Manfred Jahn
Narratology: A Guide to the Theory of Narrative

Full reference: Jahn, Manfred. 2017. Narratology: A Guide to the Theory of Narrative. English Department, University of Cologne.

Version: 2.0. May 2017.

New in this version: Multi-part mind map on WHO tells WHAT HOW (N2); expanded chapter on Focalization including 30+ paras on a 'constructivist' model (N3.2.8); updated references and bibliography.

This page:

This doc as PDF:

Project jump page:


All paragraphs in this section are labeled 'N' for 'narratology'. If you quote from this document, use paragraph references (e.g., N5.4 etc) rather than page numbers.


N1. Getting started 
N2. The narratological framework
     N2.1 Background and basics
     N2.2. Narrative genres
     N2.3. Narrative communication
     N2.4. Narrative Levels
N3. Narration, Focalization, and Narrative Situations
     N3.1. Narration
     N3.2. Focalization (point of view)
     N3.3. Narrative situation
N4. Action, story analysis, tellability
N5. Tense, Time, and Narrative Modes
     N5.1. Narrative Tenses
     N5.2. Time Analysis
     N5.3. Narrative Modes
N6. Setting and fictional space
N7. Characters and Characterization
N8. Discourses: representations of speech, thought and consciousness
N9. A Case Study: Alan Sillitoe's "The Fishing Boat Picture"
N10. References

N1. Getting started

This chapter builds a toolbox of basic narratological concepts and shows how to put it to work in the analysis of fiction. The definitions are based on a number of classical introductions -- specifically, Genette (1980 [1972]; 1988 [1983], key terms: voice, homo- and heterodiegetic, focalization); Chatman (1978, key terms: overtness, covertness), Lanser (1981; key terms: voice, human limitation, omniscience); Stanzel (1982/1984, key terms: narrative situation, authorial, figural, reflector), Bal (1985, key term: focalizer), Fludernik (1996, key term: natural narratology). In the later chapters of this script, the toolbox will serve as an organizational framework for contextualizing a large number of more specific terms and concepts.

N1.1. Normally, the literature department of a bookshop is subdivided into sections that reflect the traditional genres -- Poetry, Drama, and Fiction. The texts that one finds in the Fiction department are novels and short stories (short stories are usually published in an anthology or a collection). In order to facilitate comparison, all passages quoted in the following are taken from the first chapters of novels. Thus, as a side effect, this section will also be a survey of representative incipits (beginnings). Hey, that's one technical term out of the way already.

The foregoing decision to focus on fictional narratives is motivated by purely practical reasons. Many theorists prefer to kick off by discussing more elementary forms, especially real-world 'natural' narratives such as anecdotes, gossip, jokes etc, and then work their way up to fiction. Here, acknowledging the natural foundation of all narratives, we will jump right into fiction. Novels are an extremely rich and varied medium: everything you can find in other forms of narratives you find in the novel; most of what you find in the novel you can find in other narrative forms.

N1.2. First we need to define narrative itself. We do this by asking, What are the main ingredients of a narrative? What must a narrative have for it to count as narrative? For a simple answer let us observe that (i) all narratives have a story, and (ii) all stories are populated by characters. Stories can be told in the modes of spoken or written text, film, picture, performance, or combinations thereof. In verbally told stories, such as we are dealing with here, we also have a story-teller, a narrator. This getting started section will mainly focus on narrators and characters. Let me repeat our first simple definitions in the bullet format that will be used widely in this script:

N1.3. Let's go to the bookshelf, get out a few novels, open them on page 1, and see what we can do to get an analytical grip on them.

Note that in a real-life face-to-face story-telling situation (conversational/natural narrative), the narrator is a flesh-and-blood person, somebody who sees us and whom we can see and hear. But what do we know of a textual narrator when all we have is lines of print? Can such a narrator have a voice, and if so, how can it become manifest in a text? Consider our first excerpt, from the beginning of J.D. Salinger's Catcher in the Rye (first published 1951).

Chapter One

Even though we cannot actually see or hear the narrator, the text contains a number of elements that project the narrator's voice. It is not very hard to read it out loud and give it an appropriate intonation, perhaps making it sound like the voice of a teenage boy. If you are familiar with the text you will know that the narrator, Holden Caulfield, is actually seventeen. Much the same happens when you read an email from a friend and her voice projects from some typical expressions -- so that you can practically "hear her speak". A reader can hear a textual voice with his or her 'mind's ear', just as s/he will be able to see the story's action with his or her mind's eye. We will say that all novels project a narrative voice, some more distinct, some less, some to a greater, some to a lesser degree. Because a text can project a narrative voice we will also refer to the text as a narrative discourse. One of the narratological key texts is Genette (1980 [1972]), a study entitled Narrative Discourse; another is Chatman (1978), Story and Discourse. So, we are evidently right on target. We focus our attention on a novel's narrative voice by asking Who speaks? Obviously, the more information we have on a narrator, the more concrete will be our sense of the quality and distinctness of his or her voice.

N1.4. Which textual elements in particular project a narrative voice? Here is an (incomplete) list of the kinds of 'voice markers' that one might look out for:

N1.5. Further on pragmatic signals. In the Salinger passage, the narrator frequently addresses an addressee using the second person pronoun ("you"). Although this is exactly what we expect in ordinary conversational storytelling, if you look (and listen) closely, you will notice that Holden treats his addressee more as an imagined entity than as somebody who is bodily present. For instance, he is careful to say "if you really want to hear about it [...] you'll probably want to know". This rather sounds as if he is addressing somebody whom he does not know very closely. Nor does the addressee actually say anything. At this point, we cannot tell whether Holden has a particular addressee in mind, or whether he addresses a more general, perhaps merely hypothetical audience. "You" could be either singular or plural. Some critics assume that Holden's addressee is a psychiatrist, and "here", the place where Holden can "take it easy" after all that "madman stuff", might well refer to a mental hospital. Frankly, I have forgotten whether the question is ever resolved in the novel. What is important at this point is that it can make a difference in principle whether the narrative is uttered as a private or a public communication, to a present or an absent audience.

N1.6. Oddly enough, there is one specific audience that neither Holden Caulfield nor any other narrator in fiction can ever be concretely aware of, and that is us, the audience of real readers. We are reading Salinger's novel, not Holden's; as a matter of fact, Holden isn't writing a novel at all, he is telling a tale of personal experience (also called PEN -- personal experience narrative). The novel's text projects a narrative voice, but the text's narrator is temporally, spatially, and ontologically distant from us. Ontologically distant means he belongs to a different world, a fictional world. Fictional means invented, imaginary, not real. The narrator, his/her addressee, the characters in the story -- all are fictional beings. Put slightly differently, Holden Caulfield is a 'paper being' (Barthes) invented by Salinger, the novel's author. And again, Salinger's novel is a novel about somebody telling a story of personal experience, while Holden's story is the story of that personal experience.

Just as it is a good idea not to confuse a narrator (Holden, a fictional being) with the author (Salinger, the real person who earned money on the novel), we must not confuse a fictional addressee (the text's "you") with ourselves, the real readers. Holden cannot possibly address us because he does not know we exist. Conversely, we cannot talk to Holden (unless we do it in our imagination) because we know he does not exist. By contrast, the relationship between us and real-life authors is real enough. We can write them a letter, we can ask them to sign our copy (supposing they are still alive). Even when they are dead, readers who appreciate their work ensure their lasting reputation. There are no such points of contact with Holden. The closest analogy to a real-life scenario is when we read a message which was not intended for our eyes, or when we overhear a conversation whose participants are unaware of the fact that we are (illicitly) listening in. Fiction, one might say, offers the gratification of eavesdropping with impunity.


N1.7. What we have just established is the standard structure of fictional narrative communication. Participants and levels are usually shown in a 'Chinese boxes' model. Basically, communicative contact is possible between (1) author and reader on the level of nonfictional communication, (2) narrator and audience or addressee(s) on the level of fictional mediation, and (3) characters on the level of action. The first level is an 'extratextual level'; levels two and three are 'intratextual'.

N1.8. The beginning of Salinger's novel projects quite a distinctive narrative voice. Other novels project other kinds of voices, and sometimes it may be quite difficult to pinpoint their exact quality. What, for instance, do you make of the following incipit to James Gould Cozzens's A Cure of Flesh (first published 1933)?


Contrast this narrative discourse to the narrative discourse that we heard in Salinger's text. The Salinger passage gave us plenty of information about the pragmatic parameters of the narrative situation: there was an addressee (a "you") who was spoken to, we had rich indications of the narrator's language and emotional constitution. None of this is to be found in the present passage. Knowing the rest of the novel, I can tell you that we will never learn the narrator's name, he* will never use the first-person pronoun (that is, will never refer to himself), and he will never directly speak to his addressee. Yet we can recognize well enough that this is a narrator who begins his narrative with an intelligible exposition of the setting of the story. This is a text which has a function and a purpose and therefore projects a purposeful voice. As a matter of fact, it is difficult to imagine somebody speaking or writing without using any style at all (we will come to such a case, however). In ordinary circumstances, at any rate, one is required to speak 'co-operatively' (as pragmaticists put it) -- one selects expressions that are suitable to the purpose in hand, and suitable expressions rely on assumptions about possible readers, their informative needs, intellectual capabilities, interests, etc. Speaking, we do that all the time, or at any rate ought to. Approaching the matter from this angle, one can see that Cozzens's narrator presents a sequence of concise and carefully worded statements which very adequately serve a reader's needs. Reading the passage out loud we'd probably give it a neutral or matter-of-fact voice. But, of course, a matter-of-fact voice is definitely more than no voice at all. At the same time, compared to Holden's voice, this narrator's voice is notably less distinctive.

* Lanser's rule (N3.1.3.) will be observed throughout -- if the narrator is nameless, I will use a pronoun that is appropriate for the real-life author. Cozzens is a male author; hence I refer to the covert narrator in the passage as "he".

N1.9. Having established the foregoing difference in distinctiveness, the audibility of a narrative voice is best understood as being a matter of degrees. In fact, following Chatman (1978), narrative theorists often use the oppositional pair overtness and covertness to characterize a narrative voice, adding whichever qualification or gradation is needed. Narrators can be more or less overt, and more or less covert. Both Holden Caulfield and Cozzens' anonymous narrator are overt narrators, but Holden is clearly the more overt of the two.

Covert narrators, now, must clearly have a largely indistinct or indeterminable voice. Although we have yet to meet covert narration as a phenomenon, let us briefly speculate on how it might be possible at all. By simply inverting our definition of overtness, we can say that a covert narrator must be an inconspicuous and indistinct narrator -- a narrator who fades into the background, perhaps, one who camouflages him- or herself, who goes into hiding. What hiding strategies are there? Obviously, one can try not to draw attention to oneself -- hence a narrator who wishes to stay covert will avoid talking about him- or herself, will also avoid a loud or striking voice, and will also avoid any of the pragmatic or expressivity markers mentioned in N1.4. One can also hide behind something; if all else fails, one can hide behind someone -- keep this in mind; it will get us somewhere.

N1.10. So far we have been talking about a narrator's voice as projected by textual expressions signaling emotion, subjectivity, pragmatics, rhetoric, etc. Let us now turn to the question of the narrator's relationship to his or her story, more specifically, the question whether the narrator is present or absent in it. (The narrative types that we are going to identify here are said to be based on the 'relation criterion'). Using common terms, we know that anybody who tells a story must decide on one of two basic options: whether to present a first-person narrative or a third-person narrative. Considerable debate has raged among theorists about the suitability of these terms, and while 'first-person narrative' is still widely used (we, too, will use it presently), the term third-person narrative has generally been recognized to be misleading. In the following I will therefore additionally use the terms suggested by Genette (1980 [1972]) -- homodiegetic narrative (= roughly, first-person narrative) and heterodiegetic narrative (= third-person narrative). Diegetic here means 'pertaining to narrating'; homo means 'of the same nature', and hetero means 'of a different nature'. The detailed definitions are as follows:

N1.11. Usually (but not always, and this has turned out to be a major theoretical problem), Genette's two categorical types correlate with a text's use of first-person and third-person pronouns -- I, me, mine, we, us, our, etc., as opposed to he, she, him, her, they, their, etc. In fact, there is quite a good rule of thumb (but it is only a rule of thumb) to the effect that:

In yet other words, in order to determine the 'relation' type of a narrative, one must check for the presence or absence of an 'experiencing I' in the story's plain action sentences. Note well, the expression 'plain, story-related action sentence' refers to sentences which present an event involving one or more characters in the story. For instance, "He jumped from the bridge" (= willful action), and "She fell from the bridge" (= involuntary action), and "I said, 'Hello'" (= speech act) are all plain action sentences. By contrast, "Here comes the sad part of our story", and "It was a dark and stormy night" (i.e., a comment and a description, respectively) are not plain action sentences.

A novel is a type of text that makes use of many kinds of sentences, and not all of them are plain action sentences -- for instance, descriptions, quotations, comments, etc., are not. Indeed, as we have already seen, many novels begin with an exposition-oriented prologue (a 'block exposition'), introducing characters and setting, often via descriptive statements. While such prologues tell us a lot about the quality of the narrative voice (cp. the Salinger and the Cozzens passages above), they do not necessarily tell us whether the narrative is going to be homodiegetic or heterodiegetic. It is only when the story itself gets going, employing proper action sentences as defined above, that we get into a position to judge whether the narrator is present or absent as an acting character in the story. Actually, sometimes we have to wait quite a while until we get the full picture of which characters are involved in what ways. Sooner or later, however, a narrator's relation to his or her story becomes reasonably clear.

N1.12. We have, of course, already discussed a homodiegetic passage, namely Salinger's Catcher in the Rye (if you recall, this is a story about "what happened to me", a neat formula definition of first-person storytelling). At this point, however, an incipit which, for the reasons just mentioned, is a bit more action-oriented can serve as another straightforward case. Here is the beginning of Margaret Drabble's The Millstone (published 1965).

For analysis, I will simply repeat the text, inserting some analytical annotations:

N1.13. According to Genette, Drabble's novel is a homodiegetic narrative on the strength of the single 'relation' condition that the narrator is present as a character in her story. In order to assess the typical implications of such a scenario, and put them to work in an interpretation, we will also make use of Stanzel's theory of typical narrative situations. For this line of inquiry, it is important to realize, first of all, that a homodiegetic narrator always tells a story of personal experience, whereas a heterodiegetic narrator tells a story about other people's experiences. According to Stanzel, Drabble's text is a typical first-person narrative (in the context of narrative situations, we will prefer this term over homodiegetic narrative) because the narrator tells an autobiographical story about a set of past experiences -- experiences that evidently shaped and changed her life and made her into what she is today. Like other typical first-person narrators, she is subject to 'ordinary human limitations' (Lanser): she is restricted to a personal and subjective point of view; she has no direct access to (or authority on) events she did not witness in person; she can't be in two places at the same time (this is sometimes called the law against bilocation), and she has no way of knowing for certain what went on in the minds of other characters (in philosophy, this restriction is called the "Other Minds" problem). It is obvious that a narrator's handling of these limitations, and a text's relative closeness to, or distance from, such typicality conditions ('default conditions') can tell us a lot about the 'slant' or attitude of the narrative voice as well as the motives for telling the story.

N1.14. Let us now turn to heterodiegetic narration and consider the beginning of George Eliot's Adam Bede (first published 1859). This time, I am directly adding various annotations.



Conceivably, you may be puzzled why this has been classified as a heterodiegetic text. After all, aren't there three first-person pronouns (two "I"s, one "my") in the first paragraph? True enough, but nothing follows from this. Any narrator can refer to him- or herself using the first-person pronoun. Looking at first-person pronouns and overlooking the context in which they occur is just like walking into a trap -- the notorious "first-person pronoun trap". Re-check the definitions above to ensure that the only thing that is relevant for determining whether a text is homodiegetic or heterodiegetic is the relation of the narrator to his or her story -- if they are present in the action, they are homodiegetic, if not they are heterodiegetic. The first paragraph of Eliot's novel gives us the background setting of the story, uttered by a highly overt narrator (in this respect the three first-person pronouns are relevant, but they project a vocal quality, not a relation. We are listening to an overt narrator but whether this is going to be a story of personal experience or not is still an open question. At the same time one can already sense that the exposition is presented by somebody who is above and beyond all the people and things in the story. This is not really a remembering voice. Apparently the narrator knows all the facts, yet nobody is going to ask her how she came by her knowledge. When the story gets going in the second paragraph, all characters in it (so far, at any rate) are third-person characters. Any first-person identifying an acting or speaking character in the action itself would be significant indeed (because it would signal an experiencing I). But nothing like that happens. As a matter of fact, we'd all be a bit disoriented, I suppose, if the second paragraph began with the words "The afternoon sun was warm on the five workmen there, and I was one of them".

N1.15. Remember, a heterodiegetic narrator is somebody who is not, and never was, a character in the world of the story. The fact that a heterodiegetic narrator has a position outside the world of the story makes it easy for us to accept what we would never accept in real life -- that somebody should have unlimited knowledge and authority. Heterodiegetic narrators typically assume the power of omniscience -- knowing everything -- as if this were the most natural thing in the world. When inclined to speak overtly, heterodiegetic narrators can speak directly to their addressees, and they can liberally comment on action, characters, and storytelling itself (as happens in the Eliot excerpt above). (Homodiegetic narrators can do that too, of course, but owing to their human limitations, especially their lack of omniscience, they tend to do it differently.) Evidently, then, this is again a set of typicality conditions which we can use to enrich Genette's "pure" category of heterodiegetic narratives. Following Stanzel, we will call this type of heterodiegetic-overt narration and the typicality conditions associated with it an authorial narrative situation (or just plain authorial narration). Of course, an authorial narrator's comprehensive ('Olympian') world-view is particularly suited to reveal the moral strengths and weaknesses of the characters. Typical authorial texts are the 19C novels of 'social realism' by authors such as George Eliot, Charlotte Brontë, Charles Dickens, and Thomas Hardy.

N1.16. As pointed out above, Genette's categorical distinctions (homo- and heterodiegetic), which are based on a clear-cut 'relation' condition (narrator present or absent in the story), can be fruitfully complemented by considering the typicality features, expectations, and implications that come with Stanzel's narrative situations (first-person and authorial narration, so far). Things get a bit more complicated now because Stanzel's model has yet another typical narrative situation. Because it is a difficult type, and comes with traps of its own, I will approach it with due caution. You can probably guess what is coming.

Recall that in the preceding paragraph authorial narration was tied to a heterodiegetic and overt, i.e., distinctively voiced narrator. We are now going to refocus our attention on the question of overtness and covertness. All set? Brace yourself, then, and consider this beginning of Ernest Hemingway's For Whom the Bell Tolls (first published 1943).


[On a subsequent reading of this script, you might consider the following side issue (tricky questions department): Suppose the last sentence were "Yes," I said. Describe the consequences (a) with respect to narrative type (Genette) and (b) with respect to narrative situation (Stanzel).]

N1.17. In the Hemingway passage, the narrator's voice is much harder to determine than in all of the excerpts quoted so far, including the Cozzens passage. There are three reasons for this:

  1. We do not get any of the expressivity markers that normally project a distinctive voice -- no first-person self-reference, no value judgments, no italicized emphasis, no indications of a moral agenda, point of interest or purpose, nothing of the sort.
  2. The narrator is not a co-operative storyteller. He does not acknowledge any actual or hypothetical addressee(s); quite the contrary, he conspicuously flouts the maxim of addressee-oriented (reader-friendly) exposition normally expected at the beginning of a novel. After all, setting and characters have to be introduced somehow. Thus far into the text, however, we don't know where we are, we don't know who the characters are, how many there are, or what they are doing there. And, incidentally, if you think they are talking in English (as you are bound to do, what choice have you got?) you are dead wrong. The only thing one knows at this point is that the scene opens in some exterior natural setting, a hilly terrain, evidently; it is daytime, and there are at least two characters talking to each other.
  3. The main point, however, is that the narrator seems to withdraw or hide behind the main character whom we encounter even in the first word of the text. Minutely, from moment to moment, the text seems to render this character's perceptual horizon -- the things he sees, feels, and hears (note how cleverly this is suggested by terms such as the "pine-needled floor", the "gently sloping" ground, the wind blowing "overhead"). It won't take long and the text will also render this character's thoughts, plans, and memories, in short, the whole subjective landscape of his consciousness. Then we will also -- but always incidentally, as it were -- learn more about the story's background -- that it is set in the Spanish civil war, that the two characters are engaged in reconnoitering enemy territory, etc. Note how easy it would have been for a co-operative narrator to indicate that the characters are communicating in Spanish -- a simple "Sí" instead of a "Yes" would have been an excellent pointer, for instance. But no, he does not do it. And yet you can be dead certain that Hemingway knows exactly what he is doing by using such a narrator. Certainly no critic would be silly enough to say this is a bad story incipit!

How does the passage work? Clearly, it is both heterodiegetic (narrator not present as a character in the story) and covert (inconspicuous narrator's voice). In addition, one of the story's characters -- the central character, in fact -- acts as a 'central consciousness' (as Henry James fittingly put it). The reading experience created by such a text is quite remarkable. (1) Because the narrator is so covert, the text projects a sense of 'directness' and 'immediacy' -- which is quite logical, if one reflects on the meanings of 'direct' and 'immediate' (i.e., without intercession of a middleman). (2) Because the text is so strictly aligned with one central character's spatio-temporal co-ordinates of perception, the reader is drawn into the story and invited to co-experience what it is like to be a participant -- this particular participant -- in the unfolding events.

N1.18. Here are the technical terms that further describe the phenomena discussed above. The technique of presenting something from the point of view of a story-internal character is called internal focalization. The character through whose eyes the action is presented is called an internal focalizer (some theorists prefer the term reflector, see N3.2 for more detailed definitions). A focalizer is somebody who focuses his/her attention and perception on something. Note that the Hemingway passage has two occurrences of the verb see, and more seeing and other perception is implied by various other expressions and constructions ('perception indicators'). Even though there are two characters in the action, the subject of the various acts of perception is only one of the two. Finally, the reader's imaginative adoption of a reflector's point of view is usually called 'immersion' or (a bit quaintly) 'transposition to the phantasm' (Bühler 1990 [1934]).

Just as we asked Who speaks? in order to identify a text's narrative voice, we can now use the question Who sees? as a formula to alert us to the possible presence of an internal focalizer. And, again following Stanzel, we will call the specific configuration of a heterodiegetic-covert narrative which backgrounds the narrator and foregrounds internal focalization a figural narrative. The Hemingway passage quoted above is a 'figural' passage, and the narrative situation underlying it is a 'figural narrative situation'. The Cozzens passage quoted in N1.8 is not a figural passage because there is no reflector figure and no internal focalization in it. If you need a mnemonic, link reflector figure to figural narration. No reflector figure, no figural narration. For good measure, here is the more general definition:

The full extent of figural techniques was first explored in the novels and short stories of 20C authors such as Henry James, Franz Kafka, Dorothy Richardson, Katherine Mansfield, Virginia Woolf, James Joyce, and many others. Subduing the 19C overt narrator's intrusive presence, these authors opened the door to an unmediated access to a character's mind, and through this 'prism' or 'filter', to the story's events. Logically enough, the most radical reduction of narrative voice comes when the text presents nothing but a direct quotation of a reflector's thoughts -- as in the form of an 'interior monologue' (N8.9). Incidentally, the filmic device of the 'POV shot' (= point-of-view shot) is an instructive equivalent of the technique of internal focalization described here. (Jump to F4.3.8 for a graphic illustration.)

N1.19. To recapitulate: in addition to Genette's two basic types of narratives (homodiegetic and heterodiegetic) our toolbox now also stocks Stanzel's three typical narrative situations: first-person, authorial (heterodiegetic-overt) and figural (heterodiegetic-covert plus internal focalization).

You will be relieved to learn that most prose narratives establish their narrative situation quickly, sometimes (as we have seen) in the very first sentence, and then stick to it throughout the whole text. Be forewarned, however, that there are (i) texts that switch narrative situation from one chapter to the next (e.g., Joyce, Ulysses; Dickens, Bleak House), (ii) texts that switch narrative situations from one passage to another, and (iii) borderline cases whose narrative situation vacillates between one or more types.

N1.20. Suppose somebody asked you whether narrative theory has anything of interest to offer on "How to write a novel". What you could say -- after duly pointing out that narrative theory is more interested in how narrative texts work than in how one can make them work -- is this. The history of the novel shows that there are three tried and tested recipes. Recipe no. 1 gives you what narratologists call a homodiegetic narrative: You select one of the story's characters and let her/him tell it as a tale of personal experience. Recipe no. 2 gives you an authorial narrative: You use an overt and heterodiegetic narrator who does not belong to the cast of characters, invest him/her with far-ranging knowledge privileges (up to omniscience), and let him/her tell a story of (for instance) social realism. Finally, recipe no. 3 creates a figural narrative: You use an entirely covert narrator and present the story as if seen through the eyes of an internal focalizer.

N1.21. Applying the technical terms defined above, see what you make of the following passage from Crome Yellow by Aldous Huxley (first published 1921):

Chapter One

Can you say whether this a homodiegetic or a heterodiegetic narrative? Personally, I can't see any first person pronoun referring to somebody involved in the action. This isn't what a narrator remembers, is it? The only story-internal character present at all is somebody called Denis, and he is referred to by the third person pronoun, "he". It is surely unlikely that a first-person character -- an experiencing I -- should suddenly join him out of the blue. Therefore, most likely this is a heterodiegetic narrative. (And so it is.)

N1.22. But now for a few more challenging questions. First, what can one say about the quality of the narrative voice? Well, in the first two sentences, at least, we seem to be getting some background information (on setting and railway lines). This is roughly reminiscent of what we had in the Cozzens excerpt. So is this, too, an addressee-conscious narratorial exposition in a neutral tone of voice?

Actually, no, this is not a very satisfactory explanation. For, unlike the Cozzens excerpt, this one has plenty of emotional and subjective expressions in it -- expressions like "goodness only knew", "the green heart of England", "thank Heaven" -- and since these are strong voice markers they suggest a highly overt rather than a neutrally overt voice (as in Cozzens). So this must be heterodiegetic-overt narration then?

Nope, that isn't it, either. Note that the third sentence begins with the words "Denis knew", which is rather reminiscent of the figural style of the Hemingway excerpt (N1.12). What now? Is the text, and are we as readers, hovering between, or perhaps helplessly tossed among, different modes of narrative?

N1.23. Although this is not really a difficult text, the questions raised by it are difficult to answer on a theoretical level. Any strategy that helps explain how readers negotiate such texts is therefore most welcome.

One such strategy is the 'FID test' which Michael Toolan has proposed recently (2001: 132). FID is a common abbreviation for free indirect discourse -- a term which I am sure you have come across hundreds of times already in your studies. Put simply, FID is a technique for rendering a character's speech or thought. FID does this 'indirectly' in the sense that it transposes pronouns and tenses into the pronoun/tense system of the narrative's ordinary narrative sentences (for instance, it may shift a first person into a third person, and the present tense into the past). But there are no quotation marks, and often any identification of speaker or thinker (he said, she thought etc.) is also dropped. As a consequence, there is often no formal difference between FID (reporting a character's speech or thought) and a plain narratorial statement. Now, it may not be very important whether a sentence is the one thing or the other -- for instance, nothing may hinge on whether It was twelve o'clock; he had plenty of time to catch the plane is just the rendering of a character's thought or a piece of information given by the narrator, or even both. Then again, it may make all the difference: suppose the clock is slow, the character misses the plane, the plane crashes ... you see what I mean.

In the light of this, consider "It was the next station, thank Heaven". If we take that to be a representation of a thought going through Denis' head, then we construe the sentence as FID. Read as a narratorial statement, the sentence might express the narrator's relief ("thank heaven") to have finally come to this part of the story. Of course, this second reading is an entirely far-fetched one. In order to test whether a sentence is FID or a narratorial statement, Toolan suggests to construct two unambiguous and fully explicit versions -- one which explicitly binds the sentence to the point of view of the character, and another which explicitly binds it to the point of view of the narrator. The next step is to assess, on the strength of both content and context, which version produces the better "fit". Contrast these two versions, then:

As might be expected, given the context of the sentence and the general content of the passage, the second construction is much more plausible than the first one. Hence we conclude that the original sentence is indeed an FID representation of Denis' thought (we can even 'backshift' it to recover its original form -- "It is the next station, thank Heaven" is what Denis very likely thinks, and we see at once that it fits well). We will say that the FID test registers positively on the sentence in question. The upshot of this is that we can now claim that the emotional tone projected from "thank Heaven" is not the narrator's but Denis'.

N1.24. Let us now extend the FID test and turn it into an 'IF test' (this is not a common term), a test of internal focalization. Internal focalization is mainly concerned with what is present or goes on in a character's consciousness -- thoughts as well as perception, feeling, knowledge. For instance, that list of oddly named train stations -- is that some kind of information that the narrator provides for our benefit? Or does Denis simply rehearse this list in his mind? Again we should use context and content in order to decide this question. The sentence preceding the sentence in question actually tells us that Denis knows the names of the stations "by heart". Don't write this off as an accident; rather, take it as contextual evidence supporting the interpretation that he is now rehearsing them.

N1.25. Huxley's text really requires us to make many similar decisions, and basically they all work out in the same way. For instance, who is more likely to conceptualize the train's further progress as "creeping indolently onward", the narrator or Denis? Who does not really know (or perhaps care) where the train goes ultimately -- "goodness only knew whither" -- the narrator or Denis? (Remember: a standard authorial narrator normally has a huge knowledge privilege -- up to omniscience, we said.) Who is the originator of the image of "the green heart of England"? Well, I trust the pieces of the puzzle have long fallen into place. Apparently, one can source all judgments and expressivity markers in this passage more appropriately in the internal focalizer (i.e., Denis) than in the narrator. And, somewhat surprisingly, this even goes for the very first sentence, the sentence that perhaps looked like plain narratorial exposition at first glance. Compare:

While the IF test is never absolutely conclusive, it allows us to argue for or against a particular option. In this case, we see that the internally focalized reading is quite an appropriate one. (Admittedly, however, the story's first sentence could also be the incipit of an authorial narrative. Which ingredients would actually have to be added to the text to make it an authorial one?)

N1.26. Now see how the text, as it progresses, jells into a plain case of figural narration with all that's implied by it:

For an exercise, test your own intuitions by selectively applying the FID/IF test in this passage. Again, all distinct voice-indicating emotional expressions will attach more plausibly to the internal focalizer than to the narrator. This confirms what we found earlier, namely that any vocal quality of this text belongs to the character, not the narrator. Ultimately, we can say very little about the narrator's voice because the narrator effectively hides (himself and his voice) behind the presentation of the internal focalizer's voice (and perception and consciousness). One could also say he hides his own voice by imitating the character's voice.

N1.27. Ready for another turn of the screw? As we are coming to the end of this section, I want to test our present toolbox by looking at two further examples. The first is the incipit of Jane Austen's Emma first published in 1816). For a fair division of labor, I propose to do most of the work at first, answering the simple questions, and then you get a chance to have a go at the hard ones.


This is clearly an overt narratorial voice engaged in giving concise and reader-conscious expository information on the main character (a block characterization, in other words). The paragraphs that follow present additional background information on the Woodhouse family. The narrator introduces a governess, summarizes Emma's childhood and adolescence, and comments on the developing friendship between the two women thus:

Some of character traits attributed to Emma are obviously wholly conventional, others strike one as slightly unexpected, perhaps deserving careful attention (and intonation!). Observe the projected tone of voice in "and Emma doing just what she liked", for instance. At any rate, in the following paragraph, the narrator gets down to a crucial point -- the heroine's personality -- more directly.

Clearly, this is said in a judgmental voice, and whatever else may be entailed by the summary characterization of Emma it is not an entirely positive one. What, do you think, is it in particular that is "unperceived" by Emma (but apparently quite obvious to the narrator)?

N1.28. (Emma, continued.) The paragraphs following the preceding passage now move from plain exposition of background information (often using sentences cast in the past perfect tense) to a presentation of more concrete events and action (cast in the simple past, the novel's basic narrative tense). The novel's action proper begins on the evening of Miss Taylor's wedding day, an event which causes a major change of state in the affairs of the protagonists.

First of all, the knowledge privilege now exhibited by the narrator confirms that this is a heterodiegetic narrative situated in a typical authorial narrative situation (as you surely suspected from the beginning). There is no experiencing I in the action, and a first-person narrator would have no way of knowing how Emma spent her time on the evening of that particular day.

N1.29. More importantly, however, as you negotiate these paragraphs, you will (hopefully) notice a gradual development and shift in narrative orientation. Try to put your finger on it. First of all, the text begins to focus on single, concrete events. Whereas at the beginning of the novel we were given summary accounts of large-scale events (e.g. Emma's mother's death), we are now situated in the middle of an ongoing action sequence. Does this development go hand in hand with what we have previously identified as 'internal focalization'? Of course, we could easily ask Toolan's FID/IF test questions. Is it the narrator who, reader-friendly and duty-bound as she is, informs us of the fact that "The event had every promise of happiness for [Miss Taylor]"? In other words, is this an important piece of factual information she wants us to know? Or is there a an alternative reading? Next, who is the source of the text's reference to "all her [Emma's] advantages, natural and domestic" -- the narrator? (Actually, there is a salient textual correspondence that suggests that the answer to this question is No rather than Yes.) Again, who is a likely source for the judgment that "her father [...] was no companion for her. He could not meet her in conversation, rational or playful" -- the narrator? And what difference does it make if it were not the narrator?

We can sum up the whole of the previous line of questioning by asking, how many voices does Austen's text project? And what are the consequences? Watch out, these are loaded questions, and they come with a host of interpretive implications (which is, of course, exactly what we need).

"Emma is the climax of Jane Austen's genius and the Parthenon of fiction" (Ronald Blythe, Introduction to the Penguin edition). OTT as it is, support Blythe's judgment by showing two things: (1) that the text is entirely modern in its anticipation of a future narrative technique; (2) that the global narrative design of the novel is effectively implied and established right at the beginning (you'll have to speculate a bit on what the novel is going to be about).

N1.30. Finally, here is another incipit (from Raymond Chandler's The High Window, first published 1943). Write down a protocol of your reading experience; pay particular attention to your understanding (or non-understanding) of the narrative situation as it evolves from sentence to sentence. The bracketed note numbers in the text refer to the "questions and hints" below.

Chapter One

Questions and hints:

  1. "Cool-looking", it might be argued, is part of a textual isotopy here. Don't know what an isotopy is? Check it out in P3.5.
  2. What note is struck by indicating the size of somebody's lawn in acres?
  3. "Deodar" -- had to look it up, it's an "East Indian cedar" (Webster's Collegiate). What does that tell you, I mean, not about me, about the narrator?
  4. "They" -- as in "us and them"?
  5. What is your intuition here -- narrating I, experiencing I, or self-reference of an authorial narrator?
  6. That may be what she wanted, but was it what she got?
  7. Any comment on projected attitude, tone, etc.?
  8. It certainly took a while, but now the text's narrative situation is finally firmly established. Why did the narrator do it the way he did? By way of experiment, what would one have to do to transpose ("transvocalize", Genette would say) this passage into a figural narrative? It is absurdly simple: change four words and it is done...

N1.31. Here is a survey of the main features of the incipits discussed in this section.

Text Overtness Type (Genette) Narrative Situation (Stanzel)
Salinger: "If you really want to hear about it ..." N1.3 highly overt homodiegetic first-person
Cozzens: "The snowstorm, which began at dawn ..." N1.8 neutrally overt heterodiegetic neutral* (unobtrusively authorial)
Drabble: "My career has always been marked ..." N1.12 highly overt homodiegetic standard first-person autobiographical
Eliot: "With a single drop of ink the Egyptian ..." N1.14 highly overt heterodiegetic authorial (standard 19C pattern)
Hemingway: "He lay flat on the pine-needled ..." N1.16 covert heterodiegetic figural (standard 20C pattern)
Huxley: "Along this particular stretch of the line ..." N1.21, N1.24 covert heterodiegetic figural
Austen: "Emma Woodhouse, handsome, clever, ..." N1.27 overt heterodiegetic dynamic: authorial exposition and some  internal focalization
Chandler: "The house was on Dresden avenue ..." N1.30 (ultimately) overt homodiegetic first-person

* Stanzel (1955: 28) briefly toyed with the concept of a separate category of 'neutral narration', but this was equivalent to the heterodiegetic-covert mode rather than to the heterodiegetic-weakly-overt voice that characterizes the Cozzens passage. As a matter of fact, after two introductory paragraphs, Cozzens' text shifts gears, introduces an internal focalizer and proceeds as standard figural narration. See also N3.3.11.

Exercise. Pick some novels or short stories yourself and analyze them by working through the catalog of questions available via the toolbox. You could invite friends, let them bring some novels and do the whole thing as a group exercise, or a quiz ...

N1.32. Outline of major concepts introduced so far.

A. Narrative voice N1.3

     1) Who speaks? N1.3, N1.18

     2) expressivity markers, N1.4

     3) overt/covert voice distinction, N1.9

     4) how to hide a voice, N1.9, N1.17

B. Internal focalization N1.16, N1.24

     1) Who sees? N1.18

     2) internal focalizer/reflector, N1.18

     3) FID/IF test N1.23, N1.24, N8.6

C. Basic types and typical narrative situations

     1) Genette's basic types

           a) homodiegetic, N1.10, N1.20

           b) heterodiegetic, N1.10, N1.21, N1.28

     2) Stanzel's narrative situations (N3.3.1)

           a) first-person, N1.11

           b) authorial, N1.13, N1.20

           c) figural, N1.18, N1.20, N1.26

N1.33. This is the end of the Getting Started section, and I am sorry to say that the rest of this document is much rougher going -- one definition will simply chase another. Remember that being able to identify whether text X is homodiegetic or heterodiegetic, or authorial or figural, or what not, is fine, but not much. What is really important is that these concepts come with a huge number of assumptions, expectations, implications, and, above all, questions. The following is a rough template of possible questions.

A. Questions regarding narrative situation

  1. What is the text's major narrative situation? Or does it use several narrative situations? If so, what is the pattern or strategy behind the juxtaposition of several narrative situations?
  2. Does the text stand in the tradition of certain other texts? Or does it deviate in certain respects from the stylistic norm, perhaps to the extent that it originates a new pattern?

B. Questions focusing on the narrator

  1. Who does the author choose for a speaker? Does s/he have a name and/or a distinctive voice? Is the narrator overt or covert or somewhere in between? Is the voice quality different in specific location such as (chapter) beginnings and endings?
  2. Does the narrator make any assumptions about actual or potential addressees? Is there a clear-cut narrator-audience contract? Is the extent of the narrator's (human) limitation or omniscience ever discussed or problematized?
  3. Is the narrator largely reliable or does s/he deceive him- or herself or others? Does his or her unreliability concern value judgments or facts?
  4. If the text were 'transvocalized', i.e., narrated by another narrator and in a different narrative situation, which effects would be gained, which lost? (See Stanzel 1984: ch. 3.1 for examples, including the beginning of The Catcher in the Rye.)

C. Questions regarding focalization

  1. Does the narrator use none, one, or many story-internal focalizers? If the latter, to establish which point? In first-person narration, to what extent is the experiencing I used as an internal focalizer?
  2. How accurate are the perceptions and thoughts of the focalizers, and to what extent are they fallible filters (Chatman)? Does the narrator ever comment on the focalizer's perception from a superordinate perspective?
  3. If there are several focalizers (multiperspectival narration), do their various perceptions contradict or corroborate those of other focalizers?
  4. Is the general attitude of the narrator one of sympathy/empathy towards his or her focalizer? Are the focalizer's perceptions and thoughts reported consonantly or dissonantly (ironically)?

Hopefully, the narratological concepts introduced in this section will act like analytical tools that enable you to say because because because... And that is good because, ultimately, being able to say "because" is what theory and essay writing is all about (Aczel 1998b: 49).

N2. The narratological framework

Luckily we can boil the vastly complex field of narratology down to the question "Who narrates what how?" This allows us to make use of the following multi-part mind map (underlined items clickable).


On the text-external level, WHO is the author; on the text-internal level, WHO is the narrator (N2.3.1). Narrators come in two types, homodiegetic and heterodiegetic (telling first-person and third-person stories, respectively). WHAT do narrators do? They narrate or tell a story. Stories are made up of characters, things, and events. Events have a chronological and a causal order (plot). The HOW of narrative 'discourse' is determined by choice of tense (past? present?), speed (slow? normal? fast?), event ordering (chronological? non-chronological?), and point of view (internal focalization? external focalization?).

Note that our mind map can be vastly enriched by adding further question-words or phrases. For instance, we could ask Who tells what TO WHOM (the target audience: adult readers? children?); WHY (entertainment? education? argument?); TO WHAT EFFECT (laughter? tears?); IN WHICH SITUATION? (courtroom? doctor's office? political rally?), and so on. Widening the field in this manner is actively pursued by much of recent narratological research. Note, too, that the terms listed here, especially 'narrator' and 'discourse', relate mainly to verbally told stories. Once the scope is expanded to encompass genres like comic strip, film, drama, opera, radio play etc many of the concepts used above need to be revised and adapted -- a task that has yet to be accomplished.

N2.1. Background and basics

N2.1.1. As a discipline, narratology began to take shape in 1966, the year in which the French journal Communications published a special issue entitled "The structural analysis of narrative" -- and this is still a good working definition of narratology. The term narratology itself was coined three years later, by one of the contributors to that special issue, Tzvetan Todorov (1969: 9):

Many narratologists today consider natural narratives such as occur in everyday conversation to be the most elemental and prototypical instance of storytelling. Natural storytelling is an event in which the participants are flesh-and-blood persons engaged in direct communication. In contrast, in written narratives neither narrator nor reader can see or hear the other. However, even for writers and readers the absent party is usually evoked as an imaginary presence. Specifically, readers can re-create a mental image of the narrator from lines of text. The idea that readers habitually re-create the prototypical storytelling scenario of natural narratives is the main tenet of natural narratology as proposed by Fludernik (1996). We made use of the natural narrative hypothesis in the Getting Started section of this script (N1), where one of our tasks was to abstract narrators' voices from written texts.

Ultimately, the roots of narratology, like the roots of all Western theories of fiction, go back to Plato's (428-348 BC) and Aristotle's (384-322 BC) distinction between 'mimesis' (imitation) and 'diegesis' (narration). Chatman (1990: ch. 7) uses these concepts to distinguish diegetic narrative genres (epic narratives, novels, short stories) from mimetic narrative genres (plays, films, cartoons); most commentators, however, follow Genette's (1980 [1972]: ch. 4; 1988 [1983]: 49) proposal that narrative fiction is a 'patchwork' of both mimetic and diegetic parts, mainly to be divided into a 'narrative of words' (speech and dialogue) and a 'narrative of events' (1988 [1983]: 43).

N2.1.2. Practically all theories of narrative distinguish between WHAT is narrated (the 'story') and HOW it is narrated (the 'discourse'). Some theorists, among them Gérard Genette, opt for a narrow meaning of the term 'narrative', restricting narratives to verbally narrated texts (Genette 1988 [1983]: 17); others (Barthes 1975 [1966], Chatman 1990, Bal 1985) argue that anything that tells a story, in whatever genre, constitutes a narrative. It is this latter view which is adopted here (see N2.2 for a fuller diagram of narrative text types). On this basis, our main definitions are as follows:

N2.1.3. According to the Swiss linguist Ferdinand de Saussure (the very founding-father of structuralism), any sign consists of a 'signifier' and a 'signified' -- basically, a tangible form or substance and a non-tangible meaning. For a narrative text -- a complex sign -- the signifier is a 'discourse' (a mode of presentation) and the signified is a 'story' (an action sequence). Hence, narratological investigation usually pursues one of two basic orientations:

Further on the story/discourse distinction see Jakobson (1970 -- French terms enoncé and enonciation), Dolezel (1973: Introduction); Sacks et al. (1974 -- narrative vs conversational turns); Culler (1975a); Chatman (1978: ch. 1); Genette (1989 [1972]: 164-69; Genette (1988: 18, 61-62, 130); Lintvelt (1981: ch. 4.6.2); Bal (1983 [1977]); Fludernik (1993: ch. 1.5 -- survey of story and discourse models).

N2.1.4. The main tenets of discourse narratology are well presented in the writings of Stanzel, Chatman, Cohn, Genette, Bal, Rimmon-Kenan, Lintvelt, and Fludernik. Most of the monographs published in narratology's 'classical' period -- the nineteen seventies and eighties -- are still good introductions to the field, especially Genette (1980 [1972]), Chatman (1978), Cohn (1978), Sternberg (1993 [1978]), Todorov (1981), Prince (1982), Stanzel (1984), and Bal (1985). Particularly useful are Rimmon-Kenan's (1983, revised edition 2002) concise survey, Prince's (1987, revised ed. 2003) dictionary of terms, Onega and Garcia Landa's (1996) reader (containing reprints of many foundational essays), the critical surveys by O'Neill (1994) and Nelles (1997), and the linguistically oriented discussions and exercises in Toolan (2001).

N2.1.5. More recent variants of postclassical narratology are discussed in D. Herman, ed. (1999) and L. Herman and Vervaeck (2005). Today's narratological branches include (among others) psychoanalytic narratology (Brooks 1984), historiographic narratology (Cohn 1999), possible worlds narratology (Ryan 1991; 1998; Ronen 1994; Gutenberg 2000), legal narratology (Brooks and Gewirtz, eds. 1996); feminist narratology (Warhol 1989; Lanser 1992; Mezei, ed. 1996), gender studies narratology (Nünning and Nünning eds 2004), cognitive narratology (Perry 1979, Sternberg 1993 [1978], Jahn 1997), natural narratology (Fludernik 1996), postmodernist narratology (McHale 1987, 1992; Currie 1998), rhetorical narratology (Phelan 1996, Kearns 1999), cultural studies narratology (Nünning 2000), transgeneric narratology (Nünning and Nünning, eds. 2002, Hühn 2004), political narratology (Bal, ed, 2004), and psychonarratology (Bortolussi and Dixon 2003 [empirical approach]).

Current researchers emphasize the openness of the discipline, particularly vis à vis linguistics (Fludernik 1993a), cognitive science (Duchan et al. 1995), artificial intelligence (Ryan 1991) and pragmatics (Pratt 1977; Adams 1996). For an encyclopedic survey of approaches and trends in modern and ancient narrative theory see the Routledge Encyclopedia of Narrative Theory (Herman, Jahn, Ryan, eds 2005). For a massive (1712 pp.) collection of foundational essays see Bal, ed. (2004 -- vol. 1: Major Issues in Narrative Theory; vol. 2: Special Topics; vol. 3: Political Narratology; vol. 4 Interdisciplinarity). Recent studies include Abbott (2002), a dedicated transgeneric approach containing chapters on "narrative and life" (ch. 1), narrative rhetoric, cultural masterplots (ch. 4), closure (chs 5, 12), "overreading and underreading" (ch. 7), David Herman (2002), an investigation of the cognitive, stylistic, and linguistic basics of narratology; Marie-Laure Ryan, ed. (2004), a collection of essays on cross- and transmedial forms such as pictures, music, cinema, and computer games.

N2.1.6. For a web-based source on narratology turn to the "NarrNet" page . This is an interdisciplinary website implemented and maintained by the U of Hamburg, Germany. Among the services offered are an extensive bibliography, a list of researchers, descriptions of various current research projects, events, links, discussion lists, and plenty of other useful stuff, including an online publication entitled The Living Handbook of Narratology . The Hamburg narratologists are also the driving force behind Narratologia, a series of studies on narratological subjects. See Kindt and Müller, eds. (2003) for the first volume in this series, entitled What is Narratology? Questions and Answers Regarding the Status of a Theory.

N2.2. Narrative genres

N2.2.1. So far we have only alluded to just a few representative forms of narrative. But arguably, narrative has a far wider scope. Consider the famous list submitted by Roland Barthes (from his seminal contribution in Communications 8, mentioned in N2.1.1, above):

In this passages I have highlighted not only the individual types of narrative but also the various terms used by Barthes for the 'forms' themselves -- 'genres', 'media', 'substances', and 'vehicles'. Here is a taxonomy which imposes a kind of order on Barthes' list.


Obviously, this diagram is not exhaustive but lists representative and typical genres. Actually, it might be a good idea to assume that each tree node has an additional branch leading to an implicit "Other" category, and that this may serve as an empty slot that can be filled with any new category that might come up (this is the way Chatman 1990: 115 handles it). If you come across a genre not accounted for by any prototype -- radio plays? hypertext narratives? comic strips? -- try fitting it in. Note that some forms occur more than once in the tree diagram -- e.g., check nodes for poems and plays.

N2.2.2. As noted above, narratology is concerned with all types of narratives, literary and nonliterary, fictional and nonfictional, verbal and nonverbal. One major distinction that is usually made is that between fictional and nonfictional narratives:

Because of the systematic relatedness of these concepts, many factual narratives such as historiographic texts or biographies have fictional counterparts (historiographic fiction, fictional biographies, etc.) (Cohn 1999). On the notion of panfictionality (= no matter how factual, every narrative involves a narrator's imagination, hence is fiction) see Ryan (1997b).

N2.2.3. Here is an incomplete list of various narrative themes and genres.

N2.3. Narrative communication

N2.3.1. As is shown in the following graphic, literary narrative communication involves the interplay of at least three communicative levels. Each level of communication comes with its own set of addressers and addressees (also 'senders' and 'receivers').


This model distinguishes between the levels of action, fictional mediation, and nonfictional communication, and establishes useful points of reference for key terms like author, reader, narrator, and narratee/addressee (for a book-length study on communication in narrative see Coste 1989; for the pragmatic status of narrative statements Hamburger 1977 and Genette 1991).

For example, on the level of nonfictional (or 'real') communication, the author of the short story "The Fishing-Boat Picture" is Alan Sillitoe, and any reader of this text is situated on the same level of communication. Since author and reader do not communicate in the text itself, their level of communication is an 'extratextual' one. However, there are also two 'intratextual' levels of communication. One is the level of narrative mediation (or 'narrative discourse'), where a fictional first-person narrator named Harry tells the fishing-boat picture story to an unnamed addressee or 'narratee' (see N9 for an argument that Harry might be his own narratee). Finally, on the level of action, Harry and his wife Kathy are the major communicating characters of the story. We call this latter level the 'level of action' because we are assuming that speech acts (Austin 1962 [1955], Searle 1974 [1969]) are not categorically different from other acts.

N2.3.2. Some theorists add another intermediate level of implied fictional communication (a level below the author-reader level) comprising an implied author (a text's projection of an overarching intratextual authority above the narrator) and an implied reader (a text's overall projection of a reader role, superordinate to any narratee). The main reason for implementing this level is to account for unreliable narration. See Booth (1961), Chatman (1990) [one proposing and the other defending the concept]; Fieguth (1973); Iser (1971, 1972, 1976) [on readers and 'implied readers']; Bal (1981b: 209), Genette (1988 [1983]: ch. 19) [for critical discussion], Nünning (1993), and Kindt and Müller (1999) --

N2.3.3. Following the reception-oriented model proposed by Rabinowitz (1987), some narratologist now differentiate between the stipulated belief systems/interpretive strategies of 'authorial' and 'narrative' audiences:

The two kinds of audiences are rarely the same. In particular, readers have to decide whether they should or should not adopt the narrative audience's presuppositions as projected by or reflected in the narrator's discourse. See Prince (1980) for the first major consideration of the narratee (of which text Genette said, "I would willingly and unashamedly annex that article", 1988: 131), Rabinowitz (1987), Phelan (1996) and Kearns (1999) for further elaboration and application of the audience concepts.

N2.3.4. Although the terms person, character and figure are often used indiscriminately, modern theoretical discourse makes an effort to reach a greater degree of precision by making the following distinction.

N2.3.5. Metalepsis: transgression of levels. Normally, the levels of action, fictional mediation, and nonfictional communication (as shown in the graphic above, N2.3.1) are hermetically sealed domains indicating crucial thresholds of control and awareness. Any agent situated on a higher-level dominates and frames all lower-level agents, while lower-level agents are unaware of the existence of the higher-level agents. For instance, the characters at the level of action do not know that they are characters in some narrator's story, and they cannot complain if their acts or motives are misrepresented by this narrator. Similarly, a narrator such as Holden Caulfield is not aware of the fact that he is a fictional figure in a novel written by J.D. Salinger (the point is spelled out in more detail in N1.6).

Occasionally, however, one finds some playful and not-so-playful transgressions of levels, which Genette calls 'metalepses' (Genette 1980 [1972]: 234-237). Typical cases cited in the literature are (1) characters attempting to establish communicative contact with either audience or author (see the device of the 'aside ad spectatores' in drama and film -- D3.4, also actors 'acting out of character'), and (2) narrators and narratees seemingly joining the characters in the action. Slightly modifying the terms used in Malina (2000), the first could be called a 'diegetic-to-extradiegetic metalepsis', while the second would be 'extradiegetic-to-diegetic' (these terms differ slightly from the ones actually used by Malina because I want them to dovetail with the Genettean terms listed in N2.4, below). Here is a famous example of the second type:

Clearly, a metalepsis can either be playful and harmlessly metaphorical (as in the example above) or else a serious transgression violating the "sacred frontier between two worlds, the world in which one tells, the world of which one tells" (Genette 1980 [1972]: 236) -- in other words, the domain of the discourse and the domain of the story. See D. Herman (1997) for a formal description of metalepsis and Malina (2000) for an in-depth exploration of functions, effects, and types of 'reconstructive', 'deconstructive', 'subversive', and 'transformative' metalepses.

Whoever is interested in another batch of recent studies of the phenomenon should watch out for the proceedings of the International Colloquium "Metalepsis Today" held at the Goethe Institut, Paris, on 29-30 November 2002 (ed. John Pier). Related phenomena include alterations in prose narratives (N3.3.15), the alienation effect in drama (D6.1), the device goof in film (F5.3.3), and parabasis in classical rhetoric (the latter term refers to a character directly addressing the audience).

N2.4. Narrative Levels

N2.4.1. Story-telling can occur on many different levels. As Barth (1984 [1981]) puts it, there are "tales within tales within tales". The model presented in N2.3.1, above, provides a general framework which can easily be adapted to more complex circumstances. One such circumstance arises when a character in a story begins to tell a story of his or her own, creating a narrative within a narrative, or a tale within a tale. The original narrative now becomes a 'frame' or 'matrix' narrative, and the story told by the narrating character becomes an 'embedded' or 'hyponarrative' (Bal 1981a: 43):

N2.4.2. For a fairly logical and hence recommended analysis of embedded narratives, Rimmon-Kenan (1983: 91) suggests the following terms:

See Genette (1980 [1972]: 228-234; 1988 [1983]: ch. 14) [extradiegetic, diegetic, intradiegetic, metadiegetic]; Bal (1981: 48-50) [on 'hypo-' vs. 'meta-']; Lanser (1981); Rimmon-Kenan (1983: 91-94) ['graded' narrators and narratives]; Duyfhuizen 1992; O'Neill (1994: ch. 3); Nelles (1997: ch. 5).

N2.4.3. Genette has illustrated the basic structure of embedded narratives with the help of a naive drawing using stick-figure narrators and speech-bubble narratives (Genette 1988 [1983]: 85). In graphic (a), below, first-degree narrative A contains a second-degree story B. The other examples in the graphic are 'Chinese-boxes models' which can be drawn to great accuracy, indicating both the relative lengths of the various narratives as well as their potentially 'open' status (Lintvelt 1978; Ryan 1991: 178; Branigan 1992: 114).


In example (b), A is a first-degree narrative, B1 and B2 are second-degree narratives, and C is a third-degree narrative (Question: which ones of these are matrix narratives?). Finally, example (c) illustrates the embedding structure of Henry James's The Turn of the Screw. James's novel ends on the conclusion of a third-degree narrative (the Governess's tale) without explicitly closing its two superordinate matrix narratives.

There are a number of texts which are famous for their multiply embedded narratives: The Thousand and One Nights, Chaucer's The Canterbury Tales, Jan Potocki's The Saragossa Manuscript, Emily Brontë's Wuthering Heights, John Barth's "Menelaiad". See also Chatman (1978: 255-257), Barth (1984 [1981]), Ryan (1991: ch. 9), Baker (1992).

N2.4.4. As an exercise, work out the following problems. Some of them are quite tricky; use simple Chinese-boxes models to argue your answers.

1. Can a hyponarrative be a matrix narrative?

2. Can a matrix narrative be a hyponarrative?

3. Must a first-degree narrative be a matrix narrative?

4. Can a text have more than one first-degree narrative?

5. Can a single character be both a second-degree narrator and a third-degree narrator?

N2.4.5. Comment. The foregoing account makes short shrift of a host of rather unhappy terms that haunt the narratological literature, including the term 'frame narrative' itself (does it refer to a narrative that has a frame or one that is or acts as a frame?). With reference to graphic (a) in N2.4.3, above, Genette calls the narrator of A an 'extradiegetic narrator' whose narrative constitutes a 'diegetic' level, while B is a 'metadiegetic narrative' told by an 'intradiegetic' (or, confusingly, 'diegetic') narrator. On the next level of embedding, one would get a meta-metadiegetic narrative told by an intra-intradiegetic narrator. Against this, Bal (1981a: 43) and Rimmon-Kenan (1983: 91-93) have argued that hypo- (from Greek 'under') is a more adequate prefix than meta- (from Greek 'on, between, with') to refer to what are, at least technically (though not necessarily functionally), subordinate narratives. Oddly, however, in their system, B (in graphic [a]) is a 'hyponarrative' told by a 'diegetic narrator', and if there were an additional level, Bal and Rimmon-Kenan would be happy to have a 'hypo-hyponarrative' told by a 'hypodiegetic narrator', and so on. Although the hypo- concept is a useful one, correlating hypodiegetic narrators with hypo-hyponarratives is both awkward and counterintuitive. More drawbacks of the nomenclature become apparent when one tries to tackle the problems set in N2.4.4.

N2.4.6. Embedded narratives can serve one or several of the following functions:

N2.4.7. Hyponarratives are also often used to create an effect of 'mise en abyme', a favorite feature of postmodernist narratives (Dällenbach 1981; Ron 1987; McHale 1987: ch. 8; Wolf 1993). The graphic on the right shows a visual example.


Spence (1987: 188) cites the following example:

N3. Narration, Focalization, and Narrative Situations

This section combines the theories of Gérard Genette (1980 [1972]; 1988 [1983]) and Franz K. Stanzel (1982; 1984 [English transl.]). Additionally, it also considers various revisions and modifications suggested by Chatman (1978, 1990), Lanser (1981), Lintvelt (1981), Cohn (1981, 1999), Bal (1985), and Fludernik (1996). The best preparation for understanding the key distinctions made here is to read the "Getting started" chapter of this script (N1).

N3.1. Narration (voice)

The term 'voice' metaphorically invokes one of the major grammatical categories of verb forms -- tense, mood, and voice (Genette 1980 [1972]: 213). In terms of voice, a verb is either 'active' or 'passive'. In a more general definition, voice indicates "the relation of the subject of the verb to the action which the verb expresses" (Webster's Collegiate). In narratology, the basic voice question is "Who speaks?" (= who narrates this?). In the present account, voice is also understood as a characteristic vocal or tonal quality projected through a text.

N3.1.1. As regards the question Who speaks? Who is the text's narrative voice? we are going to use the following definition of a narrator, or 'narrative agency':

N3.1.2. In Jakobson's terms, narratorial discourse (like any other discourse) can serve a variety of 'functions', mainly (a) an addressee-oriented 'phatic function' (maintaining contact with the addressee), (b) an 'appellative function' (persuading the addressee to believe or do something), and (c) an 'emotive' or 'expressive function' (expressing his/her own subjectivity). All of these function are highly indicative of a text's projection of narratorial voice (cp. N1.4). See Jakobson (1960) for the discourse functions; Fowler (1977) on the notion of a narrator's 'discoursal stance'; Bonheim (1982) on the presence or absence of narratorial 'conative solicitude'; Chatman (1990) on narratorial 'slant' ("the psychological, sociological and ideological ramifications of the narrator's attitudes, which may range from neutral to highly charged" 1990: 143), and rhetorical approaches to narratorial discourse (Booth 1961, Phelan 1996, Kearns 1999).

N3.1.3. Whatever you may think of 'political correctness' in general, interpretive discourse must decide on how to gender a narrator grammatically, mainly because it would be stylistically awkward never to use a pronoun at all. A generic 'he' is clearly out of the question, and the option suggested by Bal -- "I shall refer to the narrator as it, however odd this may seem" (1985: 119) -- is, as Ryan (1999: 141n17) rightly points out, "incompatible with consciousness and linguistic ability". By way of compromise, most scholars now follow what has become known as 'Lanser's rule':

Hence the narrator of Dickens's Hard Times would be assumed to be male and referred to by "he", while the narrator of Austen's Sense and Sensibility would be assumed to be female and referred to as "she". See Culler (1988: 204-207) for a critique of Lanser's rule and for pointing out some interesting ramifications. Problematic in Lanser's gendered pronouns are (1) that they may attribute a narrative voice quality which is better left indeterminate, in certain cases (saying "narrative agency" and "it" poses just the opposite problem, however); (2) that they establish a questionable author-narrator link (cp. N2.3.1).

The problem of sexually indeterminate narrators usually arises with authorial narrators (heterodiegetic narrators) only. See Lanser (1995) and Fludernik (1999) for a discussion of sexually indeterminate first-person narrators in Jeannette Winterson's Written on the Body and Maureen Duffy's Love Child.

N3.1.4. Depending on how the presence of a narrator is signaled in the text, one distinguishes between 'overt' and 'covert' narrators:

See N1.4, above, for a list of typical 'voice markers' which, in addition to the pragmatic signals discussed above, consider content matter and subjective expressions.

Needless to mention, overtness and covertness are relative terms, that is, narrators can be more or less overt, and more or less covert. Usually, however, overtness and covertness vary in inverse proportion such that the presence of one is an indication of the absence of the other. In analysis, it is always a good idea to look out for typical signals (or absences) of narratorial overtness or functionality.

N3.1.5. Following Genette, we will make a categorical distinction between two principal types, homodiegetic and heterodiegetic narrators and narratives. The distinction is based on the narrator's "relationship to the story" (1980 [1972]: 248) -- i.e., whether s/he is present or absent from the story.

Usually, the two types correlate with a text's use of first-person and third-person pronouns. To repeat the rule of thumb mentioned in N1.11,

N3.1.6. In order to determine the 'relation' type of a narrative or a narrator, one must check for the presence or absence of an 'experiencing I' in the story's plain action sentences, i.e., sentences which present an event involving the characters in the story. Note well that narrative texts make use of many types of sentences which are not plain action sentences -- descriptions, quotations, comments, etc. (Cp. N1.11, N5.5.5.)

As Genette points out, the criterial feature of homodiegetic narration is whether the narrator was ever present in the world of his/her story. The bare fact that homodiegetic narrators refer to themselves in the first person is not an absolutely reliable criterion for two reasons: (1) overt heterodiegetic narrators refer to themselves in the first person, too, and (2), more rarely though, there are some homodiegetic narrators who refer to themselves in the third person (famous classical example is Caesar's De Bello Gallico). See Tamir (1976); Genette (1980 [1972]: 245-247); Stanzel (1984: 79-110, 200-224, 225-236), Edmiston (1991).

N3.1.7. At this point, let us briefly return to the concept of voice. Of course, a voice can only enter into a text through a reader's imaginary perception; hence, unless the text is an oral narrative in the first place, or is performed in the context of a public reading, voice is strictly a readerly construct. In the classical narratological model, 'voice' is primarily associated with the narrator's voice (this is also how we treated the topic in N1.3 ff. In N1.29, however, we were led to ask how many voices were projected by a particular text (Austen's Emma). Under the growing impact of Mikhail Bakhtin's theory of narrative it is now standard practice to assign all addresser agencies ('senders') in the model of narrative communication (N2.3.1) their own (potential) voices. On this basis, then,

N3.1.8. Vocal characteristics can be profitably investigated by analyzing somebody's dialect (regional features, esp. pronunciation), sociolect (speech characteristics of a social group), idiolect (singular or idiosyncratic style), and genderlect (the gender-specific style preferred by women and men, respectively).

N3.1.9. According to Bakhtin (1981a [1973]), there are two basic voice effects that can characterize a narrative text:

N3.1.10. Not surprisingly, most theorists and interpreters (including Bakhtin himself) consider the dialogic text the more sophisticated, interesting and challenging form. There are two additional Bakhtinian terms that are frequently mentioned in the context of dialogism and polyphony:

Genette (1980 [1972]: ch. 5) [voice = narrator's voice]; Bakhtin (1981a [1973]); Lanser (1981) [extra- and (intra)textual voices]; Fowler (1983) [excellent analysis of polyphony and dialect/sociolect in Dickens's Hard Times]; Fludernik (1993a: 324) [on heteroglossia]; Aczel (1998a) [voice and intertextuality; voices in Henry James].

N3.2. Focalization (point of view)

N3.2.1. Genette's model

Adopting the term focalization, Genette sets out to explore the "different points of view from which [...] the action is looked at" (1980: 161). Further definitional questions include "Who sees?", "Who perceives?", "Who is the character whose point of view orients the narrative perspective?" (1980: 186), "Who serves as a text's center of orientation?", and, ultimately, "In what way is narrative information restricted with respect to completeness of information or omniscience?" (1988: 74). Although these prompts address different features -- a text's alignment to a character's perception on the one hand and the overall scope and restriction of 'narrative information' on the other -- they are easily combined using the following general definition.

Surveying Western narrative fiction, Genette distinguishes three major types of focalization -- zero (unrestricted), internal (restricted to 'inside views', that is, views into or from within a character's mind), and external (restricted to 'outside views'). See the following paragraphs for examples. Finally, Genette also distinguishes three arrangement patterns -- fixed, variable, and multiple focalization (N3.2.5).

N3.2.2. In non-focalization or zero focalization: the story's events are narrated from a wholly unrestricted or omniscient point of view. Typical example: Henry Fielding's Tom Jones (1749) and many other 18C and 19C heterodiegetic or authorial novels.

Here is an excerpt from a 20C novel, James A. Michener's Hawaii (1961). (I am numbering the examples in this section for further reference.)

The passage exhibits a panoramic point of view encompassing huge vistas of space and time. The narrator appears to have access to limitless information which transcends what is accessible to ordinary humans. He lightly refers to a time span of "more than ten million years" and asserts that "no man will ever see" the scenery's "aching beauty of lush valleys and waterfalls". To Genette's question "Who sees" the expected, if slightly surprising, answer is nobody because no perceiving character is present. To the question concerning the scope of narrative information the answer is no restriction, the narrator is omniscient. Hence, according to Genette, the passage is nonfocalized. [But is it? (N3.2.8)]

N3.2.3. In internal focalization the story's events are focalized through a story-internal character. Narrative information is basically restricted to data available to this character's perception.

The term reflector was introduced by Henry James, who also used center and mirror. Alternate terms include focal character (Genette), figural medium (Stanzel), filter (Chatman), and internal focalizer (Bal). The proliferation of terms is an indication of the importance of the concept and the immense influence of the style.

Using a reflector character produces a subjective and 'impressionistic' view of the storyworld. It makes the reader co-experience what it is like to be in the head of somebody participating in the story's events. Third-person internal focalization is basically identical to the figural narrative situation (N3.3.4), which, strictly speaking, wasn't invented until the early 20C period called 'modernism' (however, see de Jong 2001 for a discussion of much earlier proto-forms such as in Homer around 1000 BC).

For a typical example reconsider the beginning of Hemingway's For Whom the Bell Tolls (1943) (already qtd in the Getting Started section under the heading of figural narrative situation):

[To repeat from N1.10, the passage closely represents and follows the reflector character's current perceptions -- things he sees, feels, and hears ("he could see", the "pine-needled floor", the "gently" sloping ground; the wind blowing "high overhead".) Note that all narrative information is restricted and aligned to the reflector's current spatial and temporal co-ordinates. The notable effect of this technique is that the reader is sucked into the story, invited to see the world just as the character sees it, and co-experience what it is like to be a participant in the events. It is a hugely successful stylistic device, and we squarely owe it and its many variations to Henry James, James Joyce, and Virginia Woolf.]

Many modernist novels of 'literary impressionism' built stories around carefully chosen reflector characters. These included seemingly everyday people such as Woolf's Clarissa Dalloway, an upper middle-class mother and wife, and Joyce's Leopold Bloom, an advertisement canvasser. Other popular reflector figures were intellectuals, artists, and children, or characters placed in exceptional circumstances. In Woolf's Mrs Dalloway (1925), one reflector is a shell-shocked and suicidal schizophrenic; in Graham Greene's A Gun For Sale (1936), the reflector is a murderer; and in Malcolm Lowry's Under the Volcano (1947) he is an alcoholic.

N3.2.4. External focalization is a form of presentation that restricts itself to mere "outside views", neutrally reporting what would be visible and audible to a virtual camera (plus sound recorder), without any "inside views" into the minds of the characters. (In contrast, zero focalization freely allows and internal focalization strictly depends on inside views.) Externally focalized narratives typically consist of dialogue and "stage directions" only, as in the following often quoted beginning of Hemingway's short story "The Killers" (1927). "The"

[Not an entirely convincing example either? See N3.2.8 for a comment and alternative approach.]

N3.2.5. Genette additionally distinguishes three arrangement patterns. (1) Texts employing fixed focalization are exclusively presented from the point of view of a single reflector as in Joyce's Portrait of the Artist as a Young Man (1916). (2) Variable focalization occurs in narratives that employ several reflectors (in Woolf's Mrs. Dalloway, events are variously seen through the eyes of six major characters). (3) Multiple focalization (a special case of variable focalization) occurs in texts in which the events are told two or more times, each time seen through a different reflector (Patrick White's The Solid Mandala, detailed discussion in Jahn 2007).

Genette also points out that focalization patterns can be static or dynamic along longer stretches of text. Fixed internal focalization is a static pattern by definition, other patterns dynamically shift from one type to another. For instance, Genette notes that many 19C novelists tend to introduce characters via externally focalized block description before picking one of them as a reflector and presenting the events from his or her point of view (1980: 190).

N3.2.6. Two special cases of focalization have attracted some attention in the literature, so I will briefly mention them here:

N3.2.7. Here is a selectively annotated list of references to the classical (Genettean) account: Genette (1980 [1972]: 185-194 [building on Blin's (1954) concept of restriction de champ]); Bal (1983: 35-38); Rimmon-Kenan (1983: 71-85); Nünning (1989: 41-60); Vitoux (1982); Cordesse (1988); Toolan (1988: 67-76); Kablitz (1988); Edmiston (1989; 1991: Introduction and Appendix); Füger (1993); O'Neill (1994: ch. 4); Herman (1994); Deleyto (1996 [1991]); Nelles (1997: ch. 3); Jahn (1996, 1999, 2007); Niederhoff (2013) [focalization=restriction, perspective=perception]; Herman (2009). Focalization concepts have also been put to use in analyses of films (Jost 1989, Deleyto 1996 [1991], Branigan 1992: ch. 4, Kuhn 2009, Schlickers 2009), pictures (Bal 1985: ch. 7; Bal 1990), and comic strips (O'Neill 1994: ch. 4). Controversial issues are discussed in Genette (1988 [1983]: ch. 11-12), Chatman (1986), Bal (1991: ch. 6); Fludernik (1996: 343-347), Jahn (1996, 1999, 2007), Toolan (2001), Prince (2001), Phelan (2001), Margolin (2009).


N3.2.8. A constructivist model of focalization

The model presented in the following 20+ paras is an expansion of several earlier attempts (Jahn 1996, 1999, 2007). If, along with Genette, you believe the subject has caused "enough ink to flow" (1988: 65) feel free to skip forward to N3.3.

Why another account of focalization? Is anything wrong with the original model? Let us briefly review some critical comments.

As will be shown in the following, it is not too difficult to act on the objections and suggestions listed here -- we'll get rid of nonfocalization, accept seeing narrators, use a model that equally applies to first-person and third-person texts, and treat perception as psychologically conditioned. In doing so, we may not get all issues sorted, even introduce some problems of our own, but such is theory. In the words of Walt Kelly, the author of the classic Pogo cartoons, it will be our aim here to sprinkle some blossoms around and then run through the field barefooted in order to find out where the thorns are.

N3.2.9. I am labeling the model 'constructivist' because it builds on the assumption that we can never perceive a thing X directly, let alone as "what it really is". Constructivists assume that seeing amounts to creating a mental representation of the sensory input that our sense organs are capable of recording, in effect allowing us to see a real X as a mental Y. 'Y' in this formula is a 'percept', a mental representation that our mind is able to manipulate, store, and retrieve, as opposed to both the pure 'sense data' recorded by our senses, and also the world as it really is. As Stanley Fish and many constructivists since have argued, humans -- like all sentient organisms -- have a 'shaping eye' that needs to construct what it sees, and being able to see is a function not only of the perceptive capabilities of the eye itself but of the interpretive mechanisms and strategies that an organism brings to the task. Very simply and reductively put, we see (a) what our eyes are constitutionally capable of seeing and (b) what we are interested in seeing. Note that natural perception is limited in several ways. Thus, our eyes happen to be insensitive to either extremely small or extremely large objects, such as objects on the atomic or the galactic scale. The deficiencies can be addressed by making use of (or inventing) tools like the microscope and the telescope. Another, equally important, limiting condition is that we may see sharp enough alright but simply not have the brains to see the relevant shape or pattern, such as recognizing a medical symptom. In this case, the deficiency can be cured by acquiring (learning) the interpretive strategy that enables us to do it. [Church 2000 on 'seeing as'; Fish 1980: 333 on 'shaping eyes' and 'interpretive strategies'; Jackendoff (1983: ch.8) on 'preference rules'.]

Take the case of the common or garden frog, call him Kermit. Kermit's eyes are well suited to translate certain external stimuli into the sense data that his brain is able to interpret. Kermit is particularly interested in small, black, moving objects because these might be flies. Flies, he knows, are food, so whenever he sees a small, black, moving object, he will hop to and try to catch it, errors having been known to occur. Other things he largely ignores, except maybe females and competitors, for whom, I am sure, he also has stock modes of perception and action. Kermit's perception, one can say, is driven by a specially tuned mindset. Does this amount to saying that a frog's and a human's perceptive mechanisms amount to the same thing? Indeed, the basic constructivist design seems to be just the same. What difference there is lies less in the perceptive power of a frog's and a human's eyes than in the different mindsets that drive us, and them. To Kermit, flies are food, to me they are not. If Kermit sees X as Y, I am more likely to see X as Z, a fact that not only holds for frogs vs people, but also for people vs people. There you go: this is the very phenomenon that our model seeks to theorize further.

Apart from its constructivist foundation, the present approach heavily borrows from some much earlier accounts, especially Henry James's reflections on, and experiments in, perspectivized storytelling (1881), William James's (1890) theory of subjectivity, and Karl Bühler's (1934) notion of a person's spatio-temporal co-ordinate system.

N3.2.10. While the general definition of focalization as given in N3.2.1 is still compatible with our revised approach, we can now refine it as follows.

At this point, think of Patrick Süskind's novel Perfume, in which the main focalizer is gifted with an exceptional sense of smell. Or of Rose, the main character in Ursula LeGuin's "The Diary of the Rose", who knows how to operate a 'psychoscope', a science fiction gadget that visualizes other people's thoughts. Next, think of the proverbial optimist who sees his glass as half full, or of people who have a 'one-track mind'. Or of our friend Kermit the frog, to whom most things out there either are or aren't flies. Think of a murder mystery in which one chapter presents the story's events as seen through the eyes of a pathological serial killer, while another shows us one of his attacks as filtered through the perception of a victim, and a third one lets us witness the deductive reasoning of the profiler-detective second-guessing how the murderer's mind ticks. Note how the same narrative content could be presented quite differently -- by selecting other focalizers or no internal focalizers at all, other points of attack, different sequential arrangements etc. Consider Mansfield's short story "Miss Brill" (1920), where we encounter a third-person reflector with an entirely rose-tinted worldview (to be wholly demolished in the end). Explore the hooliganized world created by and in the mind of the homodiegetic narrator-cum first-person reflector in Burgess's A Clockwork Orange (1962). Grapple with the jaded pedophiliac's mind coloring Nabokov's Lolita. Note how easy it is to accept characters and narrators as focalizers and to adduce heterodiegetic and homodiegetic examples.

For ease of reference, I will generally consider a focalizer's perception and thought acts as parts of their mental activity or 'mentation'. Narrators, performing their job of focalizing a narrative, give the narrative a perspectivized shape (this is the only sense of 'focalizer' that Genette accepts [1988: 73]). The narrative discourse itself is a product of the narrator's mentation, which usually includes reader-oriented (pragmatic) goals such as being polite, relevant, and informative. Internal focalizers, in contrast, entertain no pragmatic relations with the reader, nor have they any inkling of the fact that they are used as internal focalizers. It is the narrator-focalizer who controls everything, and, strictly speaking, all internal focalizers are only stand-ins -- substitute focalizers, used by the narrator (the 'primary' focalizer), for the special purposes and effects of internal focalization. There is only one way in which narrators and readers can interact with reflectors and that is by the unilateral process of 'transposition', to which I will come in a moment.

N3.2.11. Let us be clear about the consequences of such "coming to terms" (Chatman 1990). (1) The first and most basic premise followed here is that any narrative text has at least one narrator-focalizer. Second, a text may or may not have one or several internal focalizers. Third, because narrators are accepted as focalizers the term 'nonfocalization' no longer fits anything, not even a seemingly objective utterance like "Water boils at 100°C" (cf Genette [1988: 101]). The same goes for the examples of 'zero focalization' cited in Jahn (1999). Fourth, 'external focalization' may still meaningfully refer to a behaviorist description, or neutral report, of events, nevertheless it is here understood to proceed from, hence to express, a narratorial point of view. In order to avoid confusion, I have already suggested to deprecate the term 'external focalization' in favor of 'outside view', acknowledging that however neutral or behaviorist an outside view may be, it is based on somebody's viewpoint and perception (the narrator's 'imaginary perception', to be precise). Fifth, while it makes terminological sense to oppose the terms 'internal focalizer' and 'external focalizer' (the latter Bal's variant designation for a narrator-focalizer), the plain fact is that an "external" focalizer's focalization does not normally result in external focalization (outside view), indeed that happens only rarely. For this reason, I will stick to the term 'narrator-focalizer', denoting an agent who is free to use inside and/or outside views as s/he sees fit.

N3.2.12. In order to get a firmer grip on perception I would like to come back to a 'mental model' of vision that I introduced in an earlier essay (Jahn 1996).


The graphic displays the basic relationships between a World, an eye, and a field of vision. More specifically, it lists two 'foci' corresponding to two distinct meanings of the word focus, a key concept also in Genette's exposition. Hence F1, or 'focus-1', is the burning point of the eye's lens (the point marked "+" indicating a person's literal 'point of view'), and F2 or 'focus-2', is the area in focus including the object focused on.

Because vision stands out as the standard prime example of perception, it lends itself to be treated as prototypical and paradigmatic. Naturally there is no denying that there are important differences between the various perceptual modes, but there is also a strong general family resemblance that allows us to recognize many common features, especially things like mindset conditioning and the 'shaping-eye' effect. Therefore, by metaphoric extension, our model's eye may be taken to represent any and all sense organs, while F1 stands for a perceiving subject, a focalizer, a text's 'central consciousness' etc, and F2 indicates the 'what' or 'percept' that is seen (Bal's 'focalized'). Finally, V circumscribes the extent of a focalizer's perception, including liminal cases such as peripheral, out-of-focus, and semiconscious percepts. Note that both F2 and V are variable in location and extent and therefore already act as initial filters (reductions, restrictions) on the complexity of the world. Because filtering errors are of particular interest in this context the reader is invited to take part in the striking 'selective attention test' offered at demonstrating the so-called 'inattentional blindness' effect.

Commenting on the 'vision-centric' approach pursued here, Huck (2009) has pointed out that we do not actually have an "aural, olfactory or even a haptic equivalent to a point of view: a point of smell, maybe, or a point of taste" (2009: 202). Well, one can certainly speak of a 'point of audition'; it's a well- established film-theoretical term. But Huck is right, perhaps a more suitable common element is to be found in a pattern question like "WHO perceives WHAT as happening WHERE/WHEN" -- in which WHO is the central consciousness that shapes perception and sensation, WHAT is the X-perceived-as-Y percept, and WHERE/WHEN is the perceived spatio-temporal situatedness of Y.

N3.2.13. Along with Bickerton (1995) let us distinguish two kinds of perception:

In Marcel Proust's seven-volume novel In Search of Lost Time (1913-27) (one of the key texts of 20C fiction and Genette's master test case), both online and offline perceptions show up as recurring topics (leitmotifs), and interestingly they are all seemingly chance and trivial -- the sight of some trees from a coach, the sound of a spoon on a plate, the feel of uneven flagstones in a courtyard, the taste of a Madeleine cookie dunked in a cup of tea (that's the universally known one), the smell of a public toilet on the Champs Elysées, and bending down to open one's shoelaces.

N3.2.14. Further on offline perception consider this passage about a condemned man's vision of future events:

In fiction, the representation of imaginary perception generally uses the same styles and techniques that are used to represent characters' online perception. This can be employed for manipulative purposes as in the 'verisimilar dream' case where the reader at a late point in the narrative proceedings is told that previous happenings were "all a dream" (cp James Thurber's great short story "The Lady on 102"). That said, imaginary perception often advertises its status by being notably less realistic than online perception. It is, of course, not bound by real-life constraints and allows all sorts of spatiotemporal jumps. Add to this that it can be extremely fuzzy or 'grainy' one moment and extremely 'high density' the next, as when a significant memory detail swims into focus. The Proust examples mentioned above -- all of them loaded 'epiphanies' (N3.3.10) -- are cases in point. Perhaps the most famous example can be found in Wordsworth's "I Wandered Lonely as a Cloud" where, in the final stanza, it is the poet's remembered (offline) vision of the daffodils that finally makes him understand the true impact of the original experience:

Consider also the finely paradoxical statement "I shut my eyes in order to see" generally attributed to the French painter Paul Gauguin.

Regarding narrative texts, two special cases of offline perception are of particular importance. One is that in the process of narrating the narrator imagines or recollects the incidents of the story, or, to factor it out more succinctly, the heterodiegetic narrator imagines and the homodiegetic narrator recollects, recollection clearly being a special type of imaginary perception. The other special case, often noted by theorists, is that narrative "invites the reader's imaginative cooperation" (Genette 1993: 39-40), or as Ohmann (1971: 14) puts it, the text "leads the reader to imagine a speaker, a situation, a set of ancillary events". We, too, will consider the reader a crucial player in the 'game of focalization' (Vitoux 1982).

Another point worthy of note is that imagining sights and sounds is generally easier than imagining smells and tastes. As Ryan (2010: 470) argues, percepts of taste and smell may well rely on being associated with wider conceptual processes and structures. Indeed, it is sensible to assume that perception can involve conceptualization (thought), either concurrently or as a cause and effect process. Similarly, Herman (2009: 123) has argued that focalization needs to be correlated to general sense-making strategies available via a 'cognitive grammar'. We will take these notions on board by closely linking focalization and other mental activities.

N3.2.15. The following flow diagram lists and connects the main elements and processes of mentation.


The graphic shows the external World as opposed to a human Consciousness, with the vertical dotted line separating the external and the internal or psychological space. World meets mind at the point where our senses translate external stimuli into 'sense data'. Considering the limited sensitivity of our sense organs this first filtering is responsible for the relative coarseness or 'graininess' of the input data. The 'X to Y' module then goes on to construe percept Y as the product of the data and any or all of the mindset factors.

Fairly often, a percept will be accompanied by a concurrent stream of thought, and for this reason percept and thought have been drawn as permeable and overlapping shapes. A percept may trigger a thought, and a thought may shadow or supplement a percept (see the note on 'conceptualization' and 'perceptualization' in the next para). For offline sense-data input the model specifies imagination and memory as input-generating modules. While much of the makeup of these modules resists rational and intuition-based analysis, what we can say is that, like percepts and thoughts, imagination and memory are best treated as linked and mutually supportive faculties. For instance, imagination can flesh out a fleeting impression and make a fuzzy memory more distinct. This generally is a necessary and enabling condition, but it can also be a possible cause of error (usually called the 'false memory syndrome', see for definition and examples). The farther we go back in memory the stronger the influence of imaginary gap-filling tends to be, up to a point where we can no longer be certain that what we remember ever actually happened (see N3.2.32 for a narrator's comment on this). Moreover, any sense data played back via the secondary route of memory needs to be reprocessed by the "X to Y" module, possibly resulting in a Y' notably different from the original Y -- usually because one's mindset has changed in the course of time. But the converse is also true -- very few if any of the elements generated in and by the imagination are wholly "new" because many offline percepts can be traced back to percepts already present in memory. Of course, even though not shown in the graphic, there is also a strong linkage between memory and the various mindset components.

Note that the narrator -- the agent responsible for the text's realization -- is under no obligation to present a focalizer's mentation exhaustively. Rendering all of a character's perceptions, feelings, thoughts and emotions at any given moment just isn't a practicable option. Much of the information would be redundant, communicative efficiency would suffer severely, narrative speed would become unmanageable (N5.2.3). Add to this that readers are usually good at handling selective information and filling gaps. Nevertheless, we can of course distinguish between styles of 'rich' and styles of 'sparse' representations. In "The Killers" example, for instance (N3.2.4), we encounter a markedly sparse representation which displays the reflector's visual and auditory percepts but excludes any mention of emotions, feelings, and thoughts. The opposite case -- a rich representation covering much minute detail of many mental activities -- can be found in Péter Nádas's novel Parallel Stories (2005).

For a further distinction, consider that both imaginative perception and memory recall can happen in 'controlled' or 'spontaneous' fashion, with (obviously) various stages in between. Typically, a narrator can exert a high degree of control over his or her imaginative visions, often aided by the process of revision. We need to use the term 'spontaneous' with due caution, however, as there may be all sorts of causes that we are not necessarily conscious of. Let us also note that the perceiver of imaginative data may be aware or unaware of its offline status -- which, by the way, is often used as a deceptive narrative device). Still, as long as we are aware of the deceptive potential, the distinction generally allows us to tag a dream as 'offline/unaware' and a day-dream as 'offline/aware'.

N3.2.16. Percept and thought usually appear in correlation and interaction, and sometimes they may be linked as cause and effect. I will use the term conceptualization to describe the fact that a percept has triggered a thought or evoked a particular concept (for instance, I may recognize a squiggle on a piece of paper as uncle George's signature). Conversely, I will speak of perceptualization when a thought or concept evokes a percept ("imagine a duck"). Generally, a person who has access to specialist knowledge is likely to have more 'articulate' perceptions than others, such as when somebody is able not only to just see a tangle of wires but to identify a damaged electrical coil. Narrative texts present many instructive examples. Readers generally perceptualize what the text tells them (see N3.2.18 and N3.2.22 for examples), and narrators often take great pains to use the exactly right words to represent a character's perceptions. Note the use of typical diction and dialect in the following example:

Consider also the following passage in which the text simulates the reduced conceptual competence of an animal, a lion, who the narrator momentarily picks as a reflector. The "object" and "thing" observed by the lion is a safari jeep. There is quite a jarring effect when the text suddenly shifts to an entirely different level of conceptualization.

N3.2.17. Depending on whether mentation is based on, or triggered by, online or offline input, I will use the terms online mentation and offline mentation on the understanding that, strictly speaking, it is the sense-data input (not the mentation itself) that is online or offline. Thus in the following 'online mentation' = thought + online perception, and 'offline mentation' = thought + offline perception. This allows us to lay out the general architecture of focalization as follows:


A narrator's online mentation is grounded in the point-of-view co-ordinates of his or her discourse here-and-now (basic stance: Here I am, telling this story). Provisionally, we may take this to be the reader's reception here-and-now as well -- Here I am, reading this novel, in the presence of someone telling a story.

A narrator's offline mentation allows him or her (and us) to 'transpose' to the imaginary co-ordinates and spaces of the story, as indicated by the dotted arrow line going from discourse-NOW (read: here-and-now) to story-NOW (here-and-now). Many additional options open up at this point, for instance, Here I (the narrator) am, looking at the scene of action from a bird's eye point of view; or, Here I am, positioning myself in such a manner that I can overhear (and report) a conversation; or Here I am, in the very mind of my reflector character, seeing what s/he sees and hearing what s/he thinks. Readers, for their part, may imaginatively hear the narrator speak and, like the narrator, let themselves be transported to various locations in the story here-and-now, or into the mind of a reflector character. Of course, characters, too, imaginatively transpose to imaginary other times and places, but in their case it is from a base position of story here-and-now to offline here-and-now, and, normally, back again. As just noted, in the mode of internal focalization any reflector's offline here-and-now becomes a target transposition location for narrator and reader as well.

N3.2.18. Unsurprisingly, authors and narrators are well aware of the fact that transposition is part and parcel of the "imaginative co-operation" required from readers (Genette 1993). Note the following "invitations":

As a matter of fact, we can hardly ever refuse a narrative's invitation to "step into" the story or join the story's "party" -- unless, that is, if we decide to stop reading or listening. Interestingly, the Brontë passage was copied from N2.3.5, where it is cited as an example of narrative 'transgression'. In the light of our present theorizing I am tempted to say that it is nothing less than a narrative essential. Nevertheless, we should allow for the fact that potential transposition targets may exert a variable gravitational pull, depending on factors like the perceptual graininess of the text and the degree of narratorial or figural prominence. As we saw in our initial discussion of the figural narrative situation (N1.17 f.), features like these are often directly correlated.

N3.2.19. The concept of transposition is squarely owed to Karl Bühler, who illustrated it by referring to the saying If the mountain won't come to Mohammed then Mohammed must go to the mountain. In Bühler's adaptation, Mohammed is cast in the role of a perceiver, and the mountain is assumed to be a distant object beyond his range of online perception. Yet Mohammed doesn't necessarily have to go to the mountain. Locked in his current spatiotemporal coordinates -- his 'I-here-now point of origin' or (as Bühler called it) 'origo' -- Mohammed can (i) let the mountain come to him by picturing it to be standing right outside his window, or else (ii) he can mentally go to the mountain and see it from an assumed point of view, or (iii) he can point in the direction of where he knows the mountain to be, describing it from afar and relating it to his own bodily orientation. Type (ii) is what Bühler famously calls 'transposition to the Phantasma' - the precise move readers execute when they immerse themselves in a fairy tale, listen to a travelogue, or read a novel. For illustration Bühler presents an eye-opening observation:

Even in everyday perception and conversation, Bühler points out, we are continually transposing to virtual deictic positions, mentally rotating our body axes in order to assess where something is in relation to ourselves, or how something must appear to somebody else, or to guess what it must be like to be in a particular situation. Interestingly, the one target location Bühler does not explicitly mention is moving into somebody's head and seeing the world from a reflector's point of view. However, it is clear that this, too, is a variant of his transposition to the Phantasma.

For more recent accounts of transposition and related concepts and effects the reader is advised to consult Gerrig (1993) [on 'immersion'], Duchan et al. (1995) [a collection of essays on 'deictic shift theory'], Ryan (1991) [on the processes of 'recentering' and 'relocation' in 'possible worlds', not only fictional ones], and (Ryan 2013) [on the relationship between a 'poetics of immersion', 'telepresence', and 'interactivity']. Ryan (2013: 69) also draws attention to the fact that there may be degrees of readerly "absorption", including (at the high end of the scale) stages such as "imaginative immersion", "entrancement", "addiction", etc.

N3.2.20. Let's look at two examples of narratorial focalization, both specifically concerned with online mentation. A narrator's online mentation is usually presented in the narrative mode of comment (N5.3.2). Comment is a narrative pause which momentarily focuses not on past story events but on the narrator's current situation, as s/he is presenting (writing or speaking) the narrative discourse. Here are two examples, one homodiegetic (9) and one heterodiegetic (10).

The novel's homodiegetic narrator pauses in the act of telling the story to look at "the above lines" (this is the online perception of the manuscript that lies before him, and at the same time it is the printed text we are just reading). Then he goes on to comment on his present environment, the current date, his "state of mind", and his current difficult project, which is writing the biography of his friend. The passage helps us build a mental image of the narrator, his discourse here-and-now, his emotional state, and, last but not least, the particular mindset that drives his perception of the story matter. Logically enough, we will call this type of focalization online homodiegetic and its third-person counterpart online heterodiegetic.

For an example of the online heterodiegetic type we'll go right back to one of the earliest novels in English, Robert Greene's Pandosto, written in 1588. It begins as follows:

The heterodiegetic narrator starts out with a general reflection. Although not explicitly mentioning his current environment, his here-and-now is of course present in the very existence of the discourse text itself. Detecting certain 'voice markers' (N1.4) we can hear the narrator's voice in his emotional diction and intricate parallelisms (euphuisms). Moreover, his sweeping statements on the subject of jealousy seem to invite us to partake in the social game that the psychologist Eric Berne has called Ain't it Awful. Then, by way of perceptualization, he begins to create two actor roles: one a victim of the "hellish passion", and the other a well-intentioned but ineffectual counselor, both turning into fleshed-out characters in what follows. (In A Winter's Tale Shakespeare used a modified version of the plot.) Just like in excerpt {9}, the narrator's mindset shapes form and structure of the narrative he is going to tell.

Many critics have dismissed comment passages, especially when coming from a heterodiegetic narrator, as rambling and irrelevant excursions. True enough, momentarily foregrounding the narrator's situation, they keep the reader from getting on with what the characters do and what happens next. However, as {9} and {10} demonstrate, narrators may use such intrusions for reaching out to their readers and for creating a common focus of interest. On this basis, they are well worth looking into much more closely than is often done. On the whole, however, it seems that many modern authors tend to treat readers as emancipated agents who can be trusted to work out -- and enjoy working out -- the narrator's mindset for themselves.

N3.2.21. From narratorial online mentation we now turn to the clearly more important case of narratorial offline mentation (the narrator "looking at the action"). Using three main sets of features -- (i) presence or absence of characters in the scene, (ii) position of the narrator's scenic point of view/point of audition, (iii) outside or inside views -- we can derive five main types (plus many subtypes):

Further refinement becomes available by paying attention to the various modes, goals, and preferences of homodiegetic vs heterodiegetic narration. As readers of this script you know that homodiegetic and heterodiegetic content mainly varies with respect to the accessibility, validity, and accountability of information. That is, while the heterodiegetic narrator has conventional access to other minds and may freely and factually present third-person inside views, the homodiegetic narrator has first-hand access to the mind of the experiencing I/first-person reflector only (with category (3), above, possibly serving as a substitute option). Unlike heterodiegetic narrators, homodiegetic narrators are only witnesses to the narrated events, and the reliability of their judgment may always be challenged by the question How do you know?. The heterodiegetic narrator imagines, and what s/he imagines is narrative fact; the homodiegetic narrator recollects, and what s/he remembers may or may not be true. As for the fifth type, internal focalization, we can, of course, differentiate between homodiegetic (first-person) and heterodiegetic (third-person) reflectors, but since both types of reflectors are equally locked in the here and now of a story situation, differences between them can only be small. The Chandler excerpt that was quoted in N1.30 and the technique of interior monologue (N8.9), which is used in both homodiegetic and heterodiegetic texts, are cases in point.

N3.2.22. Scarry's (1995) annotated reading of the beginning of Hardy's Tess of the d'Urbervilles may serve to illustrate the case of heterodiegetic narratorial focalization type 2 (outside view/characters in scene). Scarry uses italics to spell out the "labour of imaginative construction" that the text puts on its readers. It is a most interesting experiment in close reading.

Actually, some authors have begun to question whether readers read as imaginatively perceptive as Scarry proposes, or indeed whether it is something a good reader should do. To my mind, whether realistic or not, Scarry's reading is quite enlightening; and if there is a choice between that and one that is cursory and careless there is every reason to prefer the former. Indeed, a particularly interesting result of Scarry's experiment is that it brings out the range of possible spatial standpoints, with the 'within earshot' option playing a particularly prominent role. More than anything else, Scarry's reading demonstrates that a passage of narrative text may be based on a sequence of camera-like 'shots' and 'angles', indeed, many of the technical terms used in cinematic practice would seem to be applicable, see paras F2, F4.2, and F5.1 of PPP's film doc. Note that some of the possible shot positions have already been mind-mapped to the 'outside view' node in N3.2.17.

At this point we are, of course, strongly reminded of Genette's call to order that "unlike the director of a movie, the novelist is not compelled to put his camera somewhere; he has no camera" (1988: 73). Against this let us posit the stark antithesis that a narrator does have a virtual camera, a recording device that selects and displays the sights and sounds that encompass a story's scenes and actions. Indeed, we can go further and say that the narrator has two cameras: one for outside views, and one for inside views (a 'scenic' camera and a 'psychocamera'). Remember, the concept of a virtual camera proved quite helpful in defining the concept of external focalization (N3.2.4). Moreover, terms like vision from behind and vision from within -- Pouillon's definitions for narratorial and reflectorial focalization (approvingly cited by Genette 1972: 189) -- are very obvious analogies of shots taken by a film camera (namely the 'over-the-shoulder' and the 'point-of-view' shot, respectively, cp F4.3.8).

N3.2.23. A focalizer, we said earlier, sees X as Y, creating a filtered and colored image of the world depending on a range of mindset dispositions. Contrasting the offline perceptions (memories) of four hypothetical travelers -- four potential focalizers -- William James offered this splendid illustration:

One can easily recognize that the four men's varying "impressions" -- all clearly filtered and colored views of the world -- are the result of perceptions caused and shaped (partly also impeded) by individual mindsets. Interestingly, James refrains from censuring any of the views as inadequate or false, even though it would surely be fair to say that the fourth man is less perceptive than the other three. Henry James, recognizing the literary potential of his brother's thought experiment, added a significant twist to it in his famous image of the "House of Fiction":

Translated into the terms used here, this means that "watching the same show", one observer sees X as Y while another sees X as Z -- invoking the very 'seeing-as' condition of focalization that the present account builds on. Possibly, James's scenario may also remind you of the famous dress color debate that went viral in the Internet Of course, the important question -- touching fiction as well as life in general -- is whether one's seeing-as interpretation of the world is correct or distorted, whether it gets us through in life, and whether it agrees with other people's perceptions. But we need to tread carefully here: views that may, at first glance, strike one as unusual or even pathological may turn out to be valid and enlightening in the long run or under special circumstances. Often enough, as readers of fiction, we encounter a strange worldview that we are happy to try on for size, on the speculative notion that it might open our minds to something new and worthwhile. For instance, consider the new genre of 'autism fiction', of which Mark Haddon's The Curious Incident of the Dog in the Night-Time is the perhaps best-known example.

N3.2.24. While the Jamesian examples strongly suggest that perception is always and inescapably subjective, successful social interaction normally relies on the fact that people are able to see things identically. Hence people can usually agree about what they have seen, especially when a percept is shaped by a common conceptualization or summarized at a level of abstraction. When a train is pulling into a station, most people -- very small children, train spotters, and space aliens excepted -- will see no more nor less than just that, a train pulling into the station. This is because irrespective of individual mindsets and preferences, percepts are often compacted to fit universally familiar 'frames'. Moreover, words like 'train', 'station' etc are so unspecific that a reader can easily accept the associated perception as inherently identical and sharable.

Normally, focalizers are singular entities, but once we can assume them to be thinking as 'social minds' (Palmer) or 'interpretive communities' (Fish) they may appear in the plural number. We can therefore distinguish between singular focalizers and collective focalizers, the latter including both plural narrators or a group of characters ('collective reflectors'). Stanzel (1984: 172); Banfield (1982: 96); Richardson (2009). Examples:

Using the label 'social minds' Palmer (2010) analyzed many cases of shared perception and thought. In scenarios like political debate, the courtroom, and war we can frequently observe groups of social minds, each characterized by specifically colored perceptions, to meet and, often enough, fight.

N3.2.25. Given that two observers may or may not see things identically we often find ourselves in the crucial position of having to compare percepts and assess degrees of difference or congruence. Comparing the percepts of two observers watching the same scene, a judgment can range from perfect congruence to total discrepancy (some critics prefer the terms 'consonance' and 'dissonance', eg, Genette 1983: 66):

Note that these terms identify two polar positions on what is clearly a sliding scale. Congruence may be partial only, percepts may count as identical or non-identical, differences may be small or big, relevant or irrelevant to a question in hand. As in James's House of Fiction, in order for us to judge a focalizer's perception we need to compare it to an alternative perception of, preferably, the same thing. That alternative perception may already be present in the text as when it juxtaposes narrator vs. character (narratorial focalization challenging internal focalization) or character vs. character (internal focalization A challenging internal focalization B). If we add the reader to the equation, as we should, we also get reader vs. narrator and reader vs. character. As a matter of fact, it was Henry James who noted that the reader's part is already presupposed in the process of storytelling: "[t]he teller of a story is primarily, none the less, the listener to it, the reader of it, too" (James [1934: 63]; qtd. Stanzel [1984: 141]).

N3.2.26. The most promising if technically intricate toolset for handling discrepant perception is offered in Gilles Fauconnier's (1994) theory of 'mental spaces'. Mental space theory focuses on the fact that ordinary thinking often needs to use bubbles of protected semantic spaces, not only for keeping things apart and orderly but also for thinking in terms of comparison, projection, and 'blending' (Fauconnier and Turner 2002). Many mental spaces are construed on the spur of the moment, while others are cordoned off more permanently as when we contrast the world of the present and the world of the past, the world of appearances and the world of scientific fact, the world of war and the world of peace, the world of facts and the world of imagined things. Our constructivist formula "seeing X as Y" is a 'space builder', too, creating mental spaces X and Y (never mind that X in the constructivist's view is inaccessible). However, when the narrator sees X as Y, and a character sees X as Z, and the reader sees X as W, then Y, Z, and W constitute well-defined mental spaces that invite the dynamics of contrast and blending. The creative reasoning that is triggered in this process may well go right to the heart of a text's narrative purpose. [See Dancygier 2012 and Schneider and Hartner eds 2012 for sample analyses.]

Fauconnier illustrates the basic mechanisms of mental spaces by referring to the seemingly odd sentence "In Len's painting, the girl with blue eyes has green eyes" (1994: 12). In order to deal with the sentence we need to construct two spaces: (i) the world of reality, where the girl has blue eyes, and (ii) the world of Len's painting, where her eyes are green. Now, although the girl is clearly the "same" girl, she resides in two spaces where she is assigned certain properties, functions and relations including some that would be incompatible or contradictory if handled within a single space. Balancing separate spaces, essential insight may become available via comparison and projection. I don't actually know what or if any insight accrues from Len's picture; however, I have come across a perfect narrative companion piece, in which the blended insight smacks the narrator in the face:

The person jointly present in both current and past mental spaces is the first-person "I". Like the girl with the blue and green eyes, this person splits into two versions: the narrating I, situated in his current discourse here-and-now, and the ten-years-younger experiencing I as it appears in recollection and photo. Surveying the mental space of the photo the narrator notes some good and some bad points of the outward and inward constitution of the younger self. Comparing this to what he has become in the mental space of the present, the narrator comes to the harsh realization that while the good points outweighed the bad ten years ago, things are just the other way round now.

N3.2.27. Let us briefly turn to the linguistics of point of view cues. On a textual level we may roughly distinguish the following types. (1) Subjective expressions comprise appellations, exclamations, emphasis, certainty qualifiers ('epistemic expressions'), idiolect, dialect, and 'mind style' (N8.12) (Banfield 1982; Fludernik 1994: ch. 8). (2) Mindset cues include direct or oblique references to a person's attitude, interest, knowledge, beliefs, value judgments, hopes, fears, etc. (eg, one person's "terrorist" may be another's "martyr") (3) Deictic expressions point to a particular speaker, thinker, or perceiver (person deixis) and his or her here-and-now (place/time deixis). Pronouns and tenses come under this rubric as do words like here, there, now, then, yesterday, tomorrow, come, go.

Consider the deictics in a narrative sentence like "She felt sad now" (Galbraith 1995: 25). Relative to the narrator's I-here-now, she has the deictic import of 'not I, the narrator, who is uttering this sentence', and the past tense has the deictic import of 'not now as I, the narrator, am speaking'. Note, however, that, there is a deictic now in the sentence that relates to the I-here-now point of origin of the third-person character (an internal focalizer) and her act of perception, which is cotemporal with story-now. Balancing these deictic pointers, we see that the narratorial deixis is largely concerned with maintaining the past-tense/third-person framework, while the character's feelings are naturally aligned to her I-here-now. We could say that in this case the narrator's presence is residual only, subliminal even, as far as the reader may be concerned. Many critics (notably Hamburger and Stanzel) have claimed that the past tense actually loses its past meaning in this context -- a good idea, actually, because it explains why now can co-occur with a past tense verb.

In the process of reading, attribution of focalization must be considered revisable and dynamic. Consider the sentences "The room was dark. John opened the door and entered." (Chatman 1990: 30). One can easily read this as 'outside view/no characters" for the first sentence and 'outside view/character in scene" for the second. Now consider, in contrast, "John opened the door and entered. The room was dark." A likely reading for this is 'outside view/character in scene" for the first sentence and 'inside view/internal focalization" for the second. On this basis, "John opened the door" can already be read (or re-interpreted) as a representation of a reflector's perception. In a similar vein, the troublesome incipit of Hemingway's "The Killers" can be seen as either (a) involving a shift from 'outside view/characters in scene' to 'inside view/internal focalization' or (b) as involving no shift because the whole passage can be re-interpreted as 'inside view/internal focalization' once Nick is established as a reflector.

N3.2.28. The juxtaposition of narratorial and internal focalization that we proposed in N3.2.17 implies a hierarchical order in which internal focalization is subordinate to, or dominated by, a level of narratorial focalization, even if the latter may only be implicit. Overall it is tempting to say that internal focalization needs to be 'embedded' in a level of narratorial focalization, just like a second-degree story can only be told from within a first-degree story (see N2.4 on particulars of 'narrative levels'). Like embedded stories, embedded focalization scenarios are often depicted in the form of 'Chinese boxes' diagrams (Nünning 2001, Dancygier 2012)

As Ryan (1991: 180-81) has argued, Chinese boxes, while instructive as final-product models, are not very specific about the mechanics and effects of shifts that occur when moving from one level to another. Perhaps, Ryan suggests, procedural aspects are better captured by a dynamic structure known as a 'stack' in computer science. The concept comes with a bit of AI jargon, which is quickly established. The particular type of stack most relevant for embedding scenarios is one called a 'LIFO' (last-in, first-out) stack. A stack is either empty or contains any number of elements. Only one element, the one on top of the stack, is visible at any one point, representing the current plane or level of story or focalization, say a third-degree narrative, or a reflector's dream. Shifts are triggered by either of two operations: a new level or plane becomes accessible by 'pushing' it on top of the stack, and an old (prior) plane becomes accessible by 'popping' other elements off the stack until it becomes the topmost and current one.

Simple as the design of stacks is, its explanatory power lies in its clever combination of structure and process. Many of the so-called 'deictic shift theorists', most notably Galbraith (1995), have suggested that if we are dealing with the reader's task of negotiating the leveled structure of a narrative text then we are constructing an 'ontological' LIFO stack, or as we shall take the liberty of saying here, a stack of focalizations. Typical elements successively pushed onto a stack of focalizations include (a) a ground level of narratorial online perception, (b) the narrator's view of the story world, (c) a character's online perception, and (d) a character's offline perception (recollection, vision, or dream). If narrator, reader, or character return to a prior level of the stack, a pop discards the top-level element. Once we come to the end of the story, a final pop (or series of pops) clears away the fiction's stack. Closing the book the reader returns to his or her own online perception.

N3.2.29. Rejecting Bal's notion of embedded focalizations, Genette drops an intriguing remark: "I do not believe", he says, "the focus of the narrative can be at two points simultaneously" (1988: 76f, his italics). Translated into stack terminology Genette's statement amounts to asserting that only the element on top of the stack is visible at any one point (which is what we also claimed above). We focus on -- we only see -- what's on top of the stack. This goes some way toward explaining our readiness to execute the Bühler transposition, to jump from discourse to story and from story into the mind of an internal focalizer. The fact that an element can be popped off the stack so that the one underneath becomes the current one is plainly analogous to the process of returning from an imaginary location to an online setting.

However, as we have noted in our discussion of the sentence "She felt sad now", texts may actually retain pointers to one or more prior orientations. Similarly, the stack of focalizations may be semi-transparent, allowing us to be aware of more than one level and as a result permitting the reader to choose between competing points of view without losing sight of the overall complex pattern. Let us take a look at three sample passages to see these options in action,

N3.2.30. First consider the very common authorial-figural borderline case (N3.3.12-13).

The passage proceeds from telling us something about Buddhism in general to presenting a "good" Buddhist's mind in action. It is done in such a manner that we are momentarily transported into U Po Kyin's head in order to witness the reflector's thoughts and perceptions more or less directly. Still, the authorial narrator plainly uses the reflector as a medium to present a world view that is largely unfamiliar to the reader, and the brief parading of the reflector's mentation serves both characterization and narrative exposition in so far as it introduces a main character, actually, the story's antagonist. The different mindsets that are at work here -- the narrator's and the character's -- can be distinguished as separate but interrelated mental spaces. Above all, we can see that "a good Buddhist" means different things to narrator and character.

N3.2.31. Next, consider the 'dual' focalization in the following childhood recollection passage from Dickens's David Copperfield (1849-50):

As is commonplace and typical, the narrator's recollection is selective and mobile. The first sentence's "now" is the narrator's current discourse-now, but his imaginary vantage point already moves to a distant point in time, also identified as "now". Only the narrator himself (the narrating I) can pass the judgment that the fruit his mother is gathering in the remembered scene is "riper and richer than fruit has ever been since, in any other garden". Looking at his younger self (experiencing I), the narrator notices that the child is "trying to look unmoved". Then, as only offline perception can manage it, time is made to pass in a rush, the narrator imaginatively leaping from the garden in summer to the parlour in winter. The same two characters are present but now his mother is fully center stage, winding her "bright curls" and looking "so pretty". Watching, narrator and character (and reader?) alike are charged with emotion. However, even as the narrator adopts the child's perception, his own view is significantly qualified by the knowledge of a past irretrievably gone.

N3.2.32. And, finally, another childhood recollection, this time from Paul Auster's Report From the Interior.

This passage begins by presenting a series of childish perceptions. Despite the fact that these are not precisely oriented in time and space, the first paragraph still invites us to co-experience the child-focalizer's 'animism' (a word used later by the narrator himself). In other words, following the Pavlovian reflex of transposition we may read the first paragraph in the mode of internal focalization. Yet on the levels of style, conceptualization, and narrative mode - summary -- we clearly remain aware of the narrator's enveloping orientation and mindset. In fact, when the reflector layer pops off in the second paragraph the narrator relocates to the "here and now of adulthood" and becomes free to pass his clear-sighted comment.

N3.2.33. What do these mixed-mode or dual focalization examples show? Even though the narratorial and reflectorial layers still push and pop freely, it seems that in all of these cases the topmost layer is semi-transparent, allowing us to transpose conditionally, being guided by and remaining aware of the level that lies underneath. The direct visual analogy would be a multi-exposure or layered picture. What also comes to mind is the analogy of audio tracks -- one track 'playing' the internal focalizer's mentation and the other the narrator-focalizer's mentation - exactly as if these were 'voices', which in a sense - following our 'naturalist' approach (N1.3) -- they are. The text itself then comes out as the blended product of both tracks. For an experiment, consider the narrative fragment shown at the bottom of the following graphic. I hasten to add that the 'oscillations' shown on the two tracks are just inventions -- because at this point we clearly cannot yet measure 'volume' or 'intensity' of mentation.


We can parse the text into six sections of one sentence each, labeled A to F. Note, to begin with, that either track can reduce to the zero line (silence equaling zero intensity mentation) as in sections A and F of the reflector's voice track and section E of the narrator's track. At no point can both tracks reduce to zero, unless we are facing a blank page. At the beginning of the text, only the narrator's track is active. I have smuggled in an "unfortunately" value judgment in order to let us have a glimpse at a piece of narratorial mindset. Section B moves into psychonarration with the narrator mildly diagnosing his character's state of mind. However, because we can imagine the character to be already liminally aware of her "state of confusion" herself, I have allowed the reflector track to register some minor pulses of mental activity. Section C has a sentence of narrated perception, more or less like "She felt sad now" discussed earlier. As in that sentence, the narratorial track shows minimal narratorial activity, just keeping person and tense alignment in check, while the reflector's track is fully active and supplies all content. Narratorial activity diminishes further as representational technique slides into free indirect thought mode in section D, and finally reduces to zero in section E for a stretch of reflectorial direct thought. Once E is done, there is likely to be a paragraph or chapter break, the narrator relocates to a different here-and-now and resumes normal narration (section F). Note the subjective expressions "hurt like hell" (section C) and "that fool Peter" (section D) indicate the internal focalizer's perception and mindset, whereas "that fool Peter" as taken up in F is the narrator's ironic mention of the character's words, now indicative of the mindset of the narrator, who may not think Peter a fool at all.

Looking at the two tracks above, we can fairly confidently identify 'chunks' of focalization. Ignoring the bits of residual narratorial activity, A+B in the narrator's track is a chunk of narratorial focalization, and a second chunk begins in section F. On the reflector's track, B+C+D+E is a chunk of internal focalization. We get and accept a certain amount of overlap in sections B, C, and D. As a whole, the "two-track three-chunks" sequence seems to describe a fairly typical arrangement pattern, but other setups and other representational techniques should by all means be considered as well. Check out N8 for a fuller list of thought and consciousness representations and N5.3 for narrative modes.

N3.2.34. To round it all off, here is a checklist of research questions.

1.  In what tradition of focalization techniques does the text stand? Is it contemporaneous with the modernist styles of literary impressionism (James/Joyce/Woolf/Mansfield) (N3.2.3) or does it predate or postdate it? Does the presentation of inside and outside views deviate from contemporary practice or norms? Is it innovative or retrogressive (reviving an earlier style)?

2.  How does the narrator fill his/her role as primary focalizer? Are there specific locations, such as chapter beginnings or endings privileging the narrator's point of view? Which kinds of online perception (comment passages) does the narrator engage in (N3.2.20)? Which kinds of offline perception (imaginary perception, recollection, etc.)? Does the narrator keep a low or high profile, is s/he covert or overt (N1.9)? When, if at all, is the narrator likely to intrude into passages of internal focalization? Does s/he prefer to use psychonarration over interior focalization? Does s/he make use of a psychocamera (N3.2.22) and who does s/he point it at?

3.  Which characters are used as internal focalizers and which are not? Does the narrator present groups of characters as social minds (plural focalizers) (N3.2.24)?

4.  How transparent are the mindsets of the focalizers? Is it easy or hard for the reader to infer or deduce them? How explicit or implicit are the pointers to mindset dispositions? What and how much is left to the reader's gap filling or speculation? Are the mindsets static or do they develop over the course of the story, or (for the narrator) the telling of the text?

5.  Which filtering devices do we encounter? To what extent is the text concerned with 'equipment filtering' (organic or artificial -- eg sense of smell in Süskind's Perfume or the psychoscope in Le Guin's "The Compass Rose"). To what extent is it concerned with 'mindset filtering' (usually all of the time, of course), possibly a combination of both (N3.2.9)?

6.  How coarse or how fine is the focalizers' mentation? Do they have any perceptual weaknesses? Cognitive weaknesses? Are perceptual/cognitive achievements or failures significant topics of story and plot? How plausible are the focalizers' thoughts, how attractive or challenging are their imaginary perceptions? In their views of the world out there, are they 'fallible filters' (Chatman), or are their misconceptions understandable, pardonable, defensible, ecologically viable? Do any of the focalizers have particular perceptive or cognitive strengths? Are they experts in one area but ignorant in another? How does the text handle specialist knowledge? Does it help the reader to understand the expert bits by offering narratorial exposition, editorial footnotes or any other paratextual or epitextual glossaries, notes, credits, hypertext links, a bibliography?

7.  How rich or sparse, how detailed or superficial is the text's representation of mentation? (N3.2.15) What is the proportion of online to offline perception? Does the text move towards central moments of online focalization or offline focalization? Does the text ever obscure the status of online vs offline perception (N3.2.14)? To what effect? Is the level of detail of the presentation constant or variable? Is it correlated to subject matter? How much does the text expect or require the reader to contribute? When filling the gaps, is the reader ever led astray or garden-pathed (Jahn 1999)? If so, then locally and revisably, or over extended periods? Is there a learning effect?

8.  Which stylistic means are used to represent different types of mentation? Specifically, to what extent does the text employ 'interior monologue,' 'free indirect discourse', and 'narrated perception' in order to achieve special effects (N8.4)? How liberally or sparingly does it use explicit perception tags such as he saw, he thought etc)? Using Nelles's terms, in what proportion does the text present and perhaps prioritize ocularization (vision), auricularization (audition), gustativization (taste), olfactivization (smell), and tactivilization (touch) (N3.2.13)?

9.  What is the role of congruent and discrepant perceptions (N3.2.25)? Do they involve the level of character vs character or narrator vs character(s)? Are they ever alluded to or even discussed explicitly? Which topics or subjects do they concern? Are the conflicts ever resolved? Can the reader negotiate the different perspectives by separating them as mental spaces ("A sees X as Y, whereas B sees X as Z") (N3.2.26)? If the narrator's and the reflector's perceptions do not markedly differ, what are the reasons -- narrator restricting him- or herself to what is 'public knowledge' in the storyworld? narrator remaining neutral or non-committal? narrator allowing (intentionally/unintentionally) his or her concepts to become 'colored' by the character's concepts (N8.13)?

10. Can one refine any of these questions by paying attention to the specific conditions of first-person and third-person narration (N3.3)? What special characteristics can be attributed to first-person vs third-person internal focalizers or to first-person vs third-person narrator-focalizers?

N3.3. Narrative situation

Both Genette (1988 [1983]: ch. 17) and Stanzel (1984) use the term narrative situation to refer to more complex arrangements or patterns of narrative features. Genette's system uses the subtypes of voice (narration) and mood (focalization) in order to explore a range of possible combinations; Stanzel is more interested in describing 'ideal-typical' or (as we shall say) prototypical configurations and arranging them on a 'typological circle' (1984: xvi). The following paragraphs will mainly focus on the interpretive implications of Stanzel's model. For an excellent comparative survey of the two approaches, including some proposals for revisions, see Cohn (1981). For alternative models see Fowler (1986), Simpson (1993), and Lintvelt (1981).

N3.3.1. Stanzel's (proto-)typical narrative situations are complex frameworks aiming at capturing typical patterns of narrative features, including features of relationship (involvement), distance, pragmatics, knowledge, reliability, voice, and focalization. This line of approach results in complex 'frames' of defaults and conditions which are extremely rich in interpretive implications (Jahn 1996). In survey, the basic definitions are as follows (more detailed definitions to follow below):

N3.3.2. Here, in more detail, are the main aspects of first-person narration.

N3.3.3. Over and above the functional roles of the I-as-protagonist and the I-as-witness (Friedman 1967 [1955]), Lanser (1981: 160) identifies a range of common subtypes: I-as-co-protagonist (Nick Carraway in The Great Gatsby), I-as-minor-character (Dickens, "The Signalman"), I-as-witness-protagonist (chapter 1 of Flaubert's Madame Bovary), I-as-uninvolved-eyewitness (Faulkner, "A Rose for Emily").

N3.3.4. Typical story patterns of the first-person narrative situation. Generally, a first-person/homodiegetic narration aims at presenting an experience that shaped or changed the narrator's life and made her/him into what s/he is today. Sometimes, a first-person narrator is an important witness offering an otherwise inaccessible account of historical or fictional events (including science-fiction scenarios). Typical subgenres of first-person narration are fictional autobiographies, initiation stories, and skaz narratives, as defined in the following.

N3.3.5. Basic features of authorial narration.

N3.3.6. Typical authorial story patterns. Usually, the authorial narrator is an omniscient and omnipresent mediator (or 'moderator') telling an instructive story (a story containing a moral or a lesson) set in a complex world. The authorial narrator's comprehensive ('Olympian') world-view is particularly suited to reveal the protagonists' moral strengths and weaknesses, and to present a tightly plotted narrative. Typical subgenres are 18C and 19C novels of social criticism. See Stanzel (1984: 141-184, 185-224); Stanzel (1964: 16, 18-25); Rimmon-Kenan (1983: 95-96); Genette (1980 [1972]: 243-245); Nünning (1989: 45-50, 84-124).

N3.3.7. Figural narration.

Note that nobody uses the term 'figural narrator': the narrative agency of a figural text is a covert authorial (heterodiegetic) narrator.

N3.3.8. Note, too, that the foregoing definition assumes that figural narration is realized as a heterodiegetic (third person) text. There is also a slightly more flexible concept of 'reflector-mode narration', however, which allows the inclusion of first-person texts:

N3.3.9. Typical figural story patterns. A figural narrative presents the story's action as seen through the eyes of a reflector figure. Often, a figural text presents a distorted or restricted view of events -- to many authors, such a distorted (but 'psychologically realistic') perspective is far more interesting than an omniscient or 'objectively true' account of events. Because figural texts have a covert narrator (a withdrawn, subdued narrator) only, figural stories typically begin 'medias in res', have little or no exposition, and attempt to present a direct (i.e., both immediate and unmediated) view into the perceptions, thoughts, and psychology of a character's mind. Typical subgenres are 'slice-of-life' and 'stream of consciousness' (N8.8) stories, often associated with 20C literary impressionism and modernism (Stevenson 1998). Indeed, many authors specifically aimed at capturing the distortive perceptions of unusual internal focalizers -- e.g., a drug addict (Dickens, The Mystery of Edwin Drood), a drinker (Lowry, Under the Volcano), a two-year old child (Dorothy Richardson, "The Garden"), a dog (Woolf, Flush), a machine (Walter M. Miller, "I Made You"). Although figural storytelling is usually considered a modern form, whose beginnings are located in the 19C, see de Jong 2001 for a discussion of proto-forms of figural storytelling in Homer.

N3.3.10. Four additional elements of figural narratives are worthy of closer attention: incipits using referentless pronouns and familiarizing articles, slice-of-life format, epiphanies, and the mirror trick.

In the practice of many authors, notably Woolf and Mansfield, epiphanies may turn out to be deceptive, misguided, or otherwise erroneous (see Mansfield's "Bliss" for a particularly striking pseudo-epiphany). In many modernist texts, epiphanies are made to serve as climaxes or endings ('epiphanic endings').

All four elements identified above can also occur, albeit to a lesser extent, in the other narrative types and situations.

N3.3.11. In addition to the three standard narrative situations, we will briefly mention four peripheral categories: we-narratives, you-narratives, simultaneous narration and camera-eye narration.

The concluding sentences of the Hemingway passage make it easier to understand why Stanzel decided to subsume neutral narration under figural narration. For narratological approaches to the Hemingway story, see Fowler (1977: 48-55); Lanser (1981: 264-276); Rimmon-Kenan (1983); Chatman (1990).

N3.3.12. Here come some problem cases, and they are largely due to the fact that a whole novel or a passage of a narrative text may exhibit features of more than one narrative situation, producing borderline cases, transitional passages, and mixed-mode narrative situations. The most common phenomenon is that of 'authorial-figural narration'.

N3.3.13. As an exercise, analyze the following passages as mixed types of narration:

N3.3.14. A decidedly rarer type of mixed-mode narration is first-person/third-person narration as exemplified by, for instance, Thomas Mann's Doktor Faustus, Sterne's Tristram Shandy, Donleavy's The Beastly Beatitudes of Balthazar B, John Barth's "Ambrose His Mark", and Fay Weldon's The Heart of the Country. In Jan Philipp Reemtsma's autobiographical story Im Keller, the episodes in the cellar (where the author was held hostage for 33 days) are narrated in the third person. As Reemtsma puts it, "there is no I-continuity that leads from my writing desk into that cellar" (p. 46).

N3.3.15. Violations of standard schemes. The narrative situations have here been described as typicality models which capture standard narratorial characteristics (function, strategy, stance, limitation) and the corresponding readerly expectations in culturally acquired 'cognitive frames'. Frequently, the conditions of these frames can also be made explicit by detailing the unwritten 'narrator-narratee contract'. Of course, sometimes a narrative has a surprise in store, either because its story takes an unexpected turn or because it becomes difficult to reconcile a present mode of presentation with the general frame or contract that we thought we could use in order to optimally read and understand. It is this second type of narrative effect which Genette terms 'transgression' or 'alteration' or 'infraction of code'.

Some of the problem cases mentioned above can clearly be analyzed as infractions/alterations in this sense. Genette further differentiates between the following two main types of alterations:

Paralepsis and paralipsis are instances of violations of Grice's (1975) famous principle of co-operation -- the notion that speakers (narrators) are socially obliged to follow an established set of 'maxims': to give the right amount of information, to speak the truth, to speak to a purpose (tell something worth telling), to be relevant, etc. Cognitive strategies for handling alterations include (a) 'naturalizing' them so that they become acceptable data consistent (after all) with one's current frame of interpretation; (b) adapting the frame so that it allows for the alteration as an 'exception'; (c) treating it as a stylistic 'error'; (d) search for a replacement frame.

Frequently mentioned cases of alterations are Agatha Christie's Murder of Roger Ackroyd (a crime novel narrated by a first-person narrator who turns out to be the murderer himself), Richard Hughes's "The Ghost" (first-person narrator "lives" to tell the tale of her own death), Ambrose Bierce's "An Occurrence at Owl Creek Bridge" and Lewis Carroll's Alice in Wonderland (containing unsignaled shifts into a character's dream world). The following case construed by Fillmore (1981), modifying the incipit of Joyce's "Eveline", shows an inconsistent shift away from reflector-mode narration:

"It would have an absolutely jarring effect on the reader", Fillmore continues, "[...] if the last line of the paragraph were to read 'She was probably tired'" (Fillmore 1981: 160). See also: Genette (1980 [1972]: 194-197); Edmiston (1991) [paralepsis/paralipsis put to excellent analytical use]; Jahn (1997) [narrative situations as cognitive frames; notion of replacement frames]; Lejeune (1989), Cohn (1999: ch. 2) [both on narrator-narratee contracts].

N4. Action, story analysis, tellability

N4.1. Although 'action' is a more or less self-explanatory term, let us try to give it a more precise and useful definition.


Events in the 'primary story line' are often kept distinct from 'external' events that take place before the beginning or after the end of the primary story line (constituting a 'pre-history' and an 'after-history', respectively). According to Sternberg (1993 [1978]: 49-50), the primary story line begins with the first scenically and singulatively presented event (N5.5.6), usually, the first dialogue. See Rimmon-Kenan (1983: 61-63).

N4.2. What should count as a "minimal sequence of events"? If one permits the limit case of one event then "the quick brown fox jumped over the lazy cow" can count as a possible minimal narrative, as do "the king died", "Pierre has come" and "I walk" (Genette 1988 [1983]: 18-20). Another example used by Genette, "Marcel becomes a writer" wittily condenses Proust's 2000-page novel A la recherche du temps perdu into a single narrative sentence. Here are some additional examples of minimal narratives:

Prince's example lists a bare sequence of action units; Forster's example illustrates the principle of causal connectivity between story units (see 'plot' in N4.6); and the third is a nursery rhyme that lends itself to being enacted by gesture and physical contact. See also Culler (1975b [on narrative units]); Branigan (1992: 11-12; 222n29); Chatman (1978: 30-31; 45-48). Propp (1969) is the first famous structuralist account of functional story units (in the Russian folktale).

N4.3. None of the foregoing examples can boast of a high degree of tellability (Labov 1972; Ryan 1991: ch. 8). Normally, a story is required to have a point, to teach a lesson, to present an interesting experience (a high degree of 'experientiality', as Fludernik 1996 calls it, promoting this element to the central feature of all narrative texts), and to arrange its episodes in an interesting progression. Sketching his project, Branigan says:

Jerome Bruner, too, considers tellability and experientiality as an essence of narrative:

For an attempt to relate universal story patterns to two prototypical narrative genres -- romantic tragi-comedy and heroic tragi-comedy -- see Hogan (2003).

S.I. Hayakawa relates tellability to offering the potential of identification and empathy. Hayakawa distinguishes identification by self-recognition and identification for wish-fulfillment:

N4.4. In the poetry section we saw that units often combine to form more complex units. Just like a number of syllables may form a metrical 'foot' (P1.7) so action units usually group into 'episodes':

This definition of episodes nicely dovetails with two graphic models of narrative trajectories that have become famous: Freytag's 1863 (!) 'triangle' and Bremond's 1970 'four-phase cycle'. Freytag's triangle originally describes the action and suspense structure of classical five-act tragedy; Bremond's model originally aims at the system of possible state changes in French folk tales. Obviously, however, both models have a far more general relevance.


Regarding his corpus of fairy tales, Bremond adds that "the cycle starts from a state of deficiency or a satisfactory state" and "ends usually with the establishment of a satisfactory state" (1970: 251), i.e., the "they lived happily ever after" formula. For a more detailed account of Freytag's model look up D7.5; for the present, however, Barth's explication is quite sufficient:

N4.5. Story grammars. Various attempts have been made to devise story grammars along the lines of Chomskyan generative grammar. Some of these grammars are still used or referred to today, especially in the context of folklore studies, empirical analysis (Stein 1982), cognitive studies and Artificial Intelligence (Ryan 1991). See also van Dijk (1972), Prince (1973), Rumelhart (1975), Mandler and Johnson (1977), Pavel (1985).

N4.6. Exercise. Using the definition of 'episode' listed above as well as the two narrative progress models (Bremond and Freytag), show that the following (proto-)stories are likely to have a relatively high degree of tellability.

N4.7. The terms 'story' and 'plot' were originally introduced in E.M. Forster's Aspects of the Novel (1976 [1927]). Ideally, one should distinguish three action-related aspects: (i) the sequence of events as ordered in the discourse; (ii) the action as it happened in its actual chronological sequence (= story); and (iii) the story's causal structure (= plot).

N4.8. General summaries or synopses normally present a plot-oriented content paraphrase. For a detailed story analysis, one usually works out a story's time line so that all main events can be situated in proper sequence and extension. Generally, a time-line model is a good point of departure for surveying themes and action units; it also helps visualize events that are presented in scenic detail as opposed to events that are merely reported in, e.g., a narrator's exposition. A time-line model can also show up significant discrepancies between story time and discourse time (N5.5.2, below). See Pfister (1977/1988: chs 6, 7.4.3); Genette (1980 [1972]: ch. 1-3).

Here is a time-line and action-unit model of Sillitoe's "The Fishing Boat Picture". For a more detailed analysis using this model see the case study essay in section N9.

Story Unit Textual detail
prehistory A various references to Harry's youth
story line
B Harry's and Kathy's walk up Snakey Wood
Harry aged 24; Kathy is 30
C married life (six years)
D book-burning incident
Kathy leaves Harry (Harry aged 30)
E 10 years pass; very few references to Harry's single life
F Kathy comes back for occasional meetings
picture is pawned several times
G Kathy is run over by a lorry
Kathy's funeral
after-history H life after Kathy's death (six years)
discourse-NOW   1951; "Why had I lived, I wonder."

N4.9. Beginnings and endings.

N5. Tense, Time, and Narrative Modes

N5.1. Narrative Tenses

N5.1.1. There are two major narrative tenses: the narrative past and the narrative present. Normally, a text's use of tenses relates to and depends on the current point in time of the narrator's speech act. Naturally, the tense used in a character's discourse depends on the current point in time in the story's action. Hence,

N5.1.2. Here is how one determines a text's narrative tense:

N5.1.3. The present tense in a narrative text can have a number of functions (Casparis 1975):

N5.1.4. Tense-categorized narratives. Depending on the anteriority or posteriority relationship between discourse-NOW and story-NOW, one can distinguish three major cases:

See Margolin (1999) for a detailed comparative survey.

N5.2. Time Analysis

Time analysis is concerned with three questions: When? How long? and How often? Order refers to the handling of the chronology of the story; duration covers the proportioning of story time and discourse time; and frequency refers to possible ways of presenting single or repetitive action units. Genette (1980 [1972]: 33-85, 87-112, 113-160); Toolan (1988: 48-67); Rimmon-Kenan (1983: 43-58). For a more general account see Ricoeur (1983; 1988).

N5.2.1. Order (When?). The basic question here is whether the presentation of the story follows the natural sequence of events. If it does, we have a chronological order. If not, we are facing a form of 'anachrony':

The first chapter of Lowry's Under the Volcano postdates the rest of the action by one year, making it either a flashforward or the rest of the action a flashback. The discourse of Graham Swift's Waterland deviates considerably from the chronology of the story. Martin Amis' Time's Arrow reverses the chronology of the story (tells the story backwards).

N5.2.2. Duration (How long?). The basic distinction that needs to be established first is that between 'story time' and 'discourse time' (see Müller 1968 [1948]).

Typical discourse-time oriented questions are, "Can the text be read at one sitting?" (Poe's definition of a short story); "How does discourse time relate to story time?", i.e., "How long does it take to tell/read this episode" versus "How long does its action last?". Müller (1968 [1948]); Genette (1980 [1972]: 33-34); Rimmon-Kenan (1983: 44-45).

Some useful questions concerning story time are "What is the global time scale of the text?" (the 'amplitude' of story time) and "How does story time differ from discourse time?". For instance, while the story time of Joyce's Ulysses (650 pages of text) is 18 hours, the following few lines cover a story time of no less than ten centuries:

N5.2.3. In order to assess a narrative passage's speed or tempo, one compares story time and discourse time. The following major types of relationship occur:

N5.2.4. Frequency (How often?). Frequency analysis investigates a narrator's strategies of summative or repetitive telling. There are three main frequential modes:

Genette (1980 [1972]: 113-160); Rimmon-Kenan (1983: 46, 56-58); Toolan (1988: 61-62). Consider also the amusing metanarrative comment given by the self-conscious authorial narrator of Lodge's How Far Can You Go?:

N5.2.5. Conduct a frequency analysis of the following excerpts:

N5.3. Narrative Modes

N5.3.1. The main narrative modes (or ways in which an episode can be presented) basically follow from the frequential and durational relationships identified above. First, however, let us make the traditional distinction between 'showing' and 'telling' (often correlated with 'mimesis' and 'diegesis', respectively):

There are only two major narrative modes: scene and summary:

N5.3.2. In addition to the two major modes, there are two minor or supportive modes: description and comment. These modes are supportive rather than constitutive because no-one can tell a story using description and comment alone.

N6. Setting and fictional space

N6.1. No-one, so far, has given literary representations of space the same kind of scrutiny that has been expended on time, tense, and chronology. For a long time, scholars simply followed Lessing's dictum that literature was a 'temporal' art as opposed to 'spatial' arts like painting and sculpture. Thus, for a long time, the general assumption was that a verbal narrative's setting simply is not as important as its temporal framework and chronology.

This was an unfortunate conclusion, however. In an important article on 'chronotopes' (literally, 'time spaces'), Bakhtin (1981b [1973]) drew attention to the fact that time and space in narrative texts are actually very closely correlated (see Riffaterre 1996 for a practical application of the concept). In 1948, Josef Frank (1963 [1948]) isolated a number of stylistic techniques that create an effect of what he termed 'spatial form'. According to Stanzel (1984: ch. 5.2), space in fiction is distinct from space in the visual arts because space in fiction can never be presented completely. Describing the entire interior of a room, to the smallest visible detail, is an impossible (and rather boring) task, but the full depiction of a room in the medium of film clearly poses no problem at all. In verbal narrative, a room can only be described by referring to a small selection of more or less 'graphic' detail -- luckily, in the process of reading, readers will complete the 'verbal picture' by imagining the rest.

N6.2. For a point of departure, one might as well begin by noting that there is a close relationship between objects and spaces. A fishbowl is an object from our human point of view, but to the goldfish it is a space; similarly, a house is an object in a larger environment (a district, a city), but to its inhabitants it is a space to move or be in. In other words, what's space and what's an object in space is a matter of adopted perspective and environmental embeddedness. Hence our definition of literary space:

Literary space in this sense is more than a stable 'place' or 'setting' -- it includes landscapes as well as climatic conditions, cities as well as gardens and rooms, indeed, it includes everything that can be conceived of as spatially located objects and persons. Along with characters, space belongs to the 'existents' of a narrative (Chatman 1978).

See Bakhtin (1981b [1973]); Kahrmann et al. (1977: ch.4); Chatman (1978: 96-106, 138-145); Hoffmann (1978); Bronfen (1986); Ronen (1994: ch. 6); Würzbach (2001).

N6.3. Paralleling the distinction between 'story time' and 'discourse time' (N5.5.2), Chatman differentiates between 'story space' and 'discourse space':

More specifically still, the terms 'story-HERE' and 'discourse-HERE' can be used to identify the current deictic 'point of origin' in story space and discourse space, respectively.

Story-HERE and discourse-HERE in conjunction with story-NOW and discourse-NOW identify the story's current 'deictic center', i.e. the origin or zero point of the text's spatio-temporal co-ordinate system.

N6.4. As Ronen (1986; 1994) has pointed out, any description of space invokes a perception of space: apart from the reader's imaginative perception, this is either a narrator's perception, or a character's perception; both can be either actual perception or imaginary perception. For this reason, fictional space is evidently strongly correlated to focalization (N3.2).

Most important among the linguistic clues to spatial perception are expressions that signal the 'deictic orientation' of a speaking or perceiving subject (representing the current 'deictic center', N6.3) -- on the most basic level, expressions like near and far, here and there, left and right, up and down, come and go, etc. Significant oppositional spaces are city vs. country, civilization vs. nature, house vs. garden, transitional space vs. permanent space, and public space vs. private space. All these spaces are culturally defined (Baak 1983: 37) and therefore variable; often, they are also very clearly associated with attitudinal stances and value judgments.

Methodologically, the most promising approach towards the semantics of fictional space is to gather the isotopies (P3.6) correlating deictic expressions, spatial opposites, and value judgments, and to identify the propositions that link the common semantic denominators involved. To practice this type of analysis, try your hand on some of the examples quoted below.

N6.5. Semantically charged space. What makes an inquiry into the semantics of literary space so promising is the fact that spatial features can significantly influence characters and events. This is often referred to as the 'semanticization' or semantic charging of space. Here are some examples:

In the Joyce passage, the spatial details of the boy's journey to the bazaar named "Araby" (a name that signifies an exotic foreign space) foreshadow his frustrating experience there. The emotive connotations of "Araby" ("the magical name") are partly mirrored, and partly contrasted in the drab Dublin environment through which he passes. (Hint: consider also the initiation aspects of this story -- N3.3.4)

This is the famous introductory description of the "valley of ashes" in Fitzgerald's The Great Gatsby (ch. 2), later the scene of a tragic car accident.

N6.6. Representations of space should always be related to the story's underlying narrative situation. Consider the two examples below:

N7. Characters and Characterization

Characterization analysis investigates the ways and means of creating the personality traits of fictional characters. The basic analytical question is, Who (subject) characterizes whom (object) as being what (as having which properties). For a general introduction, see Chatman (1978: 107-133); Rimmon-Kenan (1983: 59-70); Pfister (1988: ch. 5); Marglin (1989); Bonheim (1990: ch. 17); Fokkema (1991); Nieragden (1995); Schneider (2000); Culpeper (2001) [the latter two are cognitive approaches towards character].

N7.1. Characterization analysis focuses on three basic parameters: (1) narratorial vs. figural characterization (identity of characterizing subject: narrator or character?); (2) explicit vs. implicit characterization (are the personality traits attributed in words, or are they implied by somebody's behavior?); (3) self-characterization (auto-characterization) vs. altero-characterization (does the characterizing subject characterize himself/herself or somebody else?).

N7.2. For a reasonably complete model of the system of dramatic characterization techniques, we will use a modified version of Pfister's famous tree diagram (1988: 184). (See D8.3 for a discussion of the modifications made.)


N7.3. In figural characterization, the characterizing subject is a character. On the level of explicit characterization, a character either characterizes him- or herself, or some other character. The reliability or credibility of a character's judgment largely depends on pragmatic circumstances. (1) Auto-characterization is often marked by face- or image-saving strategies, wishful thinking, and other "subjective distortions" (Pfister 1988: 184 -- consider, e.g., lonely hearts ads, letters of applications etc.). (2) Altero-characterization is often heavily influenced by social hierarchies and "strategic aims and tactical considerations" (Pfister 1988: 184), especially when the judgment in question is a public statement made in a dialogue (as opposed to when it is made in a character's interior monologue -- N8.9), and even more so when the person characterized is present (in praesentia -- obvious case: how advisable is it to criticize a tyrant?).

N7.4. An explicit characterization is a verbal statement that ostensibly attributes (i.e., is both meant to and understood to attribute) a trait or property to a character who may be either the speaker him- or herself (auto-characterization), or some other character (altero-characterization). Usually, an explicit characterization consists of descriptive statements (particularly, sentences using be or have as verbs) which identify, categorize, individualize, and evaluate a person. Characterizing judgments can refer to external, internal, or habitual traits -- "John has blue eyes, is a good-hearted fellow, and smokes a pipe". Note that while an 'explicit' characterization is a verbal characterization, the expressions used may be quite vague, allusive, or even elliptical (as in "he is not a person you'd want to associate with"). See Srull and Wyer (1988) for a theory of character attribution in social cognition, especially their use of the concepts 'identification', 'categorization', and 'individualization'. Example:

On the one hand, this is Katie's explicit characterization of Martha ("funny", "little"); at the same time it is also an implicit self-characterization, indicative of Katie's patronizing arrogance.

N7.5. An implicit characterization is a (usually unintentional) auto-characterization in which somebody's physical appearance or behavior is indicative of a characteristic trait. X characterizes him- or herself by behaving or speaking in a certain manner. Nonverbal behavior (what a character does) may characterize somebody as, for instance, a fine football player, a good conversationalist, a coward, or a homosexual, while verbal behavior (the way a character speaks, or what a character says in a certain situation) may characterize somebody as, for instance, having a certain educational background (jargon, slang, dialect), as belonging to a certain class of people (sociolect), or as being truthful, evasive, ill-mannered, etc. Characters are also implicitly characterized by their clothing, their physical appearance (e.g., a hunchback) and their chosen environment (e.g., their rooms, their pet dogs, their cars).

Generally speaking, all explicit characterizations are always also implicit auto-characterizations (why?). Occasionally, an implicit auto-characterization can sharply clash with an explicit auto-characterization.

N7.6. The implicit self-characterization of a narrator is always a key issue in interpretation. Is the narrator omniscient? competent? opinionated? self-conscious? well-read? ironic? reliable? See Genette (1980: 182-185); Lanser (1981); Rimmon-Kenan (1983: 59-67, 100-103); Stanzel (1984: 150-152); Nünning (1997; 1998; 1999).

Some theorists make an explicit distinction between 'mimetic (un)reliability' and 'evaluative' or 'normative (un)reliability': "a narrator may be quite trustworthy in reporting events but not competent in interpreting them, or may confuse certain facts but have a good understanding of their implications" (Lanser 1981: 171). According to Cohn (1999: ch. 8), Thomas Mann's Tod in Venedig is told by a mimetically reliable but normatively unreliable narrator. See also Wall (1994), Nünning (1990; 1999); Yacobi (2000).

N7.7. E.M. Forster's distinction between flat characters and round characters concerns the psychological depth or sophistication of a person's perceived character traits:

N7.8. Here is a brief list of functionally determined character types (to be expanded):

N7.9. A text's system of denomination or naming conventions is the specific set of naming strategies used to identify and subsequently to refer to its characters. Since naming patterns often dovetail with characterization, point of view or focalization, they merit close stylistic analysis. Key questions are:

Uspensky (1973) [first close analysis of point-of-view aspects of naming]; Genette (1988 [1983]) [discussion of character identification in 19C and 20C story incipits]; Moore (1989) [naming conventions in James's What Maisie Knew]; Fludernik (1996: 246-48); Emmott (1997) [major study especially focusing on pronouns]; Collier (1999) [naming in Patrick White).

N8. Discourses: representations of speech, thought and consciousness

N8.1. With respect to verbal narratives, narrative discourse is the oral or written text produced by an act of narrating. As Dolezel puts it, "Every narrative text T is a concatenation and alternation of DN [narrator's discourse] and DC [character's discourse]" (Dolezel 1973: 4). In principle, therefore, a narrative text can be subdivided into

Although this is a useful distinction, there are many transitional and borderline phenomena such as 'narrative report of discourse', 'psychonarration', 'narrated perception', 'coloring', etc. (see below). See Dolezel (1973: Introduction); Cohn (1978: 21-57), Genette (1980 [1972]: 164-169; 1988 [1983]: 18, 43, 61-63, 130); Lintvelt (1981: ch. 4.6.2).

N8.2. When the narrative of events includes (or shifts to) a narrative of words we encounter a patchwork structure that is addressed by quotation theory:

According to Genette (1988 [1983]: 60-63), a character's consciousness can either be rendered as narrative of events or, via conventionalized 'verbalization', as narrative of words.

N8.3. A special subset of diegetic statements is 'attributive discourse':

In general, introductory tags co-occur with 'direct' and 'indirect discourse', and parenthetical tags co-occur with direct and 'free indirect discourse' (see examples below). See Page (1973: ch. 2); Prince (1978); Bonheim (1982: ch. 5 [historical and stylistic features of inquits]; Banfield (1982: ch. 1.3.1, 2.2, 2.3); Neumann (1986 [ambiguous forms in Austen]); Collier (1992b: ch. 11 [comprehensive survey, but restricted to direct discourse inquits]); Fludernik (1993a: ch. 5.2 [tag phrases and free indirect discourse]).

N8.4. As regards styles of discourse representation, we are going to distinguish the three traditional basic forms: the 'direct' style, the 'free indirect' style, and the 'indirect' style. The following table lists the general characteristics of each style; more detailed definitions and some subforms follow below.

Type Example Characteristics
direct discourse Mary said/thought: "What on
earth shall I do now?"
quoted speech formally independent of
quoting frame
free indirect
discourse (FID)
What on earth should she do
mixture of deictic elements: original
expressivity combines with person/tense
system of framing discourse
Mary wondered what she should do. diegetically oriented report; the quoted part
is a subordinate clause controlled by the
narratorial frame

N8.5. Direct discourse styles.

See Cohn (1978: 58-98); Quirk et al. (1985: ch. 14.28-14.29); Leech and Short (1981: ch. 10); Bonheim (1982: ch. 4); Sternberg (1982b); Short (1991); Fludernik (1993a: ch. 8).

N8.6. Free indirect discourse styles.

Note: Although many theorists understand 'free' to mean free of a reporting clause, as in the definition given above, recent commentators recognize that free indirect discourse does in fact often collocate with 'parenthetical' attributive discourse. It seems appropriate, therefore, to distinguish between 'tagged' and 'untagged' free indirect discourse (cf. Wales 1989: 189; Collier 1992b: 168).

See Bally (1912 [f.i.d. and French imparfait]); Pascal (1977 ['dual voice' theory]); McHale (1978 [excellent overview]); Banfield (1982 [controversial generative-grammar account, finding that f.i.d sentences are 'unspeakable']); Rimmon-Kenan (1983: 110-116); Cohn (1978: 99-140 [consonant and dissonant uses of f.i.d.]); Toolan (1988: 119-137); Short (1991 [speech-act parameters]); Fludernik (1993b [the most comprehensive account to date]; Tammi and Tommola, eds (2003) [variants and functions of f.i.d. across European languages]. Examples:

N8.7. Indirect discourse styles.

See Quirk et al. (1985: ch. 14.30-14.35); McHale (1978: 258-260); Leech and Short (1981: ch. 10); Banfield (1982: ch. 1); Short, Semino and Culpeper (1996); Toolan (1999).

N8.8. To conclude this section, we will briefly turn to terms that specifically identify certain styles of representing 'inside views' (Booth 1961: 163-168) into a character's mind. Presenting the mental processes of characters, their thoughts and perceptions, their memories, dreams, and emotions became a prime challenge for late 19C and early 20C novelists. Among the authors who became strongly interested in what was soon called 'stream of consciousness art', 'literary impressionism', 'novel of consciousness', etc, were D.H. Lawrence, Virginia Woolf, James Joyce, William Faulkner, Dorothy Richardson, Patrick White (and many others). See Cohn (1978) for an excellent introduction to the subject.

See W. James (1950 [1890]: ch. 9), Sinclair (1990 [1918]); Humphrey (1954); Steinberg (1973); Cohn (1978), Chatman (1978: 186-195); Smuda (1981); Toolan (1988: 128).

N8.9. The main techniques of representing the sound and rhythm of a character's stream of consciousness are 'interior monologue', 'direct thought', and 'free indirect thought'. Direct thought and free indirect thought have already been defined in N8.5 and N8.6, above. Interior monologue is a special case of direct thought:

See Humphrey (1954); Steinberg (1973); Chatman (1978: 178-195); Cohn (1978: 58-98); Cohn and Genette (1992 [1985]).

N8.10. Earlier forms of extended direct thought are usually identified by the term 'soliloquy' (originally a term in drama theory meaning a monologue uttered aloud in solitude, D3.4):

N8.11. Psychological states are usually rendered by diegetic statements, especially the two forms known as 'psychonarration' and 'narrated perception':

N8.12. 'Mind style' is a general term for a character's or a narrator's typical patterns of mentation:

N8.13. Following Hough (1970), the term coloring is occasionally used to refer to the local coloring (also 'tainting' or 'contamination') of the narrator's style by a character's diction, dialect, sociolect, or idiolect, often serving a comic or ironical purpose. Colouring is most functional when the narrator's and the character's voices are equally distinctive (typically, in the fiction of Austen, James, Lawrence, and Mansfield). Hough 1970; Page 1973: ch. 2; McHale 1978: 260-262; Stanzel 1984: 168-184; Fludernik 1993: 334-338. Example:

N9. A Case Study: Alan Sillitoe's "The Fishing Boat Picture"

(In the following, all page number references are to the reprint of Sillitoe's story in The Penguin Book of Modern British Short Stories, ed. Malcolm Bradbury, London: Penguin, 1988, 135-149. The story was originally published in 1959.)

N9.1. Like many first-person narratives, Sillitoe's "Fishing-Boat Picture" is a fictional autobiography. Harry is a mature narrator who looks back on his past life. Although he is only fifty-two at the time of writing the story, he feels his life is all but over. Like many first-person narrators, he has become not only older but also wiser. Looking back on his life, he realizes that he made many mistakes, especially in his behavior towards his wife Kathy. The story's first-person narrative situation is uniquely suited for presenting Harry's insights about his wasted life.

N9.2. The story is told in a straightforwardly chronological manner, and its timeline can be established quite accurately (cp N4.8). The story's action begins with Harry's and Kathy's "walk up Snakey Wood" (135). Kathy leaves Harry after six years, when he is thirty (136); so, at the beginning he must be twenty-four. Since "it's [...] twenty-eight years since I got married" (135), the narrating I's current age must be fifty-two. Kathy's weekly visits begin after a ten-year interval (139), when Harry is forty. Kathy's visits continue for six years (147), and when she dies, terminating the primary story line, the experiencing I is forty-six. A number of historical allusions indicate that Harry's and Kathy's final six years are co-extensive with World War II (140, 147). The narrative act itself takes place in 1951, six years after Kathy's death .

N9.3. The story's action episodes focus on Kathy, picking out their first sexual encounter, the violent quarrel that makes her run away, her return ten years later, her ensuing weekly visits, the repeated pawnings of the fishing-boat picture, and her death and funeral. Throughout their relationship, Harry "doesn't get ruffled at anything" (136), and he remains unemotional and indifferent to the point of lethargy. To the younger Harry, marriage means "only that I changed one house and one mother for a different house and a different mother" (136). Although he never sets foot from Nottingham (139), his main idea of a good time is reading books about far-away countries like India (137) and Brazil (139). He cannot even cry at Kathy's funeral ("No such luck", 148). And yet, her ignoble death -- in a state of drunkenness she is run over by a lorry -- causes a change in him. Now he cannot forget her as he did after she left him (139-140); the only thing he can do is obsessively review the mistakes he has made. In the final retrospective epiphany, he realizes three things with devastating clarity: that he loved Kathy but never showed it, that he was insensitive to her need for emotional involvement and communication, and that her death robbed him of a purpose in life.

N9.4. The theme of becoming aware of one's own flaws can be treated well in a first-person narrative situation. Unlike the ordinary well-spoken authorial narrator, who cannot himself be present as a character in the story, Harry's working-class voice and diction is a functional and characteristic feature in Sillitoe's story. His self-consciousness in telling the story ("I'd rather not make what I'm going to write look foolish by using dictionary words", 135) and his involvement in the story support the theme of developing self-recognition. Whereas Harry's story is an account of personal experience, an authorial narrator knows everything from the beginning and cannot normally undergo any personal development (unless this is caused by the act of telling itself).

N9.5. The theme of recollection and reflection that runs through Sillitoe's story would, however, be well manageable in a figural narrative situation, in which Harry could serve not as a narrator, but as a third-person character (an internal focalizer, a reflector figure) in the act of recollecting his past life. In fact, in a modernist short story, both main characters could be used for purposes of variable and multiple focalization. A figural beginning would filter the action through Harry's consciousness and would begin medias in res, perhaps using an incipit such as the following:

N9.6. This is clearly a more immediate beginning than Harry's self-conscious metanarrative commentary ("Take that first sentence", 135); on the other hand, a figural story usually proceeds in a more associative and less controlled manner than a first-person story. Moreover, while a figural story tends to focus on a scenic slice of life, "The Fishing-Boat Picture" spans a story-time of at least twenty-two years. In fact, Harry's telling his own story helps him think about his life and clarify his own thoughts and judgments. A reflector figure, in contrast, is not a narrator, and cannot address a narratee. It is important to Harry not only to tell his story to an anonymous audience but in a sense also to himself. The text's dialogic quality comes out in one of its key passages:

Here Harry explicitly "keeps telling himself", "answer[s]" his own indictment, and "maintain[s]" a position, stressing the self-reflective and auto-therapeutic function of his narrative. In fact, the devastating judgment "I was born dead" takes up Kathy's calling him a "dead-'ed" (137) in the quarrel that leads to their separation. Unfortunately, now that he has learned his lesson, it is "too bloody late".

N9.7. As a working-class story with occasional snippets of slang and dialect, its references to the characters' ordinary lives, their brief bouts of passion, aggression and violence ("this annoyed me, so I clocked her one", 137), Sillitoe's story is neither sentimental nor overly didactic, nor does it offer an idealized portrayal of working-class characters; it certainly does not allow the reader to feel superior. On the contrary, the protagonist's matter-of-fact account creates a strong sense of empathy, and his reflections on a wasted past and a meaningless future clearly express a general human condition.


N10. References

Abbott, H. Porter. 2002.
The Cambridge Introduction to Narrative. Cambridge: Cambridge UP.
Aczel, Richard. 1998a.
"Hearing Voices in Narrative Texts". New Literary History 29.3: 467-500.
---. 1998b.
How to Write an Essay. Stuttgart: Klett.
Adams, Jon-K. 1996.
Narrative Explanation: A Pragmatic Theory of Discourse. Frankfurt: Lang.
Applebee, Arthur. 1978.
The Child's Concept of Story: Ages Two to Seventeen. Chicago: UP.
Austin, John. 1962 [1955].
How to do things with Words. Ed. J.O. Urmson. Oxford: Clarendon.
Baak, Jan Joost, van. 1983.
The Place of Space in Narration: A Semiotic Approach to the Problem of Literary Space. Amsterdam: Rodopi.
Bakhtin, Mikhail. 1981a [1973].
The Dialogic Imagination: Four Essays by M. M. Bakhtin. Ed Michael Holquist. Austin: U of Texas P.
---. 1981b [1973].
"Forms of Time and the Chronotope in the Novel". The Dialogic Imagination: Four Essays by M. M. Bakhtin. Ed Michael Holquist. Austin: U of Texas P. 84-258.
Baker, Mary J. 1992.
"Metadiegetic Narrative in the Heptameron". Studies in the Literary Imagination 25.1: 97-101.
Bal, Mieke. 1978.
"Mise en abyme et iconicité". Littérature 29: 116-128.
---. 1981a.
"Notes on Narrative Embedding". Poetics Today 2.2: 41-59.
---. 1981b.
"The Laughing Mice or: On Focalization." Poetics Today 2.2: 203-214.
---. 1983.
"The Narrating and the Focalizing: A Theory of the Agents in Narrative." Style 17.2: 234-269.
---. 1985.
Narratology. Trans. Christine van Boheemen. Toronto: U of Toronto P.
---. 1987.
Lethal Love : Feminist Literary Readings of Biblical Love Stories. Indiana: Indiana UP.
---. 1988.
Death and Dissymmetry: The Politics of Coherence in the Book of Judges. Chicago: U. of Chicago P.
---. 1990.
"The Point of Narratology". Poetics Today 11: 727-753.
---. 1991.
On Story-Telling: Essays in Narratology. Ed. David Jobling. Sonoma, CA: Polebridge P.
---, ed. 2004.
Narrative Theory: Critical Concepts in Literary and Cultural Studies. London: Routledge.
Bally, Charles. 1912.
"Le style indirect libre en francais moderne". Germanisch-Romanische Monatsschrift 4: 549-556; 597-606.
Banfield, Ann. 1982.
Unspeakable Sentences: Narration and Representation in the Language of Fiction. London: Routledge.
---. 1987.
"Describing the Unobserved: Events Grouped Around an Empty Center". In Fabb, ed. (1987). 265-285.
Barth, John. 1968.
Lost in the Funhouse. London: Penguin.
---. 1984 [1981].
"Tales within Tales Within Tales". The Friday Book: Essay and Other Non-Fiction. New York: Putnam. 218-238.
Barthes, Roland. 1975 [1966].
"An Introduction to the Structural Analysis of Narrative." New Literary History 6: 237-272.
---. 1982.
"The Reality Effect". In Todorov, Tzvetan, ed. French Literary Theory Today. Cambridge: Cambridge UP. 11-17.
Beja, Morris. 1971.
Epiphany in th Modern Novel. London: Peter Owen.
---. 1984.
"Epiphany and Epiphanies". In Bowen, Zack; Carens, James F., A Companion to Joyce Studies. London: Greenwood.
Berne, Eric. 1964.
Games People Play: The Psychology of Human Relationships. New York: Grove.
Bickerton, Derek. 1995.
Language and Human Behavior. Seattle: U of Washington P.
Blin, Georges. 1954.
Stendhal et les problèmes du roman. Paris: Corti.
Bonheim, Helmut. 1982.
The Narrative Modes: Techniques of the Short Story. Cambridge: Brewer.
---. 1990.
Literary Systematics. Cambridge: Brewer.
Buchholz, Sabine. 2003.
Narrative Innovationen in der modernistischen britischen Short Story. Trier: WVT.
---. 2004.
"Short Stories of Female Initiation 1910-1940". Görtschacher, Wolfgang; Klein, Holger, eds. Tale, Novella, Short Story: Currents in Short Fiction. Tübingen: Stauffenburg. 91-102.
Booth, Wayne C. 1961.
The Rhetoric of Fiction. Chicago: U of Chicago P.
Bordwell, David. 2004.
"Neo-Structuralist and the Functions of Filmic Storytelling." In Ryan, ed (2004). 203-219.
Bortolussi, Marisa; Dixon, Peter; 2003.
Psychonarratology: Foundations for the Empirical Study of Literary Response. Cambridge: Cambridge UP.
Branigan, Edward. 1992.
Narrative Comprehension and Film. London: Routledge.
Bremond, Claude. 1970.
"Morphology of the French Folktale." Semiotica 2: 347-275.
Bronfen, Elisabeth. 1986.
Der literarische Raum: Eine Untersuchung am Beispiel von Richardsons Romanzyklus Pilgrimage. Tübingen: Niemeyer.
Bronzwaer, W.J.M. 1970.
Tense in the Novel. Groningen: Wolters.
Brooks, Cleanth; Warren, Robert Penn. 1959.
Understanding Fiction. New York: Appleton.
Brooks, Peter. 1984.
Reading for the Plot: Design and Intention in Narrative. New York: Random House.
---; Gewirtz, Paul, eds. 1996.
Law's Stories: Narrative and Rhetoric in the Law. New Haven: Yale UP.
Bruner, Jerome. 1986.
Actual Minds, Possible Worlds. Cambridge: Harvard UP.
Bühler, Karl.
1990 [1934]. Theory of Language: The Representational Function of Language. Trans. Donald Fraser Goodwin. Philadelphia: Benjamins.
Casparis, Christian P. 1975.
Tense Without Time. Berne: Francke.
Chatman, Seymour. 1978.
Story and Discourse: Narrative Structure in Fiction and Film. Ithaca and London: Cornell UP.
---. 1986.
"Characters and Narrators: Filter, Center, Slant, and Interest-Focus." Poetics Today 7.2: 189-204.
---. 1990.
Coming to Terms: The Rhetoric of Narrative in Fiction and Film. Ithaca: Cornell UP.
Church, Jennifer.
2000. "'Seeing As' and the Double Bind of Consciousness". Journal of Consciousness Studies 7.8-9: 99-111.
Ci, Jiwei. 1988.
"An Alternative to Genette's Theory of Order." Style 22.1: 18-41.
Cohn, Dorrit. 1978.
Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton: Princeton UP.
---. 1981.
"The Encirclement of Narrative: On Franz Stanzel's Theorie des Erzählens." Poetics Today 2: 157-82.
---. 1990.
"Signposts of Fictionality: A Narratological Perspective". Poetics Today 11.4: 775-804.
---. 1993.
"'I doze and wake': The Deviance of Simultaneous Narration". In Foltinek et al eds (1993). 9-23.
---. 1999.
The Distinction of Fiction. Baltimore: Johns Hopkins UP.
---; Genette, Gérard. 1992 [1985].
"A Narratological Exchange". In Fehn, Ann; Hoesterey, Ingeborg; Tatar, Maria, eds. Neverending Stories: Toward a Critical Narratology. Princeton: Princeton UP. 258-266.
Collier, Gordon. 1992a.
"Style at the Interface: Speech-tags". REAL: Yearbook of Research in English and American Literature 8: 33-107.
---. 1992b.
The Rocks and Sticks of Words: Style, Discourse and Narrative Structure in the Fiction of Patrick White. Amsterdam: Rodopi.
---. 1999.
"Apparent Feature-Anomalies in Subjectivized Third-Person Narration". In Pier, ed. (1999). 129-152.
Cordesse, Gérard. 1988.
"Narration et focalisation". Poétique 76: 487-498.
Cortazzi, Martin.
1993. Narrative Analysis. London: The Falmer P.
Coste, Didier. 1989.
Narrative as Communication. Minneapolis: U of Minnesota P.

Culler, Jonathan. 1975a.
Structuralist Poetics. London: Routledge.
---. 1975b.
"Defining Narrative Units". In Fowler, Roger, ed. Style and Structure in Literature. Oxford: Blackwell. 123-142.
---. 1981.
The Pursuit of Signs: Semiotics, Literature, Deconstruction. Ithaca: Cornell UP.
---. 1988.
Framing the Sign: Criticism and Its Institutions. Oxford: Blackwell.
Culpeper, Jonathan. 2001.
Language and Characterisation: People in Plays and Other Texts. Harlow: Longman.
Currie, Mark. 1998.
Postmodernist Narrative Theory. London: Macmillan.
Dällenbach, Lucien. 1980.
"Reflexivity and Reading". New Literary History 11: 435-449.
Dancygier, Barbara. 2012.
The Language of Stories: A Cognitive Approach. Cambridge: CUP.
Deleyto, Celestino. 1996 [1991].
"Focalisation in Film Narrative". In Onega and Garcia Landa, eds. (1996). 217-233.
Dolezel, Lubomír. 1973.
Narrative Modes in Czech Literature. Toronto: U of Toronto P.
Dorfman, Eugene. 1969.
The Narreme in Medieval Romance Epic: An Introduction to Narrative Structures. Toronto: U of Toronto P.
Duchan, Judith F. et al, eds. 1995.
Deixis in Narrative. Hillsdale: Erlbaum.
Dujardin, Edouard. 1991 [1887].
The Bays Are Sere [Les lauriers sont coupés]. Intr. and trans. by Anthony Suter. London: Libris.
---. 1931.
"Interior Monologue". In Dujardin (1997 [1887]). 81-147.
Duyfhuizen, Bernard. 1992.
Narratives of Transmission. London: Associated UP.
Edmiston, William F. 1989.
"Focalisation and the First-Person Narrator: A Revision of the Theory". Poetics Today 10.4: 729-744.
---. 1991.
Hindsight and Insight. University Park: Pennsylvania UP.
Emmott, Catherine. 1997.
Narrative Comprehension: A Discourse Perspective. Oxford: Clarendon.
Fabb, Nigel, et al, eds. 1987.
The Linguistics of Writing: Arguments between Language and Literature. New York: Methuen.
Fauconnier, Gilles. 1994.
Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge, Mass.: CUP.
Fehr, Bernhard. 1938.
"Substitutionary Narration and Description: A Chapter in Stylistics". English Studies 20: 97-107.
Fieguth, Rolf. 1973.
"Zur Rezeptionslenkung bei narrativen und dramatischen Texten". Sprache im technischen Zeitalter 47: 186-201.
Fillmore, Charles J. 1981.
"Pragmatics and the Description of Discourse". In Cole, Peter, ed. Radical Pragmatics. New York: Academic P. 143-166.
Fleischmann, Suzanne. 1990.
Tense and Narrativity. London: Routledge.
Flint, Christopher. 1998.
Family Fictions: Narrative and Domestic Relations in Britain, 1688-1798. Stanford: UP.
Fludernik, Monika. 1992.
"The Historical Present Tense in English Literature: An Oral Pattern and its Literary Adaptation". Language and Literature 17: 77-107.
---. 1993a.
The fictions of language and the languages of fiction: the linguistic representation of speech and consciousness. London: Routledge.
---. 1993b.
"Second person fiction: narrative you as addressee and/or protagonist". Arbeiten aus Anglistik und Amerikanistik 18.2: 217-247.
---. 1996.
Towards a 'Natural' Narratology. London: Routledge.
---. 1999.
"The Genderization of Narrative". In Pier, John, ed. GRAAT: Revue des Groupes de Recherches Anglo-Américaines de L'Université Francois Rabelais de Tours 21: 153-175.
---; Olson, Greta, eds. 2004.
In the Grip of the Law: Trials, Prisons and the Space between. Frankfurt: Lang.
Fish, Stanley. 1980.
Is There a Text in This Class? The Authority of Interpretive Communities. Cambridge, Mass.: Harvard UP.
Fokkema, Aleid. 1991.
Postmodern Characters: A Study of Characterization in British and American Postmodern Fiction. Amsterdam: Rodopi.
Foltinek, Herbert; Riele, Wolfgang; Zacharasiewicz, Waldemar, eds. 1993.
Tales and "their telling difference": Zur Theorie und Geschichte der Narrativik. Festschrift Franz K. Stanzel. Heidelberg: Winter.
Forster, Edward Morgan. 1976 [1927].
Aspects of the Novel. Harmondsworth: Penguin.
Fowler, Roger. 1977.
Linguistics and the Novel. London: Methuen.
---. 1983.
"Polyphony and Problematic in Hard Times". In Giddings, Robert, ed. The Changing World of Charles Dickens. London: Vision. 91-108.
---. 1986.
Linguistic Criticism. Oxford: Oxford UP.
Frank, Joseph. 1963 [1948].
"Spatial Form in Modern Literature". In The Widening Gyre: Crisis and Mastery in Modern Literature. New Brunswick: Rutgers UP. 3-63.
Freese, Peter. 1979.
"Über die Schwierigkeiten des Erwachsenwerdens: Amerikanische stories of initiation von Nathaniel Hawthorne bis Joyce Carol Oates". In Freese, Peter; Groene, Horst; Hermes, Liesel, eds. Die Short Story im Englischunterricht der Sekundarstufe II. Paderborn: Schöningh. 206-255.
Friedman, Norman. 1967 [1955].
"Point of View in Fiction: The Development of a Critical Concept". In Philip Stevick, ed. The Theory of the Novel. London: Free Press. 108-137.
Füger, Wilhelm. 1993.
"Stimmbrüche: Varianten und Spielräume narrativer Fokalisation". In Foltinek et al., eds. (1993). 43-60.
Galbraith, Mary. 1995.
"Deictic Shift Theory and the Poetics of Involvement in Narrative". In Duchan et al., eds. (1995). 19-59.
Genette, Gérard. 1980 [1972].
Narrative Discourse. Trans. Jane E. Lewin. Oxford: Blackwell.
---. 1988 [1983].
Narrative Discourse Revisited. Trans. Jane E. Lewin. Ithaca: Cornell UP.
---. 1993.
"Acts of Fiction." Fiction and Diction, tr. Catherine Porter. Ithaca, N.Y.: Cornell. 30-53.
---. 1997 [1982].
Palimpsests. Trans. Channa Newman and Claude Doubinsky. Lincoln: U of Nebraska P.
Gerrig, Richard J. 1993.
Experiencing Narrative Worlds: On the Psychological Activities of Reading. New Haven: Yale UP.
Gibson, Andrew. 1997.
Towards a Postmodern Theory of Narrative. Hemel Hempstead: Harvester.
Grice, H.P. 1975.
"Logic and Conversation". In Cole, Peter; Morgan, J., eds., Speech Acts. New York: Academic. 41-58.
Gutenberg, Andrea. 2000.
Mögliche Welten: Plot und Sinnstiftung im englischen Frauenroman. Heidelberg: Winter.
Hamburger, Käte. 1977 [1957].
Die Logik der Dichtung. Stuttgart: Klett.
Hayakawa, S.I. 1964.
Language in Thought and Action. New York: Harcourt.
Herman, David. 1994.
"Hypothetical Focalization". Narrative 2.3: 230-253.
---. 1997.
"Towards a Formal Description of Narrative Metalepsis". Journal of Literary Semantics 26.2: 132-152.
---. 1999.
"Introduction: Narratologies". In Herman, David, ed. (1999). 1-30.
---, ed. 1999.
Narratologies: New Perspectives on Narrative Analysis. Columbus: Ohio State UP.
---. 2002.
Story Logic: Problems and Possibilities of Narrative. Lincoln: U of Nebraska P.
---, ed. 2003.
Narrative Theory and the Cognitive Sciences. Stanford, CA.: Publications of the Center for the Study of Language and Information.
---. 2009.
"Beyond Voice and Vision: Cognitive Grammar and Focalization Theory". P. Hühn et al., eds. Point of View, Perspective, and Focalization. Modeling Mediation in Narrative (Narratologia 17). Berlin: de Gruyter. 119-42.
---; Jahn, Manfred; Ryan, Marie-Laure. 2005.
The Routledge Encyclopedia of Narrative Theory. London: Routledge.
Herman, Luc; Vervaeck, Bart. 2005.
Handbook of Narrative Analysis. Lincoln: U of Nebraska P.
Hoffmann, Gerhard. 1978.
Raum, Situation, erzählte Wirklichkeit. Stuttgart: Metzler.
Hogan, Patrick Colm. 2003.
The Mind and Its Stories: Narrative Universals and Human Emotion. Cambridge: Cambridge UP.
Holman, C. Hugh. 1972.
A Handbook to Literature. Indianapolis: Odyssey.
Hough, Graham. 1970.
"Narrative and Dialogue in Jane Austen". Critical Quarterly 12: 201-229.
Humphrey, Robert. 1954.
Stream of Consciousness in the Modern Novel. Berkeley: U of California P.
Hunter, Kathryn M. 1993.
Doctor's Stories: The Narrative Structure of Medical Knowledge. Princeton: UP.
Hühn, Peter. 2004.
"Transgeneric Narratology: Applications to Lyric Poetry." In Pier, John, ed. The Dynamics of Narrative Form: Studies in Anglo-American Narratology. Berlin: de Gruyter. 139-158.
Iser, Wolfgang. 1971.
"The Reading Process: A Phenomenological Approach". New Literary History 3.2: 279-99.
---. 1972.
Der implizite Leser: Kommunikationsformen des Romans von Bunyan bis Beckett. München: Fink.
---. 1976.
Der Akt des Lesens. München: Fink.
Jackendoff, Ray. 1983.
Semantics and Cognition. London: M.I.T. Press.
Jahn, Manfred. 1995.
"Narratologie: Methoden und Modelle der Erzähltheorie". In Nünning, ed. (1995). 29-50.
---. 1996.
"Windows of Focalization: Deconstructing and Reconstructing a Narratological Concept." Style 30.2: 241-267.
---. 1997.
"Frames, Preferences, and the Reading of Third-Person Narratives: Towards a Cognitive Narratology." Poetics Today 18.4: 441-468.
---. 1999.
"More Aspects of Focalization: Refinements and Applications". In Pier, John, ed. (1999). 85-110.
---. 2003.
'''Awake! Open your Eyes!' The Cognitive Logic of External and Internal Stories". In Herman, David, ed., Narrative and the Cognitive Sciences. Stanford, CA: CSLI Publications.
---. 2007.
"Focalization". The Cambridge Companion to Narrative. Ed. David Herman. Cambridge: CUP. 94-108.
--- ; Nünning, Ansgar. 1994.
"A Survey of Narratological Models". Literatur in Wissenschaft und Unterricht 27.4: 283-303.
Jakobson, Roman. 1960.
"Linguistics and Poetics". In Sebeok, Thomas E., ed. Style in Language. Sebeok. Cambridge, Mass: M.I.T. P. 350-377.
James, Henry. 1881.
Preface to The Portrait of a Lady. London: Penguin, 1970.
---. 1934.
The Art of the Novel: Critical Prefaces. Ed. Richard P. Blackmur. New York: Scribner's.
James, William. 1950 [1890].
The Principles of Psychology. Vol 1. New York: Dover.
Jong, Irene de. 2001.
"The Origins of Figural Narration in Antiquity". In Peer, Willie van; Chatman, Seymour, eds. New Perspectives on Narrative Perspective. Albany: State U of New York P. 67-81.
Jonnes, Denis. 1990.
Matrix of Narrative: Family Systems and Semiotics of Story. The Hague: Mouton.
Jost, Francois. 1989.
L'oeil-Caméra: Entre film et roman. 2nd ed. Lyon: Presses Universitaires.
Kablitz, Andreas. 1988.
"Erzählperspektive - Point of View - Focalisation". Zeitschrift für Romanische Sprache und Literatur 98: 237-255.
Kafalenos, Emma. 2004.
"Overview of the Music and Narrative Field". In Ryan, ed. (2004). 275-282.
Kahrmann, Cordula, et al. 1977.
Erzähltextanalyse: Eine Einführung in Grundlagen und Verfahren. 2 vols. Kronberg: Athenäum.
Kallmeier, Werner, et al. 1974.
Lektürekolleg zur Textlinguistik, I: Einführung. Frankfurt: Fischer Athenäum.
Kearns, Michael. 1999.
Rhetorical Narratology. Lincoln: U of Nebraska P.
Kenner, Hugh. 1978.
Joyce's Voices. Berkeley: U. of Berkeley P.
Kermode, Frank. 1979.
The Genesis of Secrecy. Cambridge, Mass.: Cambridge UP.
Kindt, Tom; Müller, Hans-Harald. 1999.
"The 'Implied Author': Explication and Use of a Controversial Concept". Internet document
---; Müller, Hans-Harald, eds. 2003.
What is Narratology? Questions and Answers Regarding the Status of a Theory. Berlin: de Gruyter.
Kintsch, Walter. 1976.
"Memory For Prose". In Cofer, C., ed., The Structure of Human Memory. San Francisco: Freeman.
Kozloff, Sarah. 1988.
Invisible Storytellers: Voice-over Narration in American Fiction Film. Berkeley: U of California P.
Kuhn, Markus. 2009.
"Film Narratology: Who Tells? Who Shows? Who Focalizes? Narrative Mediation in Self-Reflexive Fiction Films". P. Hühn et al., eds. Point of View, Perspective, and Focalization. Modeling Mediation in Narrative (Narratologia 17). Berlin: de Gruyter. 259-78.
Labov, William. 1972.
Language in the Inner City: Studies in the Black English Vernacular. Philadelphia: U of Pennsylvania P.
Lanser, Susan Sniader. 1981.
The Narrative Act: Point of View in Prose Fiction. Princeton: Princeton UP.
---. 1992.
Fictions of Authority: Women Writers and Narrative Voice. London: Cornell UP.
---. 1995.
"Sexing the Narrative: Propriety, Desire, and the Engendering of Narratology". Narrative 3.1: 85-94.
Leech, Geoffrey N.; Short, Michael. 1981.
Style in fiction: a linguistic introduction to English fictional prose. London: Longman.
Lejeune, Philippe. 1989.
On Autobiography. Trans. Katherine Leary. Minneapolis: U of Minnesota P.
Lintvelt, Jaap. 1978.
"Modèle discursif du récit encadré: Rhétorique et idéologie dans les Illustres Francaises de Robert Challe. Poétique 35: 352-365.
---. 1981.
Essai de Typologie narrative: le "point de vue". Paris: Corti.
Löschnigg, Martin. 1999.
"Narratological Categories and the (Non-) Distinction between Factual and Fictional Narratives". In Pier, John, ed. (1999). 31-48.
Lubbock, Percy. 1957 [1921].
The Craft of Fiction. New York: Viking.
Ludwig, Hans-Werner, ed. 1989.
Arbeitsbuch Romananalyse. Tübingen: Narr.
Malina, Debra. 2002.
Breaking the Frame: Metalepsis and the Construction of the Subject. Columbus: Ohio State UP.
Mandler, Jean M.; Johnson, Nancy S. 1977.
"Remembrance of Things Parsed: Story Structure and Recall". Cognitive Psychology 9: 11-151.
Margolin, Uri. 1989.
"Structuralist Approaches to Character in Narrative". Semiotica 75: 1-24.
---. 1996.
"Telling Our Story: On 'We' Literary Narratives". Language & Literature: Journal of the Poetics & Linguistics Association 5.2: 115-33.
---. 1999.
"Of What is Past, is Passing, or to Come: Temporality, Aspectuality, Modality, and the Nature of Literary Narrative". In Herman, ed. (1999). 142-166.
---. 2000.
"Telling in the Plural: From Grammar to Ideology". Poetics Today 21.3: 591-618.
---. 2009.
"Focalization: Where Do We Go from Here?" Ed. P. Hühn, et al., Point of View, Perspective, and Focalization. Modeling Mediation in Narrative (Narratologia 17). Berlin: de Gruyter. 48-58.
McClary, Susan. 1997.
"The Impromptu that Trod on a Loaf: or How Music Tells Stories". Narrative 5.1: 20-35.
McHale, Brian. 1978.
"Free Indirect Discourse: A Survey of Recent Accounts". PTL: Journal for Descriptive Poetics and Literature 3: 249-287.
---. 1987.
Postmodernist Fiction. London: Routledge.
---. 1992.
Constructing Postmodernism. London: Routledge.
Mezei, Kathy, ed. 1996.
Ambiguous Discourse: Feminist Narratology and British Women Writers. Chapel Hill and London: The U of North Carolina P.
Mihailescu, Calin-Andrei; Harmaneh, Walid, eds. 1996.
Fiction Updated: Theories of Fictionality, Narratology, and Poetics. Toronto: U of Toronto P.
Moore, Gene M. 1989.
"Focalization and Narrative Voice in What Maisie Knew". Language and Style 22.1: 3-24.
Müller, Günther. 1968 [1948].
"Erzählzeit und erzählte Zeit". Morphologische Poetik: Gesammelte Aufsätze. Ed. Elena Müller. Darmstadt: Wissenschaftliche Buchgesellschaft. 269-286.
Neisser, Ulric.
Cognitive Psychology. New York: Meredith, 1967.
Nell, Victor. 1988.
Lost in a Book: The Psychology of Reading For Pleasure. New Haven: Yale UP.
Nelles, William. 1997.
Frameworks: Narrative Levels and Embedded Narrative. Frankfurt: Lang.
Neumann, Anne Waldron. 1986.
"Characterization and Comment in Pride and Prejudice: Free Indirect Discourse and 'Double-Voiced' Verbs of Speaking, Thinking, Feeling". Style 20.3: 364-394.
Niederhoff, Burkhart. 2013.
"Focalization". Hühn, Peter et al. (eds.), the living handbook of narratology. Hamburg: Hamburg University.
Nieragden, Göran. 1995.
Figurendarstellung im Roman: eine narratologische Systematik. Trier: WVT.
Nischik, Reingard. 1991.
Mentalstilistik. Tübingen: Narr.
Nünning, Ansgar. 1989.
Grundzüge eines kommunikationstheoretischen Modells der erzählerischen Vermittlung: Die Funktion der Erzählinstanz in den Romanen George Eliots. Trier: WVT.
---. 1990.
"'Point of view' oder 'Focalisation': Über einige Grundlagen und Kategorien konkurrierender Modelle der erzählerischen Vermittlung. Literatur in Wissenschaft und Unterricht 23.3: 249-268.
---. 1993.
"Renaissance eines anthropomorphisierten Passepartouts oder Nachruf auf ein literaturkritisches Phantom? Überlegungen und Alternativen zum Konzept des 'implied author'". Deutsche Vierteljahrsschrift 67.1: 1-25.
---, ed. 1995.
Literaturwissenschaftliche Theorien, Modelle, Methoden. Trier: WVT.
---. 1997.
"'But why will you say that I am mad?': On the Theory, History, and Signals of Unreliable Narration in British Fiction". Arbeiten aus Anglistik und Amerikanistik 22.1: 83-105.
---, ed. 1998.
Unreliable Narration: Studien zur Theorie und Praxis unglaubwürdigen Erzählens. Trier: WVT.
---. 1999.
"Reconceptualizing the Theory and Generic Scope of Unreliable Narration". In Pier, John, ed. (1999). 63-84.
---. 2000.
"Towards a Cultural and Historical Narratology: A Survey of Diachronic Approaches, Concepts and Research Projects". In Reitz, Bernhard; Rieuwerts, Sigrid, eds. (2000). 345-373.
---. 2001.
"On the Perspective-Structure of Narrative Texts: Steps toward a Constructivist Narratology". New Perspectives on Narrative Perspective. Ed. Willi van Peer and Seymour Chatman. Albany: State University of New York P. 207-24.
Nünning, Vera; Nünning, Ansgar, eds. 2002.
Erzähltheorie transgenerisch, intermedial, interdisziplinär. Trier: WVT.
---; Nünning, Ansgar, eds. 2004.
Erzähltextanalyse und Gender Studies. Stuttart: Metzler.
Ohmann, Richard. 1971.
"Speech Acts and the Definition of Literature." Philosophy and Rhetoric 4: 1-19.
Onega Susan; García Landa, José Angel, eds. 1996.
Narratology: an Introduction. London: Longman.
O'Neill, Patrick. 1994.
Fictions of Discourse: Reading Narrative Theory. Toronto: U of Toronto P.
Orth, Eva Maria. 2000.
"The Soliloquy in 16th and 17th Century English Fiction". In Reitz, Bernhard; Rieuwerts, Sigrid, eds. (2000). 439-449.
Page, Norman. 1973.
Speech in the English Novel. London: Longman.
Palmer, Alan. 2004.
Fictional Minds. Lincoln: U of Nebraska P.
2010. Social Minds in the Novel. Ohio: Ohio State UP.
Pascal, Roy. 1977.
The Dual Voice: Free indirect speech and its functioning in the 19th century European novel. Manchester: Manchester UP.
Pavel, Thomas G. 1985a.
The Poetics of Plot: The Case of English Renaissance Drama. Manchester: Manchester UP.
---. 1985b.
"Literary Narratives". In van Dijk, Teun A., ed. Discourse and Literature: New Approaches to the Analysis of Literary Genres. Amsterdam: Benjamins. 85-103.
Perry, Menakhem. 1979.
"Literary Dynamics: How the Order of a Text Creates its Meanings". Poetics Today 1-2: 35-64; 311-361.
Pfister, Manfred. 1977.
Das Drama: Theorie und Analyse. München: Fink.
---. 1988.
The Theory and Analysis of Drama. Trans. John Halliday. Cambridge: Cambridge UP.
Phelan, James. 1996.
Narrative as Rhetoric. Columbus: Ohio State UP.
---. 2001.
"Why Narrators Can Be Focalizers-and Why It Matters." Ed. Willi van Peer and Seymour Chatman. Albany: State University of New York P. 51-64.
Pier, John, ed. 1999.
Recent Trends in Narratological Research. GRAAT: Revue des Groupes de Recherches Anglo-Américaines de L'Université Francois Rabelais de Tours 21.
Plett, Heinrich F. 1988.
"The Poetics of Quotation." In: Petöfi, Janos S.; Olivi, Terry, eds., From Verbal Constitution to Symbolic Meaning. Hamburg: Buske. 313-334.
Posner, Richard A. 1997.
"Narrative and Narratology in Classroom and Courtroom". Philosophy and Literature 21.1: 292-305.
Pouillon, Jean. 1946.
Temps et roman. Paris: Gallimard.
Pratt, Mary Louise. 1977.
Toward a speech act theory of literary discourse. Bloomington and London: Indiana UP.
Prince, Gerald. 1973.
A Grammar of Stories. The Hague: Mouton.

---. 1978. "Le discours attributif et le récit". Poétique 35: 305-313.

---. 1980.
"Introduction to the study of the narratee". In Tompkins, Jane P, ed. Reader-response Criticism: From Formalism to Post-Structuralism. London: Johns Hopkins UP. 7-25.
---. 1982.
Narratology: The Form and Functioning of Narrative. Berlin: Mouton.
---. 1987.
Dictionary of Narratology. Lincoln: U of Nebraska P.
---. 2001.
"A Point of View on Point of View or Refocusing Focalization." Ed. Willi van Peer and Seymour Chatman. Albany: State University of New York P. 43-50.
---. 2003.
A Dictionary of Narratology. Lincoln: U of Nebraska P.
Propp, Vladimir. 1968.
Morphology of the Folk-Tale. Trans. Laurence Scott. Austin: U of Texas P.
Quirk, Randolph, et al. 1985.
A Comprehensive Grammar of the English Language. London: Longman.
Rabinowitz, Peter. 1987.
Before Reading: Narrative Conventions and the Politics of Interpretation. Ithaca: Cornell.
Reichenbach, Hans. 1966 [1947].
Elements of Symbolic Logic. New York: Free Press.
Reitz, Bernhard; Rieuwerts, Sigrid, eds. 2000.
Anglistentag 1999 Mainz: Proceedings. Trier: WVT.
Richardson, Brian. 2009.
"Plural Focalization, Singular Voices: Wandering Perspectives in 'We'-Narration". 143-162.
Ricoeur, Paul. 1983.
Narrative Fiction: Contemporary Poetics. London: Methuen.
---. 1988.
Time and Narrative. Trans. Kathleen McLaughlin & David Pellauer. Chicago: U of Chicago P.
---. 1991.
"Life in Quest of Narrative". In Wood, David, ed., On Paul Ricoeur: Narrative and Interpretation. London: Routledge. 20-33.
Riffaterre, Michael. 1996.
"Chronotopes in Diegesis". In Mihailescu and Harmaneh, eds. 244-256.
Rimmon-Kenan, Shlomith. 1983.
Narrative Fiction: Contemporary Poetics. London: Methuen.
Ron, Moshe. 1987.
"The Restricted Abyss: Nine problems in the Theory of Mise en Abyme". Poetics Today 8.2: 417-438.
Ronen, Ruth. 1986.
"Space in Fiction". Poetics Today 7: 421-483.
---. 1994.
Possible Worlds in Literary Theory. Cambridge: Cambridge UP.
Rumelhart, David E. 1975.
"Notes on a Schema for Stories". In Bobrow, Daniel G.; Collins, Allan, eds. Representation and Understanding: Studies in Cognitive Science. New York: Academic. 211-236.
Ryan, Marie-Laure. 1991.
Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington and Indianapolis: Indiana UP.
---. 1997a.
"Interactive Drama: Narrativity in a Highly Interactive Environment". Modern Fiction Studies 43.3: 677-707.
---. 1997b.
"Postmodernism and the Doctrine of Panfictionality". Narrative 5.2: 165-187.
---. 1998.
"The Text as World Versus the Text as Game: Possible Worlds Semantics and Postmodern Theory". Journal of Literary Semantics 27.3: 137-163.
---. 1999.
"Cyberage Narratology: Computers, Metaphor, and Narrative". In Herman, ed. (1999). 113-141.
---, ed.2004.
Narrative Across Media: The Languages of Storytelling. Lincoln: U of Nebraska P.
---. 2010.
"Narratology and Cognitive Science: A Problematic Relation". Style 44: 4, 469-95.
---. 2013.
Narrative as Virtual Reality 2: Revisiting Immersion and Interactivity in Literature and Electronic Media. Baltimore: Johns Hopkins UP.
Sacks, Harvey; Schegloff, Emanuel A.; Jefferson, G. 1974.
"A Simplest systematic for the Organization of Turn-Taking in Conversation". Language 50: 696-735.
Scanlan, David. 1988.
Reading Drama. Mayfield: Mountain View.
Scarry, Elaine. 1995.
"On Vivacity: The Difference between Daydreaming and Imagining-Under-Authorial-Instruction". Representations 52: 1-26.
Schank, Roger C. 1995 [1990].
Tell Me a Story: Narrative and Intelligence. Evanston: Northwestern UP.
Schlickers, Sabine. 2009.
"Focalization, Ocularization and Auricularization in Film and Literature". Ed. P. Hühn, et al., Point of View, Perspective, and Focalization. Modeling Mediation in Narrative (Narratologia 17). Berlin: de Gruyter. 243-58.
Schneider, Ralf. 2000.
Grundriß zur kognitiven Theorie der Figurenrezeption am Beispiel des viktorianischen Romans. Tübingen: Stauffenburg.
Schneider, Ralf; Hartner, Marcus, eds. 2012.
Blending and the Study of Narrative: Approaches and Applications. Narratologia 34. Hamburg: de Gruyter.
Schwarze, Hans-Wilhelm. 1989.
"Ereignisse, Zeit und Raum, Sprechsituationen in narrativen Texten". In Ludwig, ed.. (1989). 145-188.
Searle, John R. 1974 [1969].
Speech Acts in the Philosophy of Language. London: Cambridge UP.
Short, Michael. 1991.
"Speech presentation, the novel and the press". In van Peer, Willie, ed. The Taming of the Text: Explorations in Language, Literature and Culture. London: Routledge. 61-81.
---; Semino, Elena; Culpeper, Jonathan. 1996.
"Using a Corpus for Stylistic Research: Speech and Thought Presentation". In Thomas, Jenny; Short, Michael, eds. Using Corpora in Language Research. London: Longman.
Simpson, Paul. 1993.
Language, Ideology and Point of View. London: Routledge.
Sinclair, May. 1990 [1918].
"The Novels of Dorothy Richardson". In: Scott, Bonnie Kime, ed. The Gender of Modernism: A Critical Anthology. Bloomington: Indiana UP. 442-448.
Smuda, Manfred. 1981.
"Stream of Consciousness und durée -- das Problem ihrer Realisation und Wirkung im modernen englischen Roman". Poetica 13: 309-326.
Souvage, Jacques. 1965.
An Introduction to the Study of the Novel. Ghent: Story-Scientia.
Spence, Donald P. 1987.
"Narrative Recursion". In Rimmon-Kenan, Shlomith, ed. Discourse in Psychoanalysis. London: Methuen. 188-210.
Srull, Thomas K.; Wyer, Robert S. 1988.
A Dual Process Model of Impression Formation. London: Erlbaum.
Stanzel, Franz K. 1955.
Die typischen Erzählsituationen im Roman. Stuttgart: Braunmüller.
---. 1959.
"Episches Präteritum, erlebte Rede, historisches Präsens". Deutsche Vierteljahrsschrift 33: 1-12.
---. 1964.
Typische Formen des Romans. Göttingen: Vandenhoeck.
---. 1982.
Theorie des Erzählens. 2nd ed. Göttingen: Vandenhoeck.
---. 1984.
A Theory of Narrative, trans. Charlotte Goedsche. Cambridge: Cambridge UP.
Stein, Nancy L. 1982.
"What's a Story: Interpreting the Interpretation of Story Grammars". Discourse Processes 5.3-5, 319-35.
Steinberg, Erwin R. 1973.
The Stream of Consciousness and the Beyond in 'Ulysses'. Pittsburgh: Pittsburg UP.
Sternberg, Meir. 1993 [1978].
Expositional Modes and Temporal Ordering in Fiction. Bloomington: Indiana UP.
---. 1982a.
"Point of View and the Indirections of Direct Speech". Language and Style 15: 67-117.
---. 1982b.
"Proteus in Quotation-Land: Mimesis and the Forms of Reported Discourse." Poetics Today 3.2: 107-156.
---. 1985.
The Poetics of Biblical Narrative. Indiana: Indiana UP.
Stevenson, Randall. 2003.
Modernist Fiction: An Introduction. Hemel Hempstead: Prentice Hall.
Tamir, Nomi. 1976.
"Personal narrative and Its Linguistic Foundation". A Journal for Descriptive Poetics and Theory of Literature 1: 403-429.
Tammi, Pekka; Tommola, Hannu, eds. 2003.
Linguistic and Literary Aspects of Free Indirect Discourse from a Typological Perspective. Tampere: Tampereen Yliopistopaino Oy.
Todorov, Tzvetan. 1969.
Grammaire du Décameron. Mouton: The Hague.
---. 1981.
Introduction to Poetics. Brighton: Harvester.
Toolan, Michael J. 1988.
Narrative: A Criticial Linguistic Introduction. London: Routledge.
---. 1999.
"New Work on Deixis in Narrative". In Grünzweig, Walter; Solbach, Andreas, eds. Transcending Boundaries: Narratology in Context. Tübingen: Narr. 147-163.
---. 2001.
Narrative: A Criticial Linguistic Introduction. 2nd ed. London: Routledge.
Torgovnick, Marianna. 1981.
Closure in the novel. Princeton, NJ: Princeton UP.
Turner, Mark. 1996.
The Literary Mind. Oxford: Oxford UP.
Uspensky, Boris. 1973.
A Poetics of Composition. Berkeley: U of California P.
Vitoux, Pierre. 1982.
"Le jeu de la focalisation." Poétique 51: 359-68.
Wall, Kathleen. 1994.
"The Remains of the Day And its Challenges to Theories of Unreliable Narration". Journal of Narrative Technique 24.1: 18-42.
Wales, Katie. 1989.
A Dictionary of Stylistics. London: Longman.
Warhol, Robyn R. 1989.
Gendered Interventions: Narrative Discourse in the Victorian Novel. New Brunswick: Rutgers UP.
White, Hayden. 1996 [1981].
"The Value of Narrativity in the Representation of Reality". In Onega and Garcia Landa, eds. (1996). 273-285.
Wolf, Werner. 1993.
Ästhetische Illusion und Illusionsdurchbrechung in der Erzählkunst. Tübingen: Niemeyer.
---. 1999.
The Musicalization of Fiction: A Study in the Theory and History of Intermediality. Amsterdam: Rodopi.
Würzbach, Natascha. 2001.
"Erzählter Raum: Funktionaler Baustein, kultueller Sinnträger, Ausdruck der Geschlechterordnung". In Helbig, Jörg, ed. Erzählen und Erzähltheorie im 20. Jahrhundert. Heidelberg: Winter. 105-129.
Yacobi, Tamar. 2000.
"Interart Narrative: (Un)Reliability and Ekphrasis". Poetics Today 21.4: 711-749.