The “Complementary Story”

August Macke 023

The title of Stanzel’s essay certainly caught my eye. Anything related to reader-oriented creation/construction currently has that effect. The 2004 article is a translated and abbreviated version of a much earlier piece by Stanzel (1977) so I’m stepping forwards by going backwards with this. Still, I found it fairly rich in terms of (stimulating) ideas and suggesting problems to be carried forward to the exploration of more recent theorists.

In keeping with the general premise of this blog, what follows documents my own reader responses with additional citations limited to confirming or eliminating what stirred in the memory in the reading process. Fortunately, this first test-case proved an accessible read. Stanzel’s article is not dense with terminology and whilst the narratological basics are assumed (not without some drawbacks), aspects that are key to his theory are given sufficient explication (I think!).

So how does the reader get involved in the creative process as they read? In his second paragraph, Stanzel provides a prompt answer and it’s the one which provides the foundation for the complementary-story theory which follows: ‘a reader can easily become a narrator or coauthor of any garden-variety of narrative text as well, for the simple reason that the author and the narrator created by that author usually produce a story that is incomplete and full of gaps’ (203).

It’s the gaps! The early sighting of ‘gaps’ and the prominence given to them in Stanzel’s outline did cause some excitement but we should take onboard some ‘basics’ of the narrative theory on which to lay Stanzel’s observations.

The first is that a text (our document) contains a finite number of words. The second is that these words form the narrative. The third is the understanding that the narrative evokes a ‘storyworld’ of (potentially) infinite possibilities. The reader’s (restricted) access to this world is via the window(s) provided by the narrator(s). Due to the finite nature of a text ‘large sections of the story necessarily remain indeterminate in the narrative’ (203). Essentially, we get a narrative but not all possible narratives from all possible angles. Stanzel’s early foundation-building offers comforting rather than challenging reading but it is with recognition of the indeterminacies of narrative that he introduces the complementary-story (Komplementärgeschichte) theory.

It’s an uncomplicated premise (at least initially) – in the process of reading, the reader supplements and fills in textual gaps. However, by way of demonstration/confirmation, Stanzel leaps to the largest ‘gap’ of all (and one which immediately prompted a few questions!) – the ending of a narrative.

Is it really a gap when there’s no far side? However large, shouldn’t a gap have a boundary?

Standing on this precipice, Stanzel remarks on a tendency in modern fiction towards rolling ‘open’ endings that encourage a reader ‘to imaginatively pursue the characters’ life story beyond the ending of the text’ (204). It would seem then that this ‘feature’ is dependent (to some degree) upon the strength (or lack thereof) of closure. If, for example, all loose ends are neatly tied up, perhaps with a ‘they lived happily ever after’ then as readers we might not be inclined to imagine further into the timeline. Some other dependencies that spring to mind here are e.g. how successful is the narrative in engaging the reader’s interest in the characters and in the storyworld they inhabit (which will vary from reader to reader)? Do we care enough to continue a speculative narrative? Or what ‘seeds’ are planted earlier on in the narrative for continuing the narrative journey? Is the reader prompted strongly in a particular direction to take their own sequel?

My own (ultimate) goal in this foraging into theories is a practical one. What can I find and use in relation to my own case-study Apollonius’ Argonautica? Stanzel’s focus is on the novel and not on Hellenistic epic poetry. The theory is not going to neatly map but I’m looking for general correspondences, for new angles of approach. We could label the Argonautica, for instance, as a quest narrative – a ‘there and back again’ adventure. Yet, its ending is both open and closed.

The Argonauts (nearly all of them) do make it home again (closed) but how satisfied is the reader with a narrative that stops short of the protagonist Jason’s dealing with Pelias, the tyrant who sent him on a quest to die and the ‘villain’ of the piece (open?)? Then there is (for a reader who recognises and privileges certain generic models) a structural sense of closure – without getting overly technical here, the ‘hymnic’ quality of the poem’s opening is echoed in the closing lines with a ‘hymnic’ address from the narrator to his characters. So, there’s both loose ends (in terms of potential story-lines) and a structural neatness to the whole. This is basic but simply to show some complications already arising in applying the observation to a specific text.

A further complication in applying a complementary-reader theory of endings to the Argonautica is that it’s also a version of an existing myth. If variations of a story are already in circulation, the reader’s creative field is likely operating under restrictions (assuming familiarity with the myth!). Additionally, a sticky end for Pelias has been very strongly hinted at earlier in the poem by its narrator. Thus, whilst the reader might feel deprived of a showdown, there are guidelines in place for where to channel future projections (at least in broad strokes). In any case, Stanzel’s leap towards endings is brief and the discussion quickly turns to opportunities for potential endings not beyond the text’s outer edge but at points within the novel itself.

These include obvious spots for the reader to pause (e.g. reaching the end of a chapter) as well as allowing for the randomness of life outside the book. For example, I could be at a particularly juicy bit but then the phone rings or maybe I realise I’m late for an appointment or perhaps I really must pull myself away and get on with that assignment now and so on. What Stanzel suggests is that at any such point, the reader will continue to create though how long this period of creation persists is not evaluated/tested (‘for a while’, 204). Now this did remind me of other recent reading I decided was worth checking and quoting on ‘the illusory experience of living the story in time’ (Lowe 2000: 23).

In reality, of course, we are not ‘in’ the story at all: we are sitting in a living-room chair or a darkened theatre, our attention absorbed in a mental universe conjured out of a stream of processed signs. The illusion of temporal succession within that universe is purely an analogue… What we really have is a mental marker of the point in the story that the main sequential narrative has reached, at the point in the text to which we have so far read.

This is only part of a cognitive model Lowe develops but what triggered the recollection was the image of the absorbed reader, a reader ‘in the story’ and engaged in world-building. So what does happen when the book is closed or eyes lifted temporarily (and this is accepting that we are engrossed when the reality of the process might likely involve minds wandering in random and entirely unrelated directions, no matter how much we’re enjoying the reading/viewing experience)? Stanzel appears to be suggesting that the building work goes on. What happened to those gaps?

Well, following Stanzel’s theory, for an unspecified time the characters persist in our imagination in following their narrative trajectory. ‘In other words, the reader comes up with a story which complements what he has so far read’ (204). Allowances are made for variation between individual readers but when we do return to our reading ‘those sections of a narrative that follow such an interruption and are not immediately specified by the text—or are left completely vacant—put the reader in a position to fill in the gaps in the story or to continue it as a kind of experiment based on his own experiences and insights’ (204).

I started to have concerns here. My own sketch of what is being set out here is ‘reading for x amount of time (no definitions or attempts to measure x time are made) – stopping and gradually emerging from my storyworld immersion (but during this emersion time still spinning out possible narrative developments with the cast I’ve met) – getting back to reading after x amount of time again, checking my complementary story against the narrative there in the text and carrying on.’ Repeat process until run out of text.

I don’t think many readers would quibble with the idea that we play out possible scenarios. If we’re ‘hooked’ on a book, if we’re engaged with its characters and their world, if we’re intrigued by how the narrative(s) are developing then the notion that we anticipate and contribute strikes me as a commonplace amongst readers. It’s also, I think, something we do whilst reading (or viewing). We don’t necessarily need to take a breather to form sketches of potential events. I suppose what I’d question is how well-developed those sketches are (whether taking time out or not).

Does a reader whilst going about their daily routine mentally formulate the next section or chapter of the narrative to be checked against the narrative they find when returning to the text? Then, if the narrative goes off in a contradictory direction do we consign our efforts to the bin? If the narrative continues in a direction harmonious with our conjectures, do we store them as complementary reading? These questions conjure to mind a reader constantly creating speculative ‘fanfic’ ideas without committing them to the page (because then the reader would become a writer?).

Short digression and another classical example culled from recent reading. In Ovid’s Fasti (a long poem in elegiac couplets on the Roman Calendar) we find a miniature Aeneid in which the festival of Anna Perenna on the Ides of March is linked to Vergil’s epic by assimilating the character of Anna Perenna with Anna, sister of Dido (Fasti 3.545-656 for Ovid’s account).

In a brief discussion of the passage, Stephen Hinds has noted how the narrative is both a sequel to the Aeneid (continuing the Carthaginian Anna’s narrative) and a re-enactment (her Ovidian sequel replays/echoes events in the narrative of Aeneas, the epic’s protagonist (1992: 108)). I suppose that I’m envisioning this (and like) examples at the outermost edge of complementary-reader theory where another writer takes up the invitation provided by a gap and continues the narrative.

However, that’s also complicated by being an entirely different work with different narrative strategies and goals (and in a different genre). It’s a clever adaptation and one that does make use of a narrative gap (what became of Anna?) but was that creativity sparked by the gap encountered by Ovid the reader of the Aeneid in the act of reading or remembered and revisited by Ovid the writer of the Fasti in the act of writing? Now, I don’t have Ovid’s notes but I would veer to the latter. I suppose the point here of the example is just to complicate the theory in practice. Gaps do have the potential to encourage reader-creativity but not necessarily in a straightforwardly linear manner.

Making Alternative Stories

Addressing the objection that reader-fantasies run contrary to ‘the demarcations set out clearly in the text’ (206), Stanzel, using an example from David Copperfield, draws attention to the embedding of alternative stories in the reflections of characters:

‘I endeavoured to convert what might have been between myself and Agnes into a means of making me more self-denying, more resolved, more conscious of myself and my defects and errors. Thus, through the reflection that it might have been [my italics], I arrived at the conviction that it could never be.’

Such alternatives posed by a novel’s character-narrators ‘can prompt the reader to make similar reflections of his own, setting out in his mind a few possible variations in the course of the character’s life at appropriate points in the plot, if only in a sketchy and vague fashion [my italics].’

Of course, this technique can be used by any form of narrator (whether character within the story-world or a narrator external to it) and, in one iteration at least, the technique is as old as Homer: the counterfactual (or in Homer, the counter-assumedtraditional?).

E.g. Iliad 7.104-8 (counterfactual, with an added-bonus example of the Homeric narrator addressing a character in the story-world).

ἔνθά κέ τοι Μενέλαε φάνη βιότοιο τελευτὴ
Ἕκτορος ἐν παλάμῃσιν, ἐπεὶ πολὺ φέρτερος ἦεν,
εἰ μὴ ἀναΐξαντες ἕλον βασιλῆες Ἀχαιῶν,
αὐτός τ᾽ Ἀτρεΐδης εὐρὺ κρείων Ἀγαμέμνων
δεξιτερῆς ἕλε χειρὸς ἔπος τ᾽ ἔφατ᾽ ἔκ τ᾽ ὀνόμαζεν

‘And then Menelaos, there would have been revealed the end of your life at Hektor’s hands, since he was much the stronger man, if the kings of the Achaians had not leapt up and held you, and Agamemnon himself, the wide-ruling son of Atreus, not taken you by the right hand and spoke to you, saying…’ (trans. Hammond)

What if Agamemnon had not intervened? What if his brother (Helen’s husband) had fought Hector and died? Would the Greeks have stayed at Troy and fought on to win her back for his corpse? Troy might still stand and Achilles might have returned home after all…

In his monograph Homeric Misdirection (a topic for a future post!), James Morrison formulates these counterfactuals as “A if not B” and considers them just one method of ‘exploring ways in which the traditional story line could be upset’ (1992: 60). Alternatively, Irene de Jong considers that the Homeric counterfactual instead reinforces the reliability of the Homeric narrator who presents himself as relating ‘facts’ (and thus assurances through negation of what did not happen). In her view, he is not an ‘extreme’ self-conscious narrator openly debating story directions. There’s a brief discussion of this (with dozens of examples cited) in her narratological analysis of the Iliad. There, she brings up a potential (and for me, exciting) correspondence with Genette’s ‘l’arbitraire du récit’ (which she summarises – ‘a narrator at every point of his story has a limitless number of way to continue his story’) but ultimately decides against it (de Jong 2004: 78–81, cf. Genette 1969: 71-99).

I suppose what did disappoint me in Stanzel’s piece (even in comparison with de Jong’s short discussion) is that when I was hoping for gateways to parallel worlds to open for exploration, there was just a cursory glance to authorial alternative stories before, in the breadth of a paragraph, Stanzel moved off to a Dublin sketched by Joyce (but not before I bookmarked for later reading Hilary Dannenberg’s 2008 monograph Coincidence and Counterfactuality: Plotting Time and Space in Narrative Fiction).

How the reader gives colours to the world.

I’ll quote a chunk here which I thought both made a good point and caused some further concerns.

It is the sketchiness and elusive quality of these parenthetical figures, as well as the protagonist’s increasing isolation from his environment, that arouse the reader’s imagination. One wants to flesh out the sketches, to stop the flight of those who are drifting past and let in colorful, cordial, and unruly Dublin to break through the cordon sanitaire of our hero’s aloofness. Today we have access to the earlier version of this novel, published as Stephen Hero, and so we know that this restrictedness of the world described in Portrait constitutes a kind of conscious artistic asceticism. Just what temptations of turn-of-the-century Dublin Joyce had to resist we can conclude from the discarded text of Stephen Hero and, even more convincingly, from Dubliners, in which more than a dozen denizens of the town, whom in Portrait of the Artist the reader senses are hovering in the background of Dublin’s squares, streets, and pubs, come into their own fictional existence.

I can get behind the reader’s impulse to colour in sketches and Stanzel’s prose gives an evocative rendering of the reader’s collaborative urge though again it prompts questions. How engaged are we with those sketches we encounter? Do we notice the extras in the background? How do we decide who to flesh out and who to leave skeletal? Is the narrative uniform in its treatment of character details? If so, do we flesh everyone out? Or no-one out? Is the level of detail an indicator of who is and isn’t a ‘player’? Again, I think it’s a commonsense observation but one that invites consideration of all manner of variables e.g. responses will vary from reader to reader and the differing presentation of the narrative will condition those response e.g. is the prose style uniformly compact or is it prone to detailed description? And so on.

The concern for me came with the intertextual and paratextual ‘proof’. I made a joke in my earlier digression about not having Ovid’s notes on the Fasti but Stanzel is pulling out more Joyce here to confirm the theory and making a speculative suggestion (as I’m reading it) that the Dubliners is Joyce’s complementary-reader story realised. On the one hand, an earlier text (Stephen Hero) is used to argue for an author’s artistic intentions in the ‘middle-text’ (Portrait of an Artist) and on the other, the ‘end-text’ (Dubliners) is offered as evidence of an extreme reader-response (not simply imagined potentials but realised in text). This seems a muddled argument to me in which Joyce responds to his own invitation to colour in what he is speculated to have left out in the first place – essentially, an extrapolation based on an assumption to be then used as a paradigm for any reader’s act of complementary-story creation.

How do we decide which characters to flesh out? Is this complementary-story creation or filling out the story-world by e.g. speculating on backstory (providing biographies for incidental characters – as though the narrative was prompting the reader to the creation of an RPG campaign).

Harry snapped out of his daydream when the glass clunked on the bar. A little of the foamy head spilled over the rim. Harry reached into his pocket and produced a crumpled note. The barman nodded and took the money, leaving Harry to stare at his pint. (Example A)

Let’s imagine the above is an extract from a larger narrative and not something I’ve just made up. The barman’s ‘function’ in the narrative is to provide Harry’s beer. Let’s imagine that the narrative continues, that the narrator’s focus remains on Harry who drinks his beer and leaves the bar. Do we continue to speculate about the barman who served that pint, about his life, even about the rest of his shift? How far does the narrative need to progress away from the scene before he’s forgotten?

In the process of reading we assign roles – principal characters, support characters, background characters. Let’s expand the narrative.

Harry took a sip. It was cold, crisp with hops, and welcoming after the day he’d had. His hand reached into his pocket to feel the folds of the letter he’d received that morning. Still there. Part of him hoped it wasn’t, that it and its contents could somehow be spirited away. ‘Cheers, Alfie,’ he heard the other punter say to the barman as he pulled the letter free, opened it and began to read it again. (Example B)

As this narrative unfolds, the barman has acquired a name. Does this elevate him above ‘background’ status? Are we more interested in him now that he, like Harry, has a name? What about his appearance? If left at ‘barman’, then every reader is invited to create their own figure based solely upon the template ‘barman’. If additional details are provided by what here appears to be an extradiegetic narrator (observing from outside the storyworld and with access to Harry’s thoughts) e.g. that the barman is burly, bald, bearded etc., then some parameters are being set to restrict/focus the remodelling. Perhaps the barman has a distinctive tattoo. If my narrator goes on to make a detailed description of that tattoo, does the reader gloss over it or become absorbed in an accurate mental rendering of it? Does that rendering process encourage further interest in the character and additional speculation/creation?

I supposed where I’m wandering with this is the link between detail (or lack of), creativity and narrative-relevance. Does the provision of detail in itself signpost a potential relevance? For example, suppose a paragraph of text has been spent describing a tattoo (to which we can add effort spent on image-construction). Does that indicate that this information should be stored as having some potential future significance to the developing narrative? Are there exterior (to the narrative-extract) factors affecting the reader’s decision-making? E.g. are there generic markers that this story is a mystery? The reader of detective fiction on the look-out for clues, collecting details as puzzle pieces, might be more inclined to reconstruct and retain information on details in their mental notebook.

The three individuals (Harry, Alfie, unnamed punter) could become embroiled in conversation at the bar – anecdotes could be exchanged, protagonist and reader could learn more about Alfie. Alternatively, Harry could read the note (whose contents may or may not yet be known to the reader – and, if not, we might be more interested in their content and how Harry has been affected than in speculating about Alfie) and exit the bar.

The point of this meandering at this juncture is two-fold.

  1. Character-interest and reader-creation is discourse-motivated. It is not the only factor but an important one – the presentation of the story in a linear reading conditions both the variety of possible reader-responses and their depth (e.g. how far a reader might be inclined to take their musings on the barman).
  2. At any given juncture, creativity options present themselves to narrator and to reader-as-narrator. There is nothing at all to stop a reader from creating a complementary narrative about the barman but in situation A it would be most peculiar (to my mind), in B not much less so and only in suggestions of potential narratives in which the characters engage in more complex interactions (e.g. with dialogue) that it becomes a more likely proposition.

Of course, roles can change. We could take A + B as the beginning of a narrative, switch focus to Alfie observing Harry, watch him watch Harry leave, stay with Alfie and never pick up Harry’s narrative again and each fill in our own complementary-reader story on the contents of his letter…

Next Post – Back to the Gaps!


Genette, G. (1969) Figures II, Paris.
Hinds, S. (1992) ‘Arma in Ovid’s Fasti Part 1: Genre and Mannerism’, Arethusa 25: 81-112.
de Jong, I. J. F. [1987] (2004) Narrators and Focalizers: The Presentation of the Story in the Iliad, Amsterdam.
Lowe, N. J. (2000) The Classical Plot and the Invention of Western Narrative, Cambridge.
Morrison, J. V. (1992) Homeric Misdirection: False Predictions in the Iliad, Michigan.
Stanzel, F. K. (2004) ‘The “Complementary Story”: Outline of a Reader-Oriented Theory of the Novel’, Style 38: 203-220.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s