‘Gap’ is such a small word to measure the boundaries of a potential abyss and Stanzel turns to the terminology of Roman Ingarden and his theory of ‘points of indeterminacy’ (and picks out Das literarische Kunstwerk and Vom Erkennen des literarischen Kunstwerks for further reading – links to English translations). Stanzel, concerned with what the reader can fill in, makes a distinction between ‘spaces’ that can be filled with help from the text and ‘gaps’ in time and/or space which the reader must face alone. The Argonautica, my own model-text, is a ‘gappy’ narrative and what Stanzel has to say here is interesting but we need to be clear (or clearer) with our labelling of ‘indeterminacies’ and ‘gaps’, what it is we can ‘concretize’ and how we can go about it.
Ingarden’s theories and ‘concretizing’ also occur in another essay on the reader, in Wolfgang Iser’s ‘The Reading Process: A Phenomenological Approach’ (and just before the extracts which frame my About). The phenomenological approach, he begins, depends upon the text and the reader’s response ‘in equal measure’.
Thus Roman Ingarden confronts the structure of the literary text with the ways in which it can be konkretisiert (realized). The text as such offers different “schematized views” through which the subject matter of the work can come to light, but the actual bringing to light is an act of Konkretisation. If this is so, then the literary work has two poles, which we might call the artistic and the aesthetic: the artistic refers to the text created by the author, and the aesthetic to the realization accomplished by the reader.
(Iser 1972: 279)
‘Concretization’ or ‘Realization’, ‘indeterminate spaces in the discourse’ and ‘passages of story time elided from the narrative’: a language of construction and emptiness but is it in the emptiness of design or necessity or both that the reader has room to build? Here is another definition of ‘gaps’ taken from the glossary in David Herman’s Basic Elements of Narrative.
Even accounting for ‘gaps’ which can or can’t be filled in, there is more than one type of ‘gap’ and they can be found on different ‘levels’ (see The Tripartite Model). Herman makes a distinction between gaps found in the ‘told’ and the ‘telling’. The ‘told’ is the storyworld where there will always be gaps because a narrative is a selective presentation of events from that world (which doesn’t generally include sections of character biographies, world topography, political structures, languages, histories etc., and which would still be incomplete, however much world-building a text contains). A reader is free to imagine how many siblings Captain Kirk had or create Bruce Banner’s family tree but those are ‘gaps’ of ‘necessity’ – information holes due to that ‘radical incompleteness of fictional worlds’.
‘Gaps’ in the ‘telling’, whether indeterminacies or true gaps (following Stanzel’s demarcation by potential ‘fillability’) are the result of selective presentation. Stanzel is concerned with ‘concretizing’; his complementary story springs from ‘the reader’s desire to recreate in his mind an entity that is, in all its aspects, as determinate, cohesive, and continuous as possible’ (Stanzel 2004: 209). Iser considers indeterminacies crucial to the dynamism between text and reader which creates the virtual work: ‘whenever the flow is interrupted and we are led off in unexpected directions, the opportunity is given to us to bring into play our own faculty for establishing connections – for filling in the gaps left by the text itself’ (Iser 1972: 284-5).
Iser’s own observations are building on Ingarden’s ‘intentional sentence correlatives’: one sentence follows another and is connected by the individual reader. Gaps break the chain, but unlike Ingarden, Iser does not consider this a flaw: ‘one text is potentially capable of several different realizations, and no reading can ever exhaust the full potential, for each individual reader will fill in the gaps in his own way, thereby excluding the various other possibilities; as he reads, he will make his own decision as to how the gap is to be filled. In this very act the dynamics of reading are revealed’ (Iser 1972: 285).
We’re drifting in this follow-up post on Stanzel’s Komplementärgeschichte theory but I cannot talk about reader-dynamics and gaps without bringing in the dynamics of narrative interest and Meir Sternberg’s Expositional Modes and Temporal Orderings in Fiction.
That’s a fair bit of material on gaps to process. It’s also material on related but different kinds of gaps, on gaps operating at different diegetic levels. Time (my selected presentation) to bring in a sample text and look for them. Pelias, the tyrant who initiates the quest for the Golden Fleece, was mentioned once or twice when I last wrote about the Complementary Story and what I want to do here is return to Pelias and think about/apply ‘gap theories/definitions’ to a specific example from a text. Argonautica spoilers ahead.
Τοίην γὰρ Πελίης φάτιν ἔκλυεν, ὥς μιν ὀπίσσω
μοῖρα μένει στυγερή, τοῦδ᾽ ἀνέρος, ὅν τιν᾽ ἴδοιτο
δημόθεν οἰοπέδιλον, ὑπ᾽ ἐννεσίῃσι δαμῆναι.
Δηρὸν δ᾽ οὐ μετέπειτα τεὴν κατὰ βάξιν Ἰήσων,
χειμερίοιο ῥέεθρα κιὼν διὰ ποσσὶν Ἀναύρου,
ἄλλο μὲν ἐξεσάωσεν ὑπ᾽ ἰλύος, ἄλλο δ᾽ ἔνερθεν
κάλλιπεν αὖθι πέδιλον ἐνισχόμενον προχοῇσιν.
For such was the oracle that Pelias heard, how in the future a hateful fate waited for him – that he would be overcome through the suggestions of the man, the one from amongst the people whom he would spot one-sandalled. And not long afterwards, Jason, in accordance with your oracle, going on foot through the streams of wintry Anaurus, saved one sandal from the mud but left the other there held fast beneath its current.
‘Such was the oracle that Pelias heard.’ We are within the Argonautica’s proem. This is the epic’s extradiegetic narrator’s preliminary exposition: information for the reader setting up the narrative to come. Within this summary is also cause; Apollo’s oracle and Pelias’ attempt to avert it initiates the sequence of events in story-time (the quest for the Golden Fleece). In Measuring Arrows in Time, I discuss Summary, Scene, Ellipsis (and fuzzy borders) in terms of Duration and the presence of the same god Apollo in the Iliad’s proem, but here we are more concerned with another aspect of Time: Order and the selective distribution of information throughout a narrative.
In the Iliad Apollo once invoked by his priest Chryses became active in the narrative as an avenger firing arrows (Il. 1.44-49). This is different. Apollo is not here a character in the story-time but words, words of an oracle and words which are reported indirectly. Whatever the oracle as intermediary of the god said to Pelias is not put before the reader in quotation. Instead, what the reader finds in the text is a character’s interpretation of the oracle since the account that follows in verses five to seven is focalised through Pelias himself.
Some of this material is (much) abridged and modified from my doctoral thesis where I argued that the manner of the presentation from the outset of the poem conditions the reader for the selective distribution of information in episodes to come later in Book 1. All I wish to do here is to draw attention to indeterminacies and to examine this extract thinking about the questions just raised in the long quotation from Sternberg (and bring in Stanzel and Iser as appropriate). Sternberg raises a host of questions – about event-sequencing, about causation, about motivation. What interests Sternberg (in 1978) is narrative interest and the manipulation of the reader via the manipulation of expositional material in the presentation. He marks out two interactive and interacting paths of that interest: ‘suspense’ and ‘curiosity’.
Both dynamics are in operation in the Argonautica passage. The king, we are told, fears he will perish. Curiosity is produced by the indeterminacies of the indirect report. We do not know the wording or when the oracle was heard. Suspense is produced by the future orientation (as is the way with prophecy). He does not know when and neither do we, only that at some unknown future point he will perish (according to the reported oracle) through the suggestions of the one-sandalled man. Suggestions? What does that entail? Will the one-sandalled man kill Pelias himself or plan it somehow? If not the one-sandalled man then who? Or will Pelias succeed in thwarting destiny? Indirect reports, ambiguities, possibilities – from the outset (and we’re only five verses in) the reader is being prompted to consider a range of potentials, a range made available by the presentation.
Sometimes indeterminacies can be resolved quickly. For example, in my first paragraph on this passage I refer to the oracle as Apollo’s. That isn’t the wording of verse five. I designated the oracle as Apollo’s because of the explicit confirmation in verse eight when the narrator addresses the source ‘your oracle’. We came in at verse five aiming straight for Pelias but the poem’s opening three words Ἀρχόμενος σέο, Φοῖβε ‘Beginning with you, Phoebus’ includes prominently the addressee who here concretizes an indeterminate space.
We are back with Ingarden’s ‘intentional sentence correlatives’; as we read, new information modifies what we have read thus far and on reaching verse eight we can rule out other potential oracle sources (‘beginning with you – such was the oracle – your oracle’). Similarly, we need not puzzle long at all over the identity of the one-sandalled man. He too is revealed by name at the end of verse eight, followed by expositional information of where and how he lost his sandal (but some critical information is being withheld from that summary). Verse eight then resolves two brief indeterminacies; the source of the oracle and the identity of that oracle’s agent. Verse eight also shifts from Pelias’ focalisation of events to the extradiegetic narrator’s focalisation. He is addressing the god. We’ll focus on narrators (and incorporate Stanzel’s thoughts on narrator-types, notably figural narrative situations) in a separate project-section on narrators , but here let’s stick with the gaps and read a little further along.
ἵκετο δ᾽ ἐς Πελίην αὐτοσχεδὸν ἀντιβολήσων
εἰλαπίνης, ἣν πατρὶ Ποσειδάωνι καὶ ἄλλοις
ῥέζε θεοῖς, Ἥρης δὲ Πελασγίδος οὐκ ἀλέγιζεν.
αἶψα δὲ τόνγ᾽ ἐσιδὼν ἐφράσσατο, καί οἱ ἄεθλον
ἔντυε ναυτιλίης πολυκηδέος, ὄφρ᾽ ἐνὶ πόντῳ
ἠὲ καὶ ἀλλοδαποῖσι μετ᾽ ἀνδράσι νόστον ὀλέσσῃ.
He came at once to Pelias to take part in the solemn feast which he was performing for his father Poseidon and the other gods, but for Pelasgian Hera he paid no mind. As soon as Pelias saw him, he made a plan and implemented the trial of a voyage full of sorrows in order that either on the sea or else amongst foreign men he would destroy his homecoming.
This is still exposition and still summary so unsurprisingly concise. We have a reason for Jason to encounter Pelias. He was attending a feast. We get some genealogical information about Pelias. We learn that he is the son of the god Poseidon. We are told that he was honouring the gods (surely a good thing), all that is but Hera (possibly not a good thing). That the omission is noted by the narrator suggests it has (potential) significance (something to keep in mind for later), but there is no further explication. What was Pelias’ motivation for ignoring Pelasgian Hera? The exposition moves straight from pointing this out to his plans for Jason. No sooner than he sees him, he recognises him as the one-sandalled man (not in the exposition but an indeterminacy filled in by me based upon what he then devises for Jason) and comes up with the quest. Now that quest was outlined in verses one to five as the voyage into the Black Sea to fetch the Golden Fleece but what we have here is Pelias’ version.
This is a return to Pelias’ focalisation (following the verbs of perception in verse fifteen ‘as soon as he saw, he made plans’) and it includes evaluation and motivation, ‘a voyage full of sorrows so that … he would destroy his homecoming’. In order to circumvent ‘hateful fate’, Pelias comes up with a mission impossible for fate’s agent. We have motivation and cause, and we have the production of suspense and very little idea of how this will play out. ‘On the sea or else amongst foreign men’ suggests/sketches two potential settings and encounters. Already we might be wondering what scenarios could unfold at sea and what people might be met on the voyage. But what was Pelias’ motivation for ignoring Pelasgian Hera?
This presentation is uneven even as it relates information about the same character. The motivation has been provided for one action (the mission) but regarding the motivation for an action that precedes it in both the narrative and story sequence – what Pelias was doing and not doing when Jason arrived – the narrator is silent. This is a not a storyworld ‘how many siblings’ gap, this is a possible causation gap in the story-sequencing. There is no explicit link made in this preliminary exposition between Pelias’ refusal to honour Hera and his fate. A reader could infer one because why mention this at all if it is not or will not become relevant? But that is for the reader to keep in mind, to put on the virtual board whilst progressing through the text.
Spoilers! Relying solely on the text of the Argonautica, the reader must be very patient for any sort of explanation (and can make all manner of hypotheses in the meantime). It comes in Book 3, 2690 lines later, when Hera herself in direct speech explains to Athena and Aphrodite why she loves Jason and why Pelias will not escape his fate (A.R. 3.56-75). She tells them she would save Jason even from the Underworld…
ὄφρα μὴ ἐγγελάσῃ Πελίης κακὸν οἶτον ἀλύξας,
ὅς μ᾽ ὑπερηνορέῃ θυέων ἀγέραστον ἔθηκεν.
καὶ δ᾽ ἄλλως ἔτι καὶ πρὶν ἐμοὶ μέγα φίλατ᾽ Ἰήσων,
ἐξότ᾽ ἐπὶ προχοῇσιν ἅλις πλήθοντος Ἀναύρου
ἀνδρῶν εὐνομίης πειρωμένῃ ἀντεβόλησεν
θήρης ἐξανιών· νιφετῷ δ᾽ ἐπαλύνετο πάντα
 οὔρεα καὶ σκοπιαὶ περιμήκεες, οἱ δὲ κατ᾽ αὐτῶν
χείμαρροι καναχηδὰ κυλινδόμενοι φορέοντο.
γρηὶ δέ μ᾽ εἰσαμένην ὀλοφύρατο, καί μ᾽ ἀναείρας
αὐτὸς ἑοῖς ὤμοισι διὲκ προαλὲς φέρεν ὕδωρ.
‘… so that Pelias does not avoid an evil fate and mock me, the man who in his arrogance deprived me of the honour of sacrifice. And besides, even before that Jason was very dear to me, from the time when by the streams of the overflowing Anaurus he met me as I was testing the piety of men and he was returning from a hunt. All the mountains and high peaks were being sprinkled with falling snow, and from them torrents came down in rolling resounds. He took pity on me in my guise of an old woman and on his own shoulders carried me across the rushing water.’
Skipping thousands of lines and everything that the narrative presentation has done to condition us in the meantime (all the other indeterminacies potentially in play and all the hypotheses on the board), we can revise with new evidence some of those early indeterminacies. Hera’s version is more expansive than the narrator’s. She remembers and relates vividly the season and the scene. Hera was there at the Anaurus testing Jason. Hera is the cause of the one sandal. Behind the oracle is Hera, the cause of ‘hateful fate’.
Hera’s κακὸν οἶτον (3.64) recalls and recasts the μοῖρα στυγερή Pelias hoped to avert. This isn’t about Jason and Pelias; this, the reader hears in Book 3 in Hera’s own words, is about Hera and Pelias. And those ‘suggestions’? Who kills Pelias? The Argonautica ends on the shore of home and before any encounter, but, again in Book 3, the narrator does foreshadow it in a comment for the reader. Following the narrative of Jason and Medea’s encounter in Colchis and following Medea providing him with the drugs and instruction needed to complete his trials, the narrator intrudes to inform us ὧς γὰρ τόδε μήδετο Ἥρη, | ὄφρα κακὸν Πελίῃ ἱερὴν ἐς Ἰωλκὸν ἵκοιτο | Αἰαίη Μήδεια, (3.1134-6). ‘For so did Hera plot this, that Aeaean Medea would come to sacred Iolcus as an evil to Pelias’. Hera will destroy Pelias through her agent Medea.
Of course, we still do not know why Pelias omitted Hera alone when making his sacrifice! The goddess is explicit about her rage, about her reason and about her plans for him. She is explicit about Jason having a role in those plans. The narrator is explicit about who will cause his destruction. But why did Pelias intentionally dishonour her? That’s a gap the Argonautica does not fill but not everything needs explaining. We have the complementary story. We could invent a motivation. We could decide it is unnecessary, that he ignored her, was punished and that is enough. Or we could look elsewhere and find in the work of a mythographer of the first century AD, the following summary of how and where Pelias killed his stepmother Sidero.
ἡ δὲ φθάσασα εἰς τὸ τῆς Ἥρας τέμενος κατέφυγε, Πελίας δὲ ἐπ᾽ αὐτῶν τῶν βωμῶν αὐτὴν κατέσφαξε, καὶ καθόλου διετέλει τὴν Ἥραν ἀτιμάζων.
Apollod. Bibliotheca 1.9.8
She [Sidero] had already fled for refuge to a precinct of Hera, Pelias murdered her on the very steps, and generally persisted in dishonouring Hera.
The man is a habitual Hera-hater (at least according to this later account) though I suppose drafting in this text as supporting evidence strengthens her motivation against him rather than elucidate his against her. What the act of searching elsewhere also does is draw attention to the added complications of processing narratives that use a shared cast of characters, that draw upon the material of myth because awareness of parallel versions and competing narratives will affect in some way the ongoing construction of the complementary story (or Iser’s virtual work, or Sternberg’s dynamics of narrative interest).
The Argonautica is a work of the third century BC, a work at the beginning of what Nick Lowe describes as a ‘three-century literary counter-current stretching from Callimachus to Ovid’ and a period in which the reader ‘is credited by the text with an encyclopaedic knowledge of all myth; he can no longer be surprised, so that even if by chance the reader is classically manipulated by a story he does not know, the pretence has to be maintained that he knows it already’ (Lowe 2000: 98).
That is a knowledge the modern reader does not have but a modern reader familiar with other extant narratives drawing on the same mythological material still has plenty to draw upon and might e.g. (from the Medea of Euripides or the Medea in Ovid’s Metamorphoses or Heroides) already have Medea in mind when reading the proem. The reader of Pindar’s Pythian 4 has much supplementary and conflicting material to bring to the summary account of Jason’s meeting with Pelias here.
And if we don’t have them? Each individual reader creates their individual virtual work. We can make our hypotheses based on e.g. story-types rather than on story-versions. We can call upon familiarity with other quest narratives featuring evil tyrants/wizards and beleaguered protagonists to put reasonable speculations on the board where indeterminacies occur.
Or at the other pole, a reader can draw upon awareness of specific models e.g. Homer in the case of the Argonautica (see my preliminary quotations of Virginia Knight in Building an Intertextual Network with Lions for the Argonautica’s Homeric debt). Sternberg, who is not a Homeric scholar, spends a chapter analysing delayed and distributed exposition in the Odyssey (1978: 56-89). It’s not without some problems for a reader familiar with Greek Myth (and a more detailed post is forthcoming in which we’ll also look at the distance he puts between his gaps and Iser’s gaps). However, his observations on e.g. the multifunctionality of Athena’s character in Odyssey 1 are quite brilliant. Her exchange with Zeus in direct speech not only gives the reader valuable information about Odysseus but also breaks the ‘expositional stalemate and sets the action going’ (1978: 60).
On Ithaca in disguise, the same goddess repeats the trick with Odysseus’ son, Telemachus (Od. 1.213f): ‘The information given in his answer is of course indispensable to the reader, not to the omniscient goddess, whose motivated pretense of ignorance enables Homer for the second time to work backwards and forwards at the same time – propelling the actional dynamics of the fictive world together with the presentational dynamics of the reading process’ (1978: 61).
There’s an omniscient god in the Argonautic proem. The text starts with him, the narrator addresses him but he’s not (here) a character in its story. I expect he knows all about Hera’s plans for Pelias. He also sets the narrative going but indirectly via an oracle (and in doing so sets up Pelias for his fall) whereas the narrator speaks to him directly. It is tempting (for this reader) to imagine a game being played here. In the Odyssey, the reader witnesses Athena in conversation first with Zeus and then with Telemachus and gains much from those exchanges. In the Argonautica, a narrator (who knows) talks to Apollo (who knows) and the reader feeds in this instance on scraps and temporary gaps.
In other instances, the narrator can be remarkably forthcoming. If I were to write a conventional quest/mission narrative, then early on in that narrative would be an assembling of the gang. The Argonautica is no different in this respect as verses 23-227 contain the Catalogue of Heroes. However, I probably wouldn’t include such spoilers as vv. 78-81 (Canthus and Mopsus marked for death in Libya) and 140-1 (Idmon marked for death, confirmed by himself 442-4!).
Now objections can be made. The Argonautica is neither the Magnificent Seven nor the Dirty Dozen. Canthus is no Boromir. Okay, but I am not here attempting to read as a contemporary Alexandrian but as a 21st century Mancunian (on Homeric Reception I can recommend Barbara Graziosi’s Inventing Homer and on memory and literacy in Classical Antiquity, Jocelyn Penny Small’s Wax Tablets of the Mind). I make these pop-culture references by way of linking to a curious and complex element of Complementary Reader theory that Stanzel introduces towards the end of his essay: ‘patterns of representation’ leading into ‘selectional grids’.
‘Patterns of representation’ are conceptually straightforward to grasp. Novels have styles (and we can expand that unproblematically into an array of narrative media), and a critical reader who reads ‘against the grain’ can impose one ‘selectional grid’ onto another when constructing the complementary story: ‘thus, he could substitute the Ulysses grid for that of Portrait of a Lady when reading James’ (Stanzel 2004: 215). This struck me as an activity not too greatly removed from my frequent comparing/contrasting of the Argonautica with Homeric models, if only more complicated. Why consciously implement a selectional grid? To what end? What insights will that give into the reading-process? How will it help fill in the gaps in a reading? I liked the idea but the examples offered were disappointing e.g. in his comments on ‘the physical facts of life’.
Perhaps that is just an unfortunate example but my answer to the last question is ‘yes’. Maybe if I were to read one immediately after the other… This reads as an attempt to bypass somehow the narrative or that the reader hankers after full narration, for the totality of a character’s existence in the storyworld. How many siblings did Captain Kirk have? These aren’t the gaps we’re looking for.
To each narrative we bring our experience of other narratives. If we recognise a story-pattern then we bring our experience of other story-patterns. If we recognise the particular story found in this narrative somewhere else, then we bring that information to this telling. All of this contributes towards creating a complementary story that will be continually modified in the process of reading a narrative. A narrative (like the Argonautica) can have a profoundly intertextual relationship with another which can assist in the filling-in of indeterminacies and in the creation of hypotheses. A narrative (like the Argonautica) can also exploit that intertextual relationship and misdirect the reader’s efforts. Still, because I like the concept if not all the associated examples, I shall let Stanzel have the last words.
Herman, D. (2009) Basic Elements of Narrative, Oxford.
Iser, W. (1972) ‘The Reading Process: A Phenomenological Approach’, New Literary History 3: 279-99.
Lowe, N. J. (2000) The Classical Plot and the Invention of Western Narrative, Cambridge.
Stanzel, F. K.  (2004), ‘The “Complementary Story”: Outline of a Reader-Oriented Theory of the Novel’, Style 38: 203-220.
Sternberg, M. (1978) Expositional Modes and Temporal Orderings in Fiction, Baltimore.