story episode: a meta scenario or register?

I am beginning to wonder if labeling “story episode” as a sort of high level meta scenario may be mixing metaphors. In addressing the structure of texts which contain stories, direct speech, indirect speech, poetry and non-narrative prose, it seems reasonable that a story episode would have a set of language-cultural specific criteria which could be understood as a set of assumptions and expectations about how a story will be told. Going back to the participle clause opening a new story episode which provides a situational (time, location, setting) frame for the story where the frame isn’t tightly bound semantically to the following episode (see previous post on Mk 5:1-2), the participle clause could be understood as a slot filler in a prototypical story episode frame, even though it doesn’t fill a slot in the scenario that is contained within that frame.

I had suggested that we view this under a different metaphor, a package with a wrapper. The story episode is the whole package, the wrapper is a container for the story scenario, but not a part of the story itself. A story episode “wrapper” might better fit in M.A.K Halliday’s[1] notion of register explained here by Liu Zequan[1].

For Halliday, register is "the clustering of semantic features according to situation type," and "can be defined as a configuration of semantic resources that the member of a culture typically associates with a situation type" (Halliday, 1978:111). Seen this way, "the notion of register is at once very simple and very powerful" and "provides a means of investigating the linguistic foundations of everyday social interaction from an angle that is complementary to the ethnomethodological one" (ibid.:31, 62). The theory of register thus derived "attempts to uncover the general principles which govern" how "the language we speak or write varies according to the type of situation" (ibid.:32). For Halliday, the central problem in text linguistics lies in how "the 'register' concept can take account of the processes which link the features of the text" "to the abstract categories of the speech situation" (ibid.:62). He warns linguists against "posing the wrong question" of "what features of language are determined by register?" (ibid.:32) in the process of seeking such a link. He tells us that we should instead seek for the factors that determine the selection of language (ibid.).

A problem I have with Halliday’s “register”[2] is vagueness. It seems to cover a lot territory and for that reason is somewhat hard to determine what is what is not included. Added to that is the difficulties of jumping back and forth between different schools of textlinguistics. Perhaps I will have more to say on this later.



[1] Liu Zequan (刘泽权) National University of Singapore
Register Analysisas a Tool for Translation Quality Assessment This is a revised version of the paper under the same title which was presented at The International Conference on Discourse and Translation held at Sun Yet-san University, Guangzhou, China from 24-26 July 2002

[2] M.A.K. Halliday 1967. "Notes on transitivity and theme in English," Journal of Linguistics, No. 3, Part 1: 37-81, Part 2: 199-244. (London, 1967)


M.A.K. Halliday 1978. Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Edward Arnold. 


M.A.K. Halliday 1994. An Introduction to Functional Grammar. 2nd Edition. London: Edward Arnold.

scenarios - packages & wrappers

scenario internal structure - packages & wrappers

R.A. Hoyle in appendix F [1] once again provides evidence for his claim that participle clauses belong to the scenario of the main verb. I am exploring another way of looking at this. Narrative scenarios appear to have an internal structure which could explained using a “package” metaphor. In NT greek a narrative episode is often preceded by a participle which provides either time, place or circumstances in regard to the following finite verb. This participle might be thought of as a “wrapper” serving as part of a “package” that contains the scenario. This participle does not always bind semantically to the scenario of the following finite verb. In some cases it binds to the preceding scenario. Take for example Mk 5:1-2:

Mark 5:1 Καὶ ἦλθον εἰς τὸ πέραν τῆς θαλάσσης εἰς τὴν χώραν τῶν Γερασηνῶν. 2 καὶ ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου εὐθὺς ὑπήντησεν αὐτῷ ἐκ τῶν μνημείων ἄνθρωπος ἐν πνεύματι ἀκαθάρτῳ,

The participle clause ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου marks the termination of a travel by boat scenario. The fact that Jesus had just got out of the boat onto the land doesn’t play a major role in the following scenario. Jesus could have arrived by other means at this location without disturbing the following scenario. On the other hand, getting off the boat is a prototypical end “wrapper” for a travel by boat scenario. Mk 5:1-5:2a serves a double purpose. It terminates the preceding scenario and it provides time and location information for the following scenario. If we think of scenarios as packages, Mk 5:1-5:2a is the end wrapper for the travel by boat scenario. The wrapper for the travel by boat scenario is found in Mk 4:35b and 5:2a. The scenario wrapper begins with the decision to travel by boat Mk 4:35b διέλθωμεν εἰς τὸ πέραν and ends with the statement that Jesus disembarked Mk 5:2a καὶ ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου.

Mark 4:35 Καὶ λέγει αὐτοῖς ἐν ἐκείνῃ τῇ ἡμέρᾳ ὀψίας γενομένης· διέλθωμεν εἰς τὸ πέραν. 36 καὶ ἀφέντες τὸν ὄχλον παραλαμβάνουσιν αὐτὸν ὡς ἦν ἐν τῷ πλοίῳ, καὶ ἄλλα πλοῖα ἦν μετ᾿ αὐτοῦ.

Mark 5:1 Καὶ ἦλθον εἰς τὸ πέραν τῆς θαλάσσης εἰς τὴν χώραν τῶν Γερασηνῶν. 2 καὶ ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου εὐθὺς ὑπήντησεν αὐτῷ ἐκ τῶν μνημείων ἄνθρωπος ἐν πνεύματι ἀκαθάρτῳ,

The wrapper in the travel by boat scenario is semantically closely associated with the situation represented in the scenario. This isn’t always true. Mk 5:1-2a also provides a time and place for the scenario that follows. But it doesn’t matter much how Jesus arrived when he was confronted by the demoniac. It is not crucial to the story that he arrived by boat. If we define a generic story episode scenario as having a wrapper which may provide time, location and circumstances, then the scenario within the wrapper can be though of as a subordinate scenario contained by the story episode scenario. In other words, the healing of the demoniac is a scenario contained within a story episode wrapper which makes it a sort of package. The package wrapper may not always be semantically tightly bound with the scenario inside. The wrapper has a narrative discourse function. It links the episodes together by place, time and circumstance.

[1]appendix F “Evidence That Greek Participial Clauses Belong in the Main Verb’s Scenario” page 523ff, but see also pages 100, 254, 255 and qualifications of this claim on pages 122, 136. Richard A. Hoyle, Scenarios, discourse and translation. SIL 2008

boundaries: scenarios & stories pt. 2

In my last post I suggested that the arrival of Jesus and the disciples by boat was not a part of the exorcism scenario. An exorcism embedded in a complex narrative episode may not have any time or location information attached specifically to the exorcism. It may inherit the time and location from a larger narrative segment. To illustrate this read Mark 9:1-29. In verse 14 we read Καὶ ἐλθόντες πρὸς τοὺς μαθητὰς, but there is no indication where the disciples were located. The last location mentioned in the story is found in Mk 9:9a Καὶ καταβαινόντων αὐτῶν ἐκ τοῦ ὄρους and that location note is dependent on the previous time and location mentioned in Mk 9:2 Καὶ μετὰ ἡμέρας ἓξ παραλαμβάνει ὁ Ἰησοῦς τὸν Πέτρον καὶ τὸν Ἰάκωβον καὶ τὸν Ἰωάννην καὶ ἀναφέρει αὐτοὺς εἰς ὄρος ὑψηλὸν κατ᾿ ἰδίαν μόνους. So there is a chain of dependency concerning time and location which links Mk 9:2,9,14.

My current thinking, which could change momentarily, is that including attributes (slots) from a generic "story episode" scenario within each species of story episode, e.g. travel by boat, storm at sea, exorcism, creates a lot of unnecessary redundancy in the the network of semantic frames/scenarios. Attributes (slots) which are universal to story episodes should be recorded at a higher level of the semantic hierarchy and inherited rather than duplicated. The attributes of an exorcism scenario should function such that a mere mention of the attribute will activate the scenario, making all the attributes of exorcism "accessible" as "hearer old" information.

To illustrate, a "reading scripture" scenario in the NT is prototypically associated with the Synagogue. For this reason, a mere mention of Synagogue as a location will make the "reading scripture" scenario available, in other words "reading scripture" becomes "hearer old" by activating Synagogue. Fishing in the gospels is generally associated with boats and the Sea of Galilee (a.k.a. Lake of Gennesaret). The mere mention of a boat in the gospels makes both the Sea of Galilee and fishing active, "hearer old" even if no fishing takes place in the pericope.

By contrast an exorcism scenario in the gospels does not associate prototypically with a type of location. Arriving by boat in Mk 5:1-2 does not make the exorcism scenario active or "hearer old". In like manner coming down of the mountain in Mk 9:9 does not activate the exorcism scenario. For this reason the location slot in the story episode frame probably belongs somewhere else, not in the definition of the exorcism scenario.

boundaries: scenarios and stories

A scenario[1] is prototypical semantic situation, e.g., taking a meal, going fishing, travel by boat, a storm at sea. A script[1] is a prototypical sequence of events that are identified with a particular scenario. The boundaries for a story episode are often not identical with the boundaries of a scenario. To illustrate, we will look at the transition between two episodes; the "Storm on the Lake" (Mk 4:35-41) and the "Healing of the Gerasene Demoniac" (Mk 5:1-20).

An episode in a story prototypically begins with a statement about time, location, setting.

Mark 4:35 Καὶ λέγει αὐτοῖς ἐν ἐκείνῃ τῇ ἡμέρᾳ ὀψίας γενομένης· διέλθωμεν εἰς τὸ πέραν. 36 καὶ ἀφέντες τὸν ὄχλον παραλαμβάνουσιν αὐτὸν ὡς ἦν ἐν τῷ πλοίῳ, καὶ ἄλλα πλοῖα ἦν μετ᾿ αὐτοῦ.

What follows is a travel by boat scenario with an embedded storm at sea scenario. A travel scenario is bounded by a start location and a destination. In this travel scenario the destination is reached in Mark 5:1.

Mark 5:1 Καὶ ἦλθον εἰς τὸ πέραν τῆς θαλάσσης εἰς τὴν χώραν τῶν Γερασηνῶν. 2 καὶ ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου εὐθὺς ὑπήντησεν αὐτῷ ἐκ τῶν μνημείων ἄνθρωπος ἐν πνεύματι ἀκαθάρτῳ,

In regard to the travel boat scenario Mark 5:2a καὶ ἐξελθόντος αὐτοῦ ἐκ τοῦ πλοίου is the end point. Someone might argue that 5:1 is the end. But that doesn't matter. The next episode, Healing of the Gerasene Demoniac. begins at 5:1-5:2a with location and temporal information. The healing scenario begins with Mk 5:2b.

The semantic situation (scenario) "healing a demoniac" does not include arrival by boat as the first item in the script. Arrival by boat, belongs to a script for a travel scenario. Arrival by boat is the end of one episode and beginning of another. It is semantically related to a travel scenario and not a healing scenario.

One might argue that a more abstract "narrative episode" scenario could be used with a script the includes arrival at some point and time where the episode begins. This would make the story episode boundaries identical with the scenario boundaries by definition. I don't think that would be a very useful procedure. For the purpose of semantic analysis and the identification of inferential associations we would still need to identify a travel by boat scenario, a storm scenario, a healing of demoniac (exorcism) scenario.


[1] I am reading a paper by R.A. Hoyle on Scenarios, Discourse and Translation which is available from SIL as a PDF . In the following posts on this subject I will assume some familiarity with the framework used in this paper.

Hoyle, Richard A. 2008. Scenarios, discourse, and translation: the scenario theory of Cognitive Linguistics, its relevance for analysing New Testament Greek and modern Parkari texts, and its implications for translation theory. SIL e-Books, 10. [Dallas]: SIL International. xi, 835 p. PDF

2 Cor 3:14b-c information structure

Today I have been looking at exegtical problems in 2 Cor 3:14, here is the context:

2Cor. 3:12 Ἔχοντες οὖν τοιαύτην ἐλπίδα πολλῇ παρρησίᾳ χρώμεθα 13 καὶ οὐ καθάπερ Μωϋσῆς ἐτίθει κάλυμμα ἐπὶ τὸ πρόσωπον αὐτοῦ πρὸς τὸ μὴ ἀτενίσαι τοὺς υἱοὺς Ἰσραὴλ εἰς τὸ τέλος τοῦ καταργουμένου. 14 ἀλλὰ ἐπωρώθη τὰ νοήματα αὐτῶν. ἄχρι γὰρ τῆς σήμερον ἡμέρας τὸ αὐτὸ κάλυμμα ἐπὶ τῇ ἀναγνώσει τῆς παλαιᾶς διαθήκης μένει, μὴ ἀνακαλυπτόμενον ὅτι ἐν Χριστῷ καταργεῖται· 15 ἀλλ᾿ ἕως σήμερον ἡνίκα ἂν ἀναγινώσκηται Μωϋσῆς, κάλυμμα ἐπὶ τὴν καρδίαν αὐτῶν κεῖται· 16 ἡνίκα δὲ ἐὰν ἐπιστρέψῃ πρὸς κύριον, περιαιρεῖται τὸ κάλυμμα.

While contemplating Paul's argument where he uses the Moses' veil as a metaphor for the inability of those under the old covenant to see the glory of Paul's gospel, after working through all the traditional exegetical options, I decided rather arbitrarily to attempt an analysis of the information structure in verse 14b-c, not because it was a passage particularly suited for this kind of analysis but just to see if I could make any sense out it using a nontraditional framework. The text segment for analysis is:

2 Cor 3:14b-c ἄχρι γὰρ τῆς σήμερον ἡμέρας τὸ αὐτὸ κάλυμμα ἐπὶ τῇ ἀναγνώσει τῆς παλαιᾶς διαθήκης μένει, μὴ ἀνακαλυπτόμενον ὅτι ἐν Χριστῷ καταργεῖται·

The first finite verb is μένει. I would suggest this is the focus, the most salient constituent. In other words the verb μένει is what this clause/sentence asserts. The initial constituent ἄχρι γὰρ τῆς σήμερον ἡμέρας is a temporal orienter and a contextualizer. The noun phrase τὸ αὐτὸ κάλυμμα is the topic which also serves as a contextualizer tying this clause to the veil κάλυμμα mentioned in v.13. My notion of a contextualizer is intentionally inclusive. Context is both textual and situational which includes temporal, local, cultural aspects. The separation of co-text from context is intentionally avoided.

The prepositional phrase ἐπὶ τῇ ἀναγνώσει τῆς παλαιᾶς διαθήκης is a setting identifier (another contextualizer) for the core clause τὸ αὐτὸ κάλυμμα ... μένει. I would suggest that the position of this phrase makes it less salient than ἄχρι γὰρ τῆς σήμερον ἡμέρας which is more inherently newsworthy and for that reason fronted.

An alternative approach might assign focus to the core clause τὸ αὐτὸ κάλυμμα ... μένει rather than the finite verb alone. What is salient is the relationship between the presupposition (the veil and the old covenant) and the current assertion that the veil remains even "today" when the old covenant is read. For that reason some might prefer to assign focus to the core clause.


The constituents following μένει raise a number of traditional exegetical problems. Here is what M.J. Harris (2Cor NIGTC 2005, p. 303) has to say about it.



more on this later ...