Printer Friendly

Peak-marking strategies in Old English narrative prose.

"peak is a zone of turbulence in the otherwise placid flow of discourse"

--Longacre, "Spectrum" 351

1. Introduction

Narrative is today probably the most thoroughly studied among the different discourse types and narratives still inspire students of language, literature, life history, folklore, and psychology, among others. Research, primarily in text and discourse linguistics but also in other fields, has shown that narratives in all languages are highly structured as to their beginnings and endings, participant tracking, episode marking, storylines, and grounding distinctions. What is particularly interesting is that narrative structuring has been shown to be signalled on the surface of discourse by various markers. Quite often these markers are features whose use remains obscure until they are examined from the perspective of the whole text or discourse. In 1976, Longacre used the terra mystery particles for these markers; nowadays, they are referred to as discourse markers, pragmatic particles, pragmatic markers, discourse particles and a myriad of other terms in text and discourse linguistics and pragmatics (cf. e.g., Aijmer and Simon-Vandenbergen; Foolen: Lenk).

Though both narrative structure and Old English are well-researched fields of inquiry, one feature of the organization of narrative text in Old English has attracted only little attention: the signalling of the peak of the story. Old English is a particularly interesting object for a study of this structural property of narratives, because Old English narrators could make use of a discourse marker, pa 'then', to signal the foregrounded main line of the narrative text and to mark the structural organization of the narrative in cooperation with other text-structuring signals (Enkvist, "More", "Problems"; Enkvist and Warvik; Foster; Hopper; Warvik, "Connective"). The purpose of this paper is to show how, in addition to its other functions, this discourse marker could be employed to highlight the peak in the progress of the narrative, through variation and interaction with other structuring signals. By comparing the ways of marking the peak in a sample of Old English narratives representing different prose genres, this study identifies preferences and general trends in the choices among alternative strategies.

2. Peak-marking strategies

The existence of a high-point, climax, turning-point, or peak is referred to in studies of different aspects of text structure, where this feature is used to elucidate and explain various phenomena, usually in narrative discourse. A general framework for describing properties of the peak is outlined by Longacre, who discusses peak-marking strategies in a variety of languages. He "use[s] the term PEAK to refer to any episode-like unit set apart by special surface structure features and corresponding to the Climax or Denouement in the notional structure" (Grammar 24). In his framework, climax "corresponds to the point of maximum tension and confrontation in a story", while denouement "corresponds to a decisive event that makes resolution of the plot possible" ("Discourse peak" 84). Either one or both of these may be marked as the action peak of the story. Longacre further notes that in addition to the action peak a narrative discourse may have a thematic or didactic peak. Other types of discourse, which do not focus on actions to the extent that narratives do can obviously have only thematic or didactic peaks. The present study is limited to action peaks, which could be identified on the basis of content features (cf. below).

Peak-marking is connected to two text-organizing principles, or patterns of narrative organization. One is grounding, that is, the distinctions between levels of foregrounded or main-line material and backgrounded or downgraded, secondary material (Warvik, "Grounding"); Longacre uses the term spectrum for the repertoire of signals for these distinctions ("Spectrum"). The other principle concerns the organization of the narrative in terms of its total structure from the beginning to the end, concentrating on the streams of increasing and decreasing tension, which culminate at peaks. In Longacre's terminology this property is modelled as the profile.

Discussing the identification of the peak, Longacre observes that "a discourse normally has a cumulative development which customarily occurs towards its end --or at least past its middle" where, in terms of surface structure features, "[t]he flow of discourse seems to quicken and grow more turbulent" ("Spectrum" 347) and "[r]outine features of the event-line may be distorted or phased out" (Grammar 25). He lists a number of strategies for creating such turbulence (Figure 1).
Figure 1. Peak-marking strategies (adapted from Longacre).

A. Change in the use of discourse markers, use of affective and
evaluative elements

B. Heightened vividness: shift in noun/verb ratio, tenses or persons,
shift between narrative and dialogue

C. Change of pace: change in unit length, variation in text type
(narrative, descriptive, instructive), change in the number and type
of connectives, change in the proportions of foregrounded and
backgrounded units

D. Concentration of participants: 'crowded stage'.

E. Rhetorical underlining: parallelism, paraphrase, repetition

F. Change of point of view, shift in focalization

The basic property in the peak-marking strategies listed in Figure 1 is a contrast with what has come before and what follows, and peak-markers can in this sense be characterized as contextually marked (cf. Chvany; Fleischman). Thus the peak zone of a text can be marked not only by adding signals in the discourse but also by changing or leaving out expected signals. Moreover, the different strategies are not independent of each other; for instance, the peak may be characterized by an increase in the frequency of a discourse marker, a decrease of unit length, and a shift to more verbs in proportion to nouns, but it will not be possible to determine which of these features is the primary peak marker.

Though peak-marking signals have mostly been studied in present-day data in non-Western languages (e.g., Brye; Hopkins; also Grimes; Longacre, Grammar, "Discourse peak"), some studies have investigated English narratives from different time periods and they have shown that there are several features which can be attributed the function of peak-marking. As examples of these, we can mention two lexical items. Enkvist ("Old English") calls the Old English adverbial pa 'then' an action marker, which occurs in passages of action and dramatic climaxes, but later characterizes pa as a foreground marker that can also function as a "'dramatizer' highlighting a dramatic view of stative conditions" ("More" 306). Brinton ("The importance") shows how anon 'at once' marks salient actions in Middle English narratives. Among grammatical features, the historical present is characterized in several studies as a feature associated with key events, turning-points and vividness. Its earliest instances in English are in Middle English texts (e.g. Fludernik; Ness and Duncan Rose). Schiffrin and Wolfson observe that tense switching between present and past, rather than the historical present alone, has discourse organizational functions in Present-Day English narratives, including episode shifts and evaluation. As another example, do has been characterized by Richardson as "a peak event marker" in Middle English poetry (qtd. in Brinton Pragmatic Markers 16), by Stein as a marker of discourse prominence in different types of text in Early Modern English, and by Rissanen as giving "textual emphasis" or "[underlining] the particular importance of a passage" (164) in American English texts from the early seventeenth century.

In an overview of peak signals in Old English narratives, Warvik ("Peakmarking") observes that the alternative strategies listed in Figure 1 can be used to describe the ways of marking the peak in a variety of extant Old English narratives. There is, however, a great deal of variation in the strategies and signals favoured in different texts, and thus it is necessary to look at peak-marking in more detail. Longacre's framework is chosen also in the present study, not only because it has proved applicable to Old English narratives, but also because, as a language-independent framework, it gives a basis for further studies comparing the findings with peak-marking strategies in other kinds of Old English texts and even cross-linguistically.

Let us finish this section with two example narratives. The annals for the years 449 and 1066 in the Anglo-Saxon Chronicle recount crucial events in the early history of England.

(1) AN. .ccccxlix. Her Martianus & Ualentines onfengon rice & ricsodon .vii. winter. & On hiera dagum Hengest & Horsa from Wyrtgeorne gelealaade Bretta kyninge gesohton Bretene on pam stape pe is genemned Ypwinesfleot, aerest Brettum to fultume, ac hie eft on hie fuhton. Se cing her hi feohtan agien Pihtas, & hi swa dydan & sige haefdan swa hwar swa hi comon. Hi da sende to Angle & heton heom sendan mare fultum & heom seggan Brytwalana nahtnesse & daes landes cysta. Hy da sendan heom mare fultum. Pa comon pa men of prim maegpum Germanie, of Ealdseaxum, of Anglum, of Iotum. Of Iotum comon Cantware & Wihtware, paet ys seo maeid de nu eardao on Wiht, & daet cynn on Westsexum pe man gyt haet Iutna cyn. Of Ealdseaxon comon Eastsexa & Sudsexa & Westsexan. Of Angle comon, sea siddan stod westi betwyx Iutum & Seaxum, Eastengla, Midelangla, Mearca & ealle Nordhymbra. (ChronA (Bately) 449) (1)

A.D. 449. This year Martianus and Valentinian assumed the empire and reigned seven winters. And in their days Hengest and Horsa, invited by Vortigern, king of the Britons landed in Britain in a place that is called Ipwinesfleet, at first to help the Britons, but they afterwards fought against them. The king commanded them to fight against the Picts, and they did so, and had victory wherever they came. They then sent to Angel and commanded them to send more assistance and to tell them of the worthlessness of the Britons and the excellence of the land. They then sent them more assistance. Then came the men of three nations of Germany, of Old Saxons, of Angles, of Jutes. Of the Jutes came the people of Kent and Isle of Wight, that is, the tribe that even now lives on the Isle of Wight, and that race in Wessex that people still call the race of the Jutes. Of the Old Saxons came the East-Saxons and South-Saxons and West-Saxons. Of the Angles, which has ever since remained waste between the Jutes and the Saxons, came the East Anglians, the Middle Anglians, the Mercians, and all the Northumbrians.

(2) AN. im.lxvi. Her fordferde Eaduuard king, & Harold eorl feng to dam rice & heold hit .xl. wucena & aenne daeg, & her com Willelm & gewann AEngla land. & her on dison geare barn Cristes cyrce. & her atiwede cometa .xiiii. kalendas Mai. (ChronA (Bately) 1066)

A.D. 1066. This year died king Edward, and earl Harold succeeded to the kingdom and held it forty weeks and one day, and this year came William and conquered England. And in this year Christ-Church [Canterbury] was burned. And this year appeared a comet on the fourteenth before the kalends of May.

These stories are well-organized as to the introduction, the beginning of the action, and the order of the events on the main story-line, but they are rarely characterized as interesting or exciting. However, they are from a chronicle and as the main purpose in that genre is to recount the events of the year in chronological order, the rather monotonous progress of the narrative is in fact the expected norm. The structural feature that is so remarkably absent from typical chronicle stories is the peak: the narrative proceeds evenly from one event to the next, but the story does not develop, there is no turning-point, no mounting tension, and no loosening point.

3. Materials and Methods

The sample of narratives investigated in the present study consists of forty stories selected from each of the following religious and historical works, representing four different genres: sermons, saints' lives, chronicles, and histories. 2 The study is limited to narratives to allow for looking into variation only within that one text type, and it is limited to prose texts in order to focus on genre-dependent variation by excluding stylistic differences between prose and poetry. It is a task for another investigation to study to what extent narrative poetry employs the same kinds of strategies of peak-marking.

AElfric's Catholic Homilies, Second Series (Godden ed.)

AElfric's Lives of Saints (Skeat ed.)

Anglo-Saxon Chronicle, Manuscript A (Bately ed.)

The Old English Orosius (Bately ed.)

These works vary in the proportion of narrative text type. The saints' lives by AElfric are all narrative, whereas the sermons in the Catholic Homilies range from purely narrative to fully instructive and argumentative. 3 While the Chronicle is almost exclusively narrative, the Orosius's history includes argumentative comments. Most of the narratives in these works can be described as 'compound' narratives, because they do not recount only one single story from the beginning to the end, but contain several stories, or substories, told successively or embedded into each other. Consequently, they may have several peaks, one (or more) in each story. However, as the stories in (1) and (2) show, not all narratives have a peak: they are episodic narratives.

The narratives selected for this study of peak-marking all include specific turning-points, and thus they are good candidates for what Longacre calls climactic narratives, that is, narratives characterized by tension (cf. Longacre Grammar 4-6). Obviously, for methodological validity, the potential signals of the peak could not be used as a basis for the study. To avoid this circularity, the peaks in the narratives were identified on the basis of key events that constitute the typical turning-points in the religious and historical stories: miracles, conversions, victories or defeats, martyrdoms and other such crucial deaths of main characters. The narratives in these works also include other kinds of turning-points, but as the decisions about their peak status would have involved a high degree of subjectivity, they are not included in the sample.

4. Peak-marking in Old English narrative prose

The features characterizing the potential peaks in the sample narratives can be grouped into contextually marked uses of discourse markers and variation of structural properties of narrative text. Table 1 presents the numbers of occurrences of the types of signals in the forty narratives selected from each text.

The figures show differences in peak-marking strategies between the texts, both in the preferences for certain types of signals and in the proportions of episodic narratives with unmarked peaks. In 26 cases we find no indications of a peak in the text where a key event is depicted; these 'peakless' narratives are episodic, like the stories of the conquests cited in (1) and (2). Not unexpectedly such narratives are most common in the Chronicle, where over half of the sample narratives have no marked peak. Moreover, a simple addition of the figures per text shows that many of the peaks are marked by a combination of signals. In those cases the different strategies together highlight the peak and their individual contributions cannot be exactly determined.

The biggest category of peak markers are the discourse markers gathered under the headings pa, 'suddenly', and 'until'. Of the total 160 potential peaks in the sample narratives, this category covers 112 cases; in 37 of them we find more than one signal. If we exclude the 26 cases with no marked peaks, there are only 22 narratives employing other strategies. However, discourse markers have a role in these cases as well, because as many as 14 of them involve a marked absence of pa in the story-line clauses at the peak (included in the category labeled 'structure'). Each of these categories is presented in turn below.

The Old English discourse marker that has probably attracted most attention is the adverb pa 'then', which, as noted earlier, is involved in peak-marking in addition to its other text-structuring tasks. Enkvist ("Old English") observed that in Beowulf high frequencies of pa correlate with fitts that are generally characterized as dramatic. In those fitts, the foregrounded story-line starts with a regular use of pa, which marks the progress of the story, until we reach the peak, where pa becomes more frequent, and after the peak pa returns to the regular pattern. Clusters of pa are also found in the peak zones of close to half of the sample narratives. To properly illustrate such changes in the frequency of pa would require longer examples than is possible within the limits of an article, and thus one short narrative passage will have to suffice. After the introduction in (3), the story proceeds in clauses linked by pa or and, depicting the alternating dialogue turns and the actions of praying and at the end departing. The only point in the story where two consecutive clauses contain pa is where Ephrem's miraculous acquisition of Greek is reported; it is also the only pa-clause which is co-ordinated by and.

(3) AEfter paera halgan messan, mid ham pe hi gereordodon, cwaed se halga Effrem to ham arwurdan biscope, Ic bidde pe, arwurda faeder, baer pu me anes binges tydige, Ic wat baet pu byst tyoa swa hwaes swa pu bytst aet Gode. Bide nu aet Gode paet ic grecisc cunne, pa cwaecd se biscop him to, pu baede ofer mine maede, ac uton swa peah biddan pas bena aet Gode. Hi feollan ba on gebedum, and Basilius cwaed, Hwi nelt pu la Effrem, nu du swa arwurde eart, beon maessepreost? And se oder him cwaed to, fordan be ic eom synful. Pa saede se biscop, eala gif ic haefde pine synna ana. Hi cneowdon ba aeft, and aeffrem pa spraec mid greciscum Gereorde, God herigende, and se halga biscop hine hadode to messepreost, and his wealhstod to diacone. And hi wendon eft ongean to loam widgyllan westene, wuldrigende God. (AELS (Basil), 510-525)

After the holy mass, while they broke their fast, the holy Ephrem said to the venerable bishop, 'I pray thee venerable father, to grant me one thing; I know that thou art a dispenser of whatsoever thou askest of God. Pray now to God that I may know Greek.' Then said the Bishop to him, 'Thou hast asked beyond my power, but let us, nevertheless, ask this boon of God.' They fell then to prayers, and Basil said, 'Behold! why wilt thou not, Ephrem, since thou art so worthy, become a mass-priest?' And the other said to him, "Because I am sinful.' Then said the bishop, 'Oh, if I had but thy sins only!' Then they knelt again; and Ephrem then spake in the Greek tongue, praising God, and the holy Bishop ordained him as mass-priest, and his interpreter as deacon, and they went back again to the wider-stretching desert, glorifying God. (Translation from Skeat 79.)

The adverb pa is a multifunctional discourse marker, whose various tasks can be subsumed under the core function of signalling the main line of the story. The hierarchical organization of the narrative text into episodes is expressed by an interaction of pa and other expressions of temporal relations. While the semantic content of pa is simple temporal sequentiality, 'then', at episode boundaries it is supplemented or replaced by semantically heavier expressions of temporal relations, such as aefter paeere maessan, mid pam pe hie gereordodon 'after the holy mass, while they broke their fast' in example (3). At the peak of the story, pa cooperates with expressions of 'suddenness', such as paerrihte, sona, faerlice, hraedlice, which can be rendered in Present Day English 'suddenly, immediately, straightway, at once' (cf. Brinton "The importance" on anon 'at once' in Middle English). There are 29 cases in the sample where, pa co-occurs with such expressions and 17 cases where an expression of suddenness occurs alone at the peak. These expressions are common in the religious texts by AElfric, with a preference for paerrihte in the homilies (13 cases) and sona in the saints' lives (11 cases). In the following examples/ha marks the progress of the story by heading strings of co-ordinated clauses, until we get to the miracle, where instead of pa we find paerrihte (4) and sona (5),

(4) Sum gelyfed man gebaed aet paere cyrcan for his adligan dehter, and hire reaf pider abaer, efne da da he ham gecyrde, pa urnon his hiwan him togeanes, and hire fordsid him gecyddon; He da mid pam reafe paet lic oferwreah, and seo dohtor paerrihte to life aras; (AECHom II, 2, 13.62-65)

A certain religious man prayed at the church for his sick daughter, and brought her garment there. Indeed when he returned home, then his family hurried towards him, and told him about her death; He then covered her dead body with the garment, and the daughter arose straightway to life;

(5) AEfter disum weard paes eadigan cudberhtes cneow mid heardum geswelle alefed, swa paet he mid criccum his fedunge underwredode; pa gesaet he sume daege under sunnbeame ana on sundran, and his sceancan bedode, him com da ridende to sum arwurde ridda sittende on snawhwitum horse, and he sylf mid hwitum gyrlum befangen waes. and he done halgan mid gesibsumum wordum swaeslice grette, biddende paet he him daegwistes gedafenlice tidode; Cudberhtus da to dam engle anmodlice cwaed; lc wolde dine denunge sylf nu gearcian. gif ic me mid fedunge ferian mihte; Min adlige cneow is yfele gehaefd. paet ne mihte nan laecewyrt awiht gelidian. deah de heo gelome to geled waere; pa gelihte se cuma. and his cneow grapode, mid his halwendum handum and het hine geniman hwaetene smedeman. and on meolce awyllan, and swa mid daere haetan paet todundene lim gewridan, and aefter disum wordum his hors bestrad, on dam sidfaete de he dider com aweg ferende; Hwaet da cudberhtus aefter paes engles lare his cneow bedode. and he sona gesundfull his faereldes breac, and ongeat paet god purh his engel hine geneosode, se de giu aer pone blindan Tobian durh his heahengel Raphahel mihtelice onlihte:

(AECHom II, 10, 82.28-44)

After this the holy Cuthben's knee was afflicted by a severe swelling, so that he supported his gait with crutches; Then he sat one day in the sunshine alone by himself and warmed his leg. To him came then riding an honorable rider sitting on a snow-white horse, and he himself was clothed in snow-white robes, and he greeted the saint pleasantly with kind words, asking that he would give him a meal; Cuthbert then said to the angel without hesitation; I would myself prepare your meal, if! could move by walking. My sore knee is so badly diseased that no medicinal herb can give any relief at all though it was frequently applied; Then the stranger alighted, and felt his knee with his healing hands, and told him take fine wheaten flour, and boil it in milk, and so with that heat bandage the swollen limb and after these words remounted his horse going away on the journey on which he came there; Indeed then Cuthbert according to the angel's advice bathed his knee, and he at once could enjoy his ability to walk, and understood that God through his angel visited him, he who had earlier through his archangel Raphael mightily given sight to the blind Tobias;

Another group of items which accompany pa at the peak are the interjections hwaet 'indeed, what!, lo!' and efne 'behold!, truly!, indeed!', which function as intensifying elements emphasizing the dynamicity of pa (cf. Brinton, Pragmatic Markers, on hwaet pa marking result or consequence). In the sample narratives, such elements occur only in the religious texts: both occur three times in the homilies, and efne appears twice and hwaet five times in the saints' lives. In example (5) above, we find hwaet together with pa in the clause preceding the miracle. In example (6), AElfric highlights the miraculous events accompanying the death of Servulus by a cluster of intensifying elements occurring together with ba: efne da da .. pa 'indeed when .. then', hwaet da faerlice da da .. da 'Lo then suddenly when .. then', efne da mid pam pe .. da 'indeed then as ... then'. Each stage gets additional salience from the parallel structures with the temporal subordinate clauses presenting one successive event as backgrounded to the other, depicted in the main clause. The final stage of the miracle contrasts with these in being presented in a clause introduced by pa alone.

(6) pa da se tima becom paet his miccle gedyld wurde gewuldrod fram gode da awende seo sarnyss ealra his lima to daere heortan; Efne da da he ongeat paet se dead him genealaehte. pa baed he da aeldeodigan weras de on cuman hiwe him mid wunodon. paet hi astodon and on his fordside heora sealmas sungon; Hwaet da faerlice da da he sylf mid pam aeldeodigum preostum sang. da clypode he mid micclum ogan. and heora sang gestilde, and cwaed. Su wiad. Hwaet la. ne gehyre ge hu myrige lofsangas swegad on heofonum? Efne da mid loam toe he hlyste daes heofonlican sanges. da gewat his sawul of dam geswenctan lichaman, to ecere reste; pa weard paet hus afylled mid wunderlicum braise, swa paet ealle da licmenn wurdon afyllede mid dam wynsumum stence, and se braed on heora nosdyrlum ne ateorode. od paet se halga lichama bebyriged waes; Swa ageaf pes goda mann his waestm gode purh gedyld, for dan pe he forbaer godes swingele swide emlice, and siddan to edleanes aecre becom; (AECHom II, 6, 58.182-195)

When the time came that his great patience was to be glorified by God, then the pain of all his limbs turned to the heart; Indeed when he understood that death was approaching him, then he begged the foreign men who were staying with him as guests that they stand up and sing psalms at his departure; Lo then suddenly when he himself sang with the foreign priests, then he called out in great terror and silenced their song and said. Be silent. Why! Do you not hear how merry songs of praise resound in heaven? Indeed then as he listened to the heavenly song, then his soul left the afflicted body for eternal rest; Then the house was filled with a wonderful odour so that all the pallbearers were filled with that pleasant scent, and the odour in their nostrils did not leave until the holy body was buried; So this good man gave his produce to God through patience because he endured God's afflictions very patiently and afterwards came to the land of retribution;

While the peak-marking strategies presented so far involve contrast in the frequencies of items, the category labeled 'until' in Table 1 includes 27 cases where the contrast arises from the semantic sense of the connecting item (op, oppe, oppaet) and the situation types of the story-line elements. In the historical narratives, particularly in Orosius, we find instances of peak events reported in a sequence of clauses which depict an activity or state, i.e., a durative or stative situation without an end-point, which is terminated by an accomplishment or achievement, i.e., a durative or punctual situation with an end-point. In 15 cases 'until' appears on its own at the peak; example (7) is one of the small stories, where op marks the turning-points (for small stories, see Bamberg). Example (8) is one of the eight cases where this kind of sequence occurs together with ha. In the remaining four instances, this strategy is accompanied by structural variation, one of them with hraedlice 'at once, forthwith'.

(7) & sona aefter paem, toy ilcan geare, Darius gefor Persa cyning. & his II suna ymb toast rice wunnon, Artecserses & Cirus, op hiora aeegper paet maeste folc ongean operne geteah, & pa unsibbe mid gefeohtum dreogende waeron op Cirus ofslagen weard, se paer gingra waes. (Or 2, 7.51.9-13)

and soon after that, the same year, the Persian king Darius passed away, and his two sons fought over the kingdom, Artexerxes and Cyrus, until both of them drew most of the people against the other and they were fighting the battles with hostilities until Cyrus was killed, who was the younger (= vassal) there.

(8) He pa Romulus aefter piosan underfeng Cirinen gewinn loara burgwarana, for pon pe he pagiet lytel landrice haefde buton paere byrig anre, for loon toe Romulus & ealle Romware operum folcum unweorde waeran, for pon de hie on cnihthade waeron operra manna niedlingas. pa hie da haefdon Cirinensa pa burg ymbseten, & daer micelne hungor poliende waeron, pa gecwaedan hie baer him leofre waere baer hie on daem iermbum heora lif geendodon bonne hie daet gewinn forleten, odde frid genamen. Hie daer pa winnende waeron od hie pa burg abraecon, & aefter paem wid pa londleode on aelce healfe unablinnendlice winnende waeron, od hie daerymbutan haefdon monega byrig begietena. (Or 2, 2.39.25-35)

He then Romulus after this undertook war on the citizens of Caenina, because he still had little territory except the town only, because Romulus and all Romans were considered worthless by other people, because they had in their youth been bondmen of others. When they then bad besieged the city of Caenina, and were there suffering great hunger, then they agreed that they thought it better that they ended the lives of the wretches than that they abandon the baule or make peace. They were then fighting there until they destroyed the city and after that were fighting incessantly against the people of the country on all sides, until they had taken many cities thereabouts.'

As we have seen, the majority of the peaks in the sample narratives are characterized by uses of discourse markers, but other kinds of strategies also occur, though often accompanied by discourse markers. There are 19 instances where shifts between narration and dialogue coincide with peak events, but all of them also involve uses of discourse markers; 11 coincide with pa, two with 'suddenly' and six with both. Of the 47 cases where the structural properties of narratives are used for creating contrasts at the peak, 25 reinforce the turning point with discourse markers, which leaves 22 peak events marked by other strategies than the presence of discourse markers. In as many as 14 of these we find a contrasting absence of pa in the story-line clauses at the peak, which is a strategy that could be counted among uses of discourse markers, but these cases also involve changes of pace through unit length and types of connectives. This strategy is most common in the Chronicle, where we find six peak events reported like the victory ending the sequence of events in (9). In five cases in the Orosius the peak event is accompanied by shifts between narration and description or commentary, as at the end of the chapter cited in (10).

(9) & py ilcan geare sende Aelfred cyning sciphere on Eastengle. Sona swa hie comon on Stufe mulaan, pa metton hie .xvi. scipu wicenga & wip da gefuhton & pa scipo alle gerehton & pa men ofslogon. Pa hie pa hamweard wendon mid paere herehype, pa metton hie micelne sciphere wicenga & pa wip pa gefuhton py ilcan daege, & pa Deniscan ahton sige. (ChronA (Bately) 885)

And the same year king Alfred sent a fleet into East Anglia. As soon as they came to the mouth of the Stour, then they met sixteen ships of the pirates and fought with them and captured all the ships and killed the men. When they then returned homeward with their booty, then they met a large fleet of the pirates and then fought with them the same day, and the Danes had the victory.

(10) AEfter paere pe Romeburg getimbred waes iii hund wintra & lxxxviii, paette Gallie oferhergedan Romana lond od iiii mila to daere byrig, & pa burg mehton eade begitan gif hie paer ne gewacadon, for pon Romane waeron swa forhte & swa aemode, paet hie ne wendon paet hie pa burg bewerian mehton. Ac paes on morgenne Titus heora ladteow, pe odre noman waes haten Quintius, hie mid firde gesohte, paer gefeaht Mallius anwig, pe odre noman waes haten Tarcwatus, wid anne Galliscne monn, & hiene ofslog; & Titus Cuintius pa odre surne gefliemde, sume ofslog. Be paem mon mehte ongietan hwaet daer ofslagen waes, pa heora fela dusenda gefongen waes. (Or 3, 4.57.23-58.4)

After the city of Rome was built three hundred and eighty-eight winters, (it happened) that Gauls plundered the Roman territory until four miles into the city, and might easily have gained possession of the city if they had not become sluggish, for the Romans were so fearful and disheartened, that they did not think that they could defend the city. But in the morning, Titus their general, who by other name was called Quintus, attacked them with an army. There Manlius, who by other name was called Torquatus, fought in a single combat against a Gaul, and killed hire; and Titus Quintus then put some of the rest to right and killed others. By that one may understand how many there were killed, when many thousand were captured.

5. Conclusion

While the overview of peak markers by Warvik ("Peak-marking") listed alternative ways of signalling the peak in Old English narratives, the present study of a sample of narratives has revealed that there are differences between the frequencies of and the preferences for different strategies in different kinds of narrative texts.

Most importantly, this study has provided more evidence for the crucial role of discourse markers, especially pa, in the organization of Old English narrative text. When we count both the occurrences and the marked absences of discourse markers in reports of peak events, we find discourse markers involved in 126 peaks, which is 94 per cent of the marked peaks in the sample narratives. Pa operates at 90 peaks, in 76 cases by presence and in 14 by contrasting absence, which together account for over two thirds of all cases of discourse marker uses. The prominent structuring role of pa is hardly surprising in view of earlier research, but the focus of earlier studies has not been on the function of pa as a peak signal and its cooperation with other ways signalling the peak. A question that these findings give rise to concerns the vividness traditionally associated with the verb-subject order. As pa is a highly frequent word in Old English and in initial position it is by rule followed by this inverted word order, it is possible that the vividness of pa-verb-subject clauses is not a feature arising from the word order but from the use of pa. This idea is reinforced not only by discourse linguistic studies attributing various discourse functions to pa, but also by syntactic research showing how the interaction of pa, word order and other syntactic factors contribute to expressing discourse relations (cf. e.g., Los). More research is needed to identify the roles of the individual factors in their interaction.

In addition to emphasizing the role of discourse markers in structuring Old English narratives, the findings of this study underline the importance of genres and text types in the study of discourse features also in historical materials. Though discourse markers play a role in peak-marking in all the four texts from which the sample narratives are selected, there are great differences between the preferred strategies. The Chronicle stands apart from the other texts in its high proportion of episodic narratives, which have no marked peaks in the reports of key events; this is totally in line with characterizations of chronicles as rather unexciting lists of events. Though the Orosius also reports a great number of chronologically ordered events, the sample narratives display a more complex structure with marked peaks in climactic narratives. Comparing the historical narratives with the religious ones, we find preferences for different kinds of peak markers. Both the sermons and the saints' lives make more use of pa and expressions of suddenness than either of the historical texts. Instead, Orosius and to a lesser extent the Chronicle have more instances of 'until' sequences at the peak. It is tempting to see this difference as a concomitant of emphases on different kinds of sequences of events: the reports of miracles and conversions in the religious texts focus on the suddenness of the change, whereas the accounts of victories and defeats in the historical narratives highlight the duration of the situation leading to the resolving event. Furthermore, we find differences in the numbers of markers. The peaks in the religious texts tend to be both more systematically marked and more often marked by multiple signals than the peaks in the historical narratives. This tendency tallies nicely with the underlying instructive purpose of sermons and saints' lives, which is prominent in AElfric's use of stories (cf. Szarmach; Warvik "Teaching").

Though the narratives investigated in the present study are limited to samples of prose from four texts, the findings still suggest noteworthy differences in the ways of signalling the peak between different genres of religious and historical narratives. By focusing on peak-marking, which is an aspect of narrative structure that has not received much attention earlier, this study underlines the central role of discourse markers, especially Ira, in the organization of Old English narrative text.

Works Cited

Aijmer, Karin, and Anne-Marie Simon-Vandenbergen. "Pragmatic Markers." Handbook of Pragmatics Online. Amsterdam: John Benjamins, 2009. Web. 31 Jan. 2012 <>.

Bamberg, Michael G. W. "Stories: Big or small. Why do we care?" Narrative: State of the Art. Ed. Michael G. W. Bamberg. Amsterdam: John Benjamins, 2007. 165-174. Print.

Bately, Janet M., ed. The Old English Orosius (= EETS, SS, 6). London: Oxford University Press, 1980. Print.

--. The Anglo-Saxon Chronicle: A Collaborative Edition. Volume 3 : MS A. Cambridge: Brewer, 1986. Print.

Brye, Ann Elizabeth. "Features marking peak and peak' in Hausa folktales." Journal of West African Languages 21, 1991. Web. 24 Jan. 2012. <>.

Brinton, Laurel J. Pragmatic Markers in English: Grammaticalization and Discourse Functions. Berlin: Mouton de Gruyter, 1996. Print.

--. "The importance of discourse types in grammaticalization: The case of anon." Textual Parameters in Older Languages. Ed. Susan Herring, Pieter van Reenen, and Lene Schosler. Amsterdam: John Benjamins, 2000. 139-162. Print.

Chvany, Catherine V. "Backgrounded perfectives and plot line imperfectives: Toward a Theory of Grounding in Text." The Scope of Slavic Aspect. Ed. Michael S. Flier and Alan M. Timberlake. Columbus, Ohio: Slavica, 1985. 247-73. Print.

Enkvist, Nils Erik. "Old English adverbial pa--an action marker?" Neuphilologische Mitteilungen 73 (1972): 90-96. Print.

--. "More about the textual functions of Old English adverbial pa." Linguistics across Historical and Geographical Boundaries: In Honour of Jacek Fisiak on the Occasion of his Fiftieth Birthday. Vol. 1: Linguistic Theory and Historical Linguistics. Ed. Dieter Kastovsky and Aleksander Szwedek. Berlin: Mouton de Gruyter, 1986. 301-309. Print.

--. "Problems raised by Old English pa." Writing vs. Speaking: Language, Text, Discourse, Communication. Ed. Svetla Cmejrkova, Frantisek Danes, and Eva Havlova. Tubingen: Gunter Narr, 1994. 55-62. Print.

Enkvist, Nils Erik and Brita Warvik. "Old English/ha, Temporal Chains, and Narrative Structure." Papers from the 7th International Conference on Historical Linguistics. Ed. Anna Giacalone Ramat, Onofrio Carruba, and Giuliano Bernini. Amsterdam: John Benjamins, 1987. 221-37. Print.

Fleischmann, Suzanne. Tense and Narrativity: From Medieval Performance to Modern Fiction. Austin: U of Texas P, 1990. Print.

Fludernik, Monika. "The Historical Present Tense in English literature: An oral pattern and its literary adaption." Language and Literature 17 (1992): 77-107. Print.

Foolen, Ad. "Pragmatic Particles." Handbook of Pragmatics Online. Amsterdam: John Benjamins, 1996. Web. 31 Jan. 2012. <>.

Foster, Robert. "The use of pa in Old and Middle English narratives." Neuphilologische Mitteilungen 76 (1975): 404-14. Print.

Georgakopoulou, Alexandra. "Narrative". Handbook of Pragmatics Online. Amsterdam: John Benjamins, 1997/2005. Web. 31 Jan. 2012. <>.

Godden, Malcolm, ed. AElfric's Catholic Homilies. The Second Series Text (= EETS, SS, 5). London: Oxford UP, 1979. Print.

Grimes, Joseph E., ed. Papers on Discourse. Arlington, Texas: Summer Institute of Linguistics, 1978. Print.

Hopkins, Mary L. "Narrative Peak in Xangufa Zapotek." SIL-Mexico Workpapers 11 (1995): 17-36. Web. 24 Jan. 2012. < scans/WS001i-WS012i-WorkpaperScans.htm>.

Hopper, Paul J. "A discourse Perspective on Syntactic Change: Text-building strategies in Early Germanic." Reconstructing Languages and Cultures. Ed. Edgar C. Polome and Werner Winter. Berlin: Mouton de Gruyter, 1992. 217-238. Print.

Lenk, Uta. "Discourse markers." Handbook of Pragmatics Online. Amsterdam: John Benjamins, 1997. Web. 31 Jan. 2012. <>

Longacre, Robert E. "'Mystery' particles and affixes." Papers from the Twelfth Regional Meeting Chicago Linguistic Society, April 23-25, 1976. Ed. Salikoko S. Mufwene, Carol A. Walker, and Sanford B. Steever. Chicago: Chicago Linguistic Society, 1976. 468-77. Print.

--. "A spectrum and profile approach to discourse analysis." Text 1 (1981): 337-59. Print.

--. The Grammar of Discourse. New York: Plenum Press, 1983. Print.

--. "Discourse peak as a zone of turbulence." Beyond the Sentence: Discourse and sentential form. Ed. Jessica R. Wirth. Ann Arbor, MI: Karoma Publishers, 1985.83-98. Print.

Los, Bettelou. "Onginnan/beginnan with bare and to-infinitive in AElfric." Pathways of Change: Grammaticalization in English. Ed. Olga Fischer, Anette Rosenbach, and Dieter Stein. Amsterdam: John Benjamins, 2000. 251-274. Print.

Ness, Lynne and Caroline Duncan-Rose. "A syntactic correlate of style switching in the Canterbury Tales". Papers from the 3rd International Conference on Historical Linguistics. Ed. J. Peter Maher, Allan R. Bomhard, and E.F. Konrad Koerner. Amsterdam: John Benjamins, 1982. 293-322. Print.

Richardson, Peter. 1991. Narrative strategies in early English poetry. Unpublished Ph.D. dissertation, University of California, Berkeley, 1991. Print.

Rissanen, Matti. "Periphrastic do in affirmative statements in Early American English" Journal of English Linguistics 18 (1985): 163-183. Print.

Schiffrin, Deborah. "Tense variation in narrative." Language 57 (1981): 45-62. Print.

Skeat, Walter, W. ed. AElfric's Lives of Saints, Volume I (= EETS, 76 & 82). London: Oxford University Press, 1966. Print. [Originally published in two volumes in 1881 (76)and 1885 (82)]

Stein, Dieter. The Semantics of Syntactic Change: Aspects of the evolution of do in English. Berlin: Mouton de Gruyter, 1990. Print.

Szarmach, Paul E. "AElfric as exegete: Approaches and examples in the study of the Sermones Catholici." Hermeneutics and Medieval Culture. Ed. Patrick J. Gallacher and Helen Damico. Albany, NY: State University of New York Press, 1989. 137-147. Print.

Virtanen, Tuija and Brita Warvik. "Observations sur les Types de Textes". 8e Rencontre de Professeurs de Francais de l'Enseignement Superieur. Communications Ed. Juhani Harma, and Iris Makinen-Schwanck. Helsinki: Departement des Langues Romanes de Universite de Helsinki, 1987. 91-114. Print.

Warvik, Brita. "Peak-marking in Old English narrative." Organization in Discourse: Proceedings from the Turku Conference. Ed. Brita Warvik, Sanna-Kaisa Tanskanen, and Risto Hiltunen. Turku: University of Turku, 1995. 549-558. Print.

--. "Grounding". Handbook of Pragmatics Online. Amsterdam: John Benjamins, 1996/2011. Web. 30 Jan. 2012. <http://>.

--. "Teaching by stories: AElfric's instructive narratives." Instructional Writing in English: Studies in Honour of Risto Hiltunen. Ed. Matti Peikola, Janne Skaffari, and Sanna-Kaisa Tanskanen. Amsterdam: John Benjamins, 2009. 13-34. Print.

--. "Connective or 'disconnective' discourse marker? Old English pa, multifunctionality and narrative structuring." Connectives in Synchrony and Diachrony in European Languages (Studies in Variation, Contacts and Change in English, 8). Ed. Anneli Meurman-Solin and Ursula Lenker. 2011. Web. 28 Dec. 2011. <>.

Wolfson, Nessa. "The conversational historical present alternation". Language 55 (1979): 168-182. Print.

Brita Warvik

Abo Akademi University


(1) The references follow the system of the Dictionary of Old English at http://

(2) Narratives were identified by the criteria of temporal sequentiality and agent orientation (cf. Georgakopoulou; Longacre Grammar; Virtanen and Warvik).

(3) AElfric's works were selected for this study because his sermons include a great number of stories. This distinguishes him from Wulfstan, whose sermons are almost fully hortatory and argumentative as to their text type and thus the peaks in his texts are typically didactic or thematic peaks. In the few cases where Wulfstan tells a story which has an action peak, his ways of marking the peak are similar to AElfric's.

(4) The category 'suddenly' subsumes various expressions of suddenness: poerrihte, sona, foerlice, hroedlice.

(5) The category 'until' includes the conjunctions op, oppe and oppoet.
Table 1. Peak-marking strategies in the sample narratives
(40 narratives / text)

                'suddenly'   'until'
text       pa      (4)         (5)     speech   structure   no marking

AECHom     18           22         1        4          10            1
AELS       37           17         1       14          19            0
ChronA      5            1         8        0           7           23
Or         16            6        17        1          11            2
Total      76           46        27       19          47           26
COPYRIGHT 2013 Northern Illinois University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Warvik, Brita
Article Type:Critical essay
Geographic Code:4EUFI
Date:Jun 22, 2013
Previous Article:Framing the reader in early science fiction.
Next Article:Carpaccio and D'Annunzio's Il Fuoco.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters