Communicative dynamism and prosodic prominence in presentation sentences with initial rhematic subjects
Abstract
57
Full text
Communicative dynamism and prosodic prominence in presentation sentences with initial rhematic subjects Martin Adam— Irena Headlandová Kalischová (Brno) ABSTRACT Within the framework of the theory of functional sentence perspective (Firbas 1992), the distinction between presentation and quality scale sentences plays avital role. The present paper proposes to shed light on one of the most common configurations of presentation sentences, viz. structures with initial rhematic subject (e.g. an uninvited dwarf came), examining the way native speakers place the intonation centre in such structures, i.e. to map the correspondence between the degrees of communicative dynamism and prosodic prominence. For the purpose of the investigation selected chapters from Tolkien’s The Hobbit are used. KEYWORDS FSP, presentation, scale, intonation centre, communicative dynamism 1. INTRODUCTION Apart from acareful selection of data and aminute and profound analysis of it, an appropriate interpretation of the text material under analysis seems to stand at the root of linguistic research of all sorts across schools, approaches and methodologies. It follows that within the theory of functional sentence perspective (henceforward referred to as FSP)1— in relation to its context-orientation and multifaceted character— this postulate holds true perhaps even more so. Among other aspects, it is especially the relatively blurred and at times perhaps indistinct borderline between the two dynamic semantic scales (Presentation Scale and Quality Scale) that has raised many aquestion in recent FSP research (i.e. Dušková, 1999, 247–272, 2015, 202–208, 256–268; Svoboda, 2005, 2006; Chamonikolasová, 2007, 2010; Adam, 2013, 157–165, 2014, 2016; Rohrauer, 2015, 148–150, 162–164; Brůhová and Malá, 2017). The present paper proposes to discuss one of the most common configurations of presentation sentences (Pr-sentences), viz. structures with an initial rhematic subject. Since the prototypical Pr-sentences of this sort actually violate the end-focus principle (with the most prominent, rhematic element occupying the initial position), the prosodic treatment of these utterances in spoken discourse then becomes an important issue. The research objective of this paper is to examine the way native speakers place the intonation centre in such structures, i.e. to map the correspondence between the degrees of communicative dynamism and prosodic prominence. 1 For the full list of abbreviations used, see the List of Abbreviations at the end of the article.
58 LINGUISTICA PRAGENSIA 1/2017 2. CORPUS AND METHOD For the purpose of the proposed discussion, the authors decided to analyse J. R. R. Tol - kien’s novel The Hobbit, specifically the written form against its spoken version (an audiobook narrated by Rob Inglis2). The corpus data encompasses chapters I-Vof the novel, representing ca 27,800 words, which amounts to 2,268 basic distributional fields. The actual procedure comprised several stages: first, all Pr-sentences were extracted manually and further subdivided into four types, second, the recorded text was subjected to aphysical perception test and the prosodic treatment of the selected utterances was assessed on an auditory basis, and finally, the correspondence between the distribution of communicative dynamism and that of prosodic prominence was determined. Below are charts showing basic corpus statistics (the incidence of Presentation and Quality Scale sentences; see Table 1) and the statistical occurrence of individual types of Pr-sentences (Table 2) respectively. The ratio of Prand Q-Scale sentences (8% vs. 92%) virtually corresponds with other recent research data (Adam, 2013, 153–156, 2016; Rohrauer, 2015, 148–150; Stehlíková, 2016, 109–111, 123). Also the syntactic realization of Pr-sentences (with Type Ibeing the most frequent and Type IV represented just marginally) is in concord with the usual make-up of fiction narrative texts examined previously (Dušková, 1998; Adam, 2013). Scales Abs. % Pr 182 8 Q 2,086 92 Total 2,268 100 table 1.Corpus statistics Type of Pr-sentences Abs. % I(existential there-construction) 86 47.3 II (initial rhematic subject) 74 40.7 III (fronted adverbial + subject-verb inversion) 20 10.9 IV (locative subject) 2 1.1 Total 182 100.0 table 2.Types of presentation sentences 3. PRESENTATION SENTENCES The idea of dynamic semantic scales, implemented in sentences, was introduced by Firbas (see especially 1992, 66–69); the scales functionally reflect the distribution of 2 The audiobook is available at http://esl-bits.net/ESL.English.Learning.Audiobooks/ The.Hobbit/indice.html
MARTIN ADAM— IRENA HEADLANDOVá KALISCHOVá 59 communicative dynamism (CD) and operate irrespective of word order. In principle, Firbas distinguishes two types of such scales: the Presentation Scale and the Quality Scale. In these scales, each element is ascribed one of the dynamic-semantic functions (DSFs). In contrast to astatic approach towards semantic functions of sentence constituents, the dynamic-semantic functions may change in the course of an act of communication; the same element may thus perform different functions in different contexts and under different conditions (Chamonikolasová, 2010; cf. also Firbas, 1992, 66ff; Svoboda, 2005, 225). As has been mentioned above, the present paper focuses on the sentences that implement the Presentation Scale. This scale generally includes three basic dynamic- -semantic functions: proceeding in the interpretative arrangement from the least to the most dynamic element, the first position is taken by the thematic Setting of the action (Set), usually temporal and spatial items regarding when and where the action takes place. Second, the existence or appearance on the scene is typically conveyed by averb by means of the Presentation of Phenomenon (Pr); as Firbas says, aPr-verb expresses “existence or appearance on the scene with explicitness or sufficient implicitness” (Firbas, 1995, 65). Finally, the major, most dynamic element, the Phenomenon to be Presented (Ph), is literally ushered onto the scene. The average occurrence of Pr-sentences in texts has been determined to range from 5 to 12% of all distributional fields (Adam, 2013, 61ff, 153–156). In previous research as well as in the data under analysis, four principal syntactic patterns have been identified and labelled as Types I-IV (see Table 2 above; for more details on these, see e.g. Adam, 2013, 59ff, 156; Dušková, 2015, 202–208). To exemplify each of the types, selected examples (1–4) excerpted from the research corpus are adduced below; the rhematic Phenomena are presented in bold and the number in square brackets indicates the item number in the corpus. — Type I: Existential there-construction (1) There were lots of dragons in the North in those days... [18] — Type II: Preverbal Rh-subject (2) ...and awaning moon appeared above the hills between the flying rags. [115] — Type III: Fronted Adverbial + S-VInversion (3) Then suddenly out of the gloom came asharp hiss. [179] — Type IV: Locative Subject + Rh-object (4) It (=the cave) had adry floor and some comfortable nooks. [182] As proposed, the present discussion is above all preoccupied with the second most recurring subtype of Pr-Scale sentence pattern, viz. that with arhematic subject in the initial, i.e. preverbal position. Such aconfiguration is perceived as the prototypical, canonical type connected with the Presentation Scale (cf. Dušková et al., 1988, 62, 531–532). In it, the initial sentence element is typically represented by acontext-independent subject, which is only then followed (in concord with the requirements of
60 LINGUISTICA PRAGENSIA 1/2017 the English word order principles) by the verb expressing existence or appearance on the scene (cf. Dušková, 1999, 248–250); the sentence may also open with ascene-setting temporal or spatial thematic adverbial, the position of which is rather volatile. As aresult, the end-focus principle is not applied. It should be noted, however, that sentences with arhematic subject in apreverbal position are, as arule, considered unmarked by native speakers of English (exceptions may be observed in connection to prosodic re-evaluating intensification; cf. Firbas, 1992, 154–156). In Czech, for that matter, the situation is reversed: the end-focus principle is most closely observed while an initialrhematic subject is untypical if not highly exceptional and can be justified only under special prosodic or other emphatic conditions (cf. Dušková, 1999, 281). 4. INTONATION: AFACTOR IN FSP INTERPLAY In the framework of the theory of FSP, the sentence is viewed as “afield of semantic and syntactic relations that in its turn provides adistributional field of degrees of communicative dynamism (CD)” (Firbas, 1992, 7–8); individual sentence elements then serve as communicative units with different degrees of CD. The degrees of CD are determined by the interplay of FSP factors involved in the distribution of degrees of CD. The FSP factors (formative forces) are (i) linear modification, (ii) context, and (iii) semantics. In spoken language, their interplay is logically joined by afourth factor— (iv) intonation (Firbas, 1985, 1992, 14–16). In harmony with the principal research question, the present discussion ta ckles the difference between the distribution of degrees of communicative dynamism (triggered by non-prosodic FSP factors) on the one hand and the degrees of prosodic prominence (PP) on the other. Intonation can confirm or modify the interplay of the other three (non-prosodic) factors and it is also capable of disambiguating possible cases of interpretative potentiality triggered by syntactic-semantic structures (cf.Headlandová Kalischová, 2010a, 45–47); thus, intonation may even change the overall distribution of CD within acommunicative field. Major phonological studies maintain that there is aconnection between the nucleus placement and the information load of individual lexical units in that the nucleus mostly occurs on words carrying new information (see e.g. O’Connor and Arnold, 1973, 5; Cruttenden, 1986, 94; or Roach, 1994, 176). In his key monograph, Firbas (1992) examined the relation between the operation of non-prosodic FSP factors (linear modification, context and semantics) and the features affecting degrees of PP. Even though the relationship of the two concepts is extremely close, one cannot say that their coincidence is absolute. Basically, there are three types of relationship between the distributions of CD and PP (Firbas, 1992, 148–172): (I) PERFECT CORRESPONDENCE In perfect correspondence between CD and PP, the intonation exactly reflects the information structure as determined by the non-prosodic factors; in other words, the nuclear stress falls on the rhematic element. In example (5) below, the expression aflame is the carrier of the most dynamic piece of information (in bold) and, at the
MARTIN ADAM— IRENA HEADLANDOVá KALISCHOVá 61 same time, it carries the nuclear stress (underlined), i.e. it becomes the intonation centre (IC) of the utterance. The example is supplemented with the FSP functions and basic tonetic marks inserted. (5) Suddenly in the wood beyond The Water aflame leapt up. [101] ' Th \Rh Tr (II) NON-REEVALUATING INTENSIFICATION The following example demonstrates that this kind of intensification does not affect the basic relation between the theme and the rheme although it increases the amount of CD assigned to one of the units (Headlandová Kalischová, 2010a, 46). The non-prosodic FSP analysis is accompanied by the last line showing the prosodic intensification of the thematic subject You. The reason for this intensification can most likely be put down to emotive colouring; consequently the degree of CD of this unit (communicatively the least dynamic) is increased to that of adiatheme (DTh). (6) [...and so Itold him the truth.—] You must be joking! Th Tr Rh \DTh Tr \Rh (III) RE-EVALUATING INTENSIFICATION Sometimes, on the other hand, the intensification produces astronger deviation from the perfect correspondence between the distribution of the degrees of CD and PP. The point is that intonation can re-evaluate the information structure (i.e. affect the theme-rheme articulation) and in that way play down the non-prosodic factors. In the example sentence (7), the element That represents the only nucleus and thus the most prominent prosodic item (the intonation centre) in the distributional field; it is re-evaluated from an originally thematic unit into arhematic one (Headlandová Kalischová, 2010b, 89). As aresult of this re-evaluating prosodic intensification, the sentence is emotionally marked. Such markedness may be regarded as an irretrievable piece of information that enables the otherwise dynamically weak element to become the rheme of the sentence. (7) That’s good! → That’s good! Th Tr Rh \/ Rh Tr DTh 5. ANALYSIS AND INTERPRETATION The overall results based on the analyses of all four types of Pr-sentences and the distributions of communicative dynamism vs. prosodic prominence within them were very positive, as may be seen in Table 3. The first column indicates the type of Pr-sentences, column 2 shows the absolute number of identified examples of each type, columns 3–4 refer to cases displaying perfect correspondence between the distribution
62 LINGUISTICA PRAGENSIA 1/2017 of CD and that of PP, while columns 5–6 relate to the number of cases with discrepancies between the CD and PP distributions. Type Abs. Perf. corresp. CD & PP Discrepancy CD vs. PP № of occur. %№ of occur. % I 86 76 88 10 12 II 74 65 88 9 12 III 20 18 90 2 10 IV 2 2 100 0 0 table 3.Distribution of CD vs. PP in Types I–IV All four types show very consistent results, which is rather remarkable given the difference in the absolute numbers of their occurrences. Abrief commentary on the findings for individual types follows, with Type II being dealt with separately in section 5.2. 5.1 TYPES I, III, IV Type Iwith the existential there-construction comes up most frequently, and individual examples range from short utterances to rather extensive ones, cf. examples (8) and (9): (8) There was apurse in it... [32] (9) Just before tea-time there came atremendous ring on the front-door bell... [14] It has to be noted, though, that in 41% of cases the IC is placed on afinal element, which is in line with the general end-weight principle and therefore represents adefault prosodic treatment, such as in example (10): (10) There was afine toothsome smell. [30] Arecurring deviation from the expected IC placement in this type is found in cases where the general tendency in structures with afinally placed verb would suggest that “afinal verb is usually deaccented, and the nucleus goes on apreceding noun” (Wells, 2011, 170). In our corpus, however, it is sometimes the verb that becomes the IC bearer, cf. sentence (11a) as opposed to what was expected (sentence (11b)). (11a) ...where there were no people LEFT. [24] (11b) ...where there were no people left. [24] Apossible explanation may be found in the so-called “reading stereotype” (Bolinger, 1986, 77); aphenomenon related to the fact that the end position is perceived (and preferred) as the most powerful and climatic for the nucleus placement, which can be particularly noticed when people read out loud. As Bolinger points out, “readers, including actors who mentally read their lines, often put the accent in the sup posedly
MARTIN ADAM— IRENA HEADLANDOVá KALISCHOVá 63 most effective place regardless of the sense.” (ibid.) Considering the amount of text the narrator of The Hobbit had to record, it is more than probable that occasionally there could be aslip of this kind. Type III (fronted adverbials followed with subject-verb inversion) demonstrates an even higher concordance rate between the analysed rhematic elements and assessed IC bearers than Type I(90% vs. 88%). However, it must be stated that in most of the utterances (78%) the rhematic element occupies the very last position (examples (12) and (13)), which is likely to have had ashare in the appropriate IC placement. (12) And here is our little Bilbo Baggins... [164] (13) … pop! went anasty smelly sack over his head… [170] Example (14) below shows one of the discrepancies identified between the distributions of communicative dynamism and prosodic prominence. The focalizer too is one of the list of function words which tend to attract nuclear stress (see e.g. Wells, 2011, 148), all the more so when it is placed finally, as is the case here. Cruttenden terms this phenomenon “lexical focusing”, i.e. certain words have “afixed relationship with nucleus placement” (1986, 73). (14) ...and out of it comes the dragon TOO... [163] The FSP analysis of the distribution of communicative dynamism over the sentence ele ments would assign the highest degree of CD to the communicative unit the dragon. Type IV, i.e. structures containing alocative subject, is evidenced by just two utterances, both showing perfect correspondence between the elements which are prosodically and communicatively the weightiest (see example (4) in section 3 and example (15) below). (15) It (=the hobbit hole) had aperfectly round door like aporthole... [181] 5.2 TYPE II Type II with an initial rhematic subject, the principal area under our examination, proportionally represents the second largest part of the corpus (40.7%), and as is the case with the other types, it also shows avery high percentage (88%) of perfectly corresponding distributions of CD and PP. The analysed material provides agreat range of utterances pertaining to this type of Pr-sentences in terms of their length as well as the position of the IC bearer within the communicative field. Even though the average length of these utterances (6.5 words) seems to tally rather closely with the general estimates of the length of tone units/intonation groups/intonational phrases (see e.g. Cruttenden, 1986, 72; Chamonikolasová, 2007, 73; or Gimson, 1994, 286) as well as that of basic distributional fields as viewed by FSP, the actual numbers of words in individual utterances vary from two to as many as twenty in one case. The following examples (16) and (17) may help to illustrate the extreme range.
64 LINGUISTICA PRAGENSIA 1/2017 (16) Whistles blew... [71] (17) Boulders, too, at times came galloping down the mountain-sides… [133] In terms of intonation centre placement, the IC bearers occupy either an initial3 position (65%) or they are placed medially (35%), see examples (18) and (19). (18) Abig jug of coffee had just been set in the hearth ... [94] (19) Suddenly in the wood beyond The Water aflame leapt up... [101] To digress slightly at this point, let us remark that an interesting occurrence applicable to all the Pr-sentences in the corpus has been observed: if the sentence contains one of the words Gandalf, sun or moon, identified as rheme proper, then this word becomes the IC bearer in the spoken version with unfailing regularity (see examples (20) and (21)). (20) Now Gandalf led the way. [125] (21) ...and awaning moon appeared above the hills between the flying rags. [114] Whether this observation may be ascribed to peculiarities of the auctorial style (narrative techniques, aparticular way of description or even fondness of acharacter) or perhaps genre constraints (such as reiterative appearance of some phenomena— Gandalf, moon or sun in Tolkien’s narrative), it does recur with identical periodicity and FSP interpretation throughout the text. In any case, it appears that various other texts also manifest analogous features; recently, for instance, research data examining fantasy fiction have revealed similar recurrences concerning the appearance/disappearance of RhPr elements such as cave, wind, moon, sun, rainbow, cloud, etc. (see Adam, 2013, 69, 120–121, 134). The presentation verbs operating in these sentences almost prototypically originate from the semantic category of verbs denoting natural or even supernatural phenomena that are typically unaffected by people, such as weather or the transcendental, and so, to acertain degree, show semantic affinity with the phenomena construed as rhematic subjects. As shown in Table 3 above, there are nine instances (12%) where the analyses of the spoken and written texts reveal differences in the identification of the bearers of the greatest prosodic prominence and the carriers of the highest degree of communicative dynamism (see examples (22)-(24) in section 6). The common denominator for all nine utterances in their spoken versions is arightward shift of the intonation centre, i.e. the narrator placed the IC on adifferent element than expected and this new IC is located closer to or at the very end of the utterance. The first possible explanation at hand was that the reason could be the length of the post-IC prosodic shade, i.e. the number of units in the communicative field coming after the IC bearer. According to Firbas, elements placed after the intonation centre find themselves in the “shade”, which makes their prosodic prominence lower 3 Please note that the initial position does not apply to the very first word only but refers to the first sentence element.
MARTIN ADAM— IRENA HEADLANDOVá KALISCHOVá 65 (1992, 152); phonologically speaking, in certain instances it is not the final nucleus but the preceding one that counts as the most prominent prosodic feature (Cruttenden, 1986, 84; Gimson, 1994, 242). It seemed only natural that aspeaker would find it rather difficult to maintain asomewhat subdued prosody on alarger stretch of text coming after the main peak and therefore might add extra emphasis to the final words; yet this theory had to be dismissed as unfounded. The corpus yielded evidence contrary to this presupposition, showing clearly that there was ahigh percentage among the utterances with corresponding CD and PP distributions where the post-IC prosodic shade stretched over as many as five words and more (see Table 4 below). Comparatively extensive parts of communicative fields following the IC bearers can also be found in examples (17), (18) and (21) above, each of which displays perfect correspondence between CD and PP. Length of post-IC prosodic shade % 1–2 words 41 3–5 words 38 6–12 words 21 table 4.Length of post-IC prosodic shade 6. SUMMARY AND CONCLUSIONS In the course of analysis and interpretation, while searching for existing analogies between the discrepancies so as to find some reasons for the narrator’s prosodic treatment, we realized what a“slippery phenomenon” intonation indeed was (Brazil, 1985, 240) and how difficult it was to assess it. In order to ascertain that our original FSP analysis (viewed by the non-prosodic factors) was correct we also worked with the Czech translation of The Hobbit (first published in 1978) by the acclaimed Czech translator František Vrba; it confirmed our assessment of the most dynamic rhematic ele ments in each of the nine instances. Eventually, it was possible to identify, albeit tentatively, three types of deviations and their potential causes. In the following examples the rheme-proper units are shown in bold while the newly assigned IC bearers are capitalized; the line below the example provides the Czech equivalent with the rheme-proper unit in bold and marked with F.V. (František Vrba) at the end, and finally aword-for-word translation back to English follows in square brackets. (I) ERRONEOUS RE-EVALUATING INTENSIFICATION (22) Supper is preparing over THERE... [130] Chystá se tam večeře. (F.V.) [is-preparing (refl.) there supper] In this example the prosodic intensification of the last word results in are-evaluation of the sentence perspective. The unit supper now appears to be acontextually dependent