The position and FSP role of -ly adverbials in small distributional fields

The position and FSP role of -ly adverbials in small distributional fields Vladislav Smolka (České Budějovice) ABSTRACT In the theory of FSP, context-independent adverbials are usually interpreted as constituting either the setting or specification, with the line between the two sometimes difficult to draw. Adverbials in -ly have been shown to be nearly always context-independent, functioning as aspecification and generally being more dynamic than the verb. At the same time, the CD of these adverbials is thought to be higher when they are placed after the verb than when they precede it. The present research explores sentence structures such as she smiled coldly, where the adverbial is the only complement of the verb in addition to anon-rhematic subject. Examples of these structures retrieved from the BNC show that the SVA sequence is considerably more frequent than SAV (she coldly smiled), while the proportion of the two variants is more balanced when there are other postverbal complements in addition to the adverbial, as in she coldly analysed his features/she analysed his features coldly. This paper aims to show that among the SVA structures in question, there are instances where the verb is actually more dynamic than the context-independent adverbial, and where linear modification is weakened as an indicator of FSP. These conclusions seem to be supported, among other clues, by translation into Czech. KEYWORDS -ly adverbials, FSP, position DOI https://doi.org/10.14712/18059635.2019.2.2 1. INTRODUCTION One of the factors co-determining the FSP status of aclause constituent is linearity. Compared with inflected languages with ahighly variable word order, the positional mobility of clause constituents is considerably constrained in English, because the position of agiven constituent within the sentence primarily indicates its syntactic function, rather than its communicative importance. Still, adverbials in English show ahigher degree of mobility than other constituents; this is especially true about adverbials that are not obligatory because their use is not prompted by the valency of the verb. The link between the position of adverbials in the sentence and their communicative importance has been extensively discussed in linguistic literature, but the discussion has been complicated by the fact that the category of adverbials is extremely diverse in several respects: in terms of their formal realisations, ranging from AdvP, NP, PP to verbless, nonfinite and finite clauses (Quirk et al. 1985: 489), their structural weight (length), and also in terms of the range of semantic functions (ibid. 479). It is therefore obvious that the respective subtypes of adverbials must be examined sepa- 148 LINGUISTICA PRAGENSIA 2/2019 rately, and it is why the present paper focuses solely on the subcategory of one-word –ly adverbials used within small distributional fields of communicative dynamism. 2. ADVERBIALS IN THE THEORY OF FSP In Firbasian terms, adverbials may perform one of three dynamic semantic functions in communication: aspecification, asetting, or atransition proper oriented element (Firbas 1992: 49, 77). In order to classify them as belonging to one of the abovementioned categories, two parameters must be considered: whether they represent context-dependent or context-independent elements, and whether they are obligatory or non-obligatory complements of the verb. According to Firbas, context-dependent adverbials invariably serve as settings, irrespective of their semantics and position in the sentence. On the other hand, obligatory context-independent adverbials serve as specifications of the verb, irrespective of their position, and are therefore more dynamic. The interpretation of non-obligatory context-independent adverbials depends on the kind of information they contribute: they perform the function of setting if they convey background, non-obligatory information; conversely, they are regarded as specifications if they convey essential information. The third category of adverbials, transition proper oriented elements, includes adverbials such as often, sometimes, naturally, of course, etc., that is, adverbials of indefinite time or sentence adverbials (Firbas 1992: 77–79). Unlike obligatory context-independent adverbials, which are interpreted as specifications regardless of their position, non-obligatory context-independent adverbials allow linear modification to assert itself as afactor determining their communicative dynamism: such adverbials tend to be interpreted as settings when they are initial, and as specifications when they are final or near-final. The difficulty in interpreting the latter type of adverbials lies in the fact that the assessment is made on semantic grounds, namely in considering whether the adverbial conveys information that is an essential or non-essential amplification of the semantic content of the verb. The difficulty of identifying adividing line in aphenomenon of scalar nature leads to frequent cases of potentiality. 3. SHORT –LY ADVERBIALS Within the category of adverbials, the –ly manner adverbials deserve particular attention because they show some specific features. They are nearly always contextindependent and therefore considered as more dynamic than the verb. As to the role of linear modification, Chládková (1979: 93) suggests that amanner adverbial carries alower degree of CD than anon-thematic element it precedes, but ahigher degree of CD than anon-thematic element that it follows; however, this does not apply to non-thematic subjects and to the verb. According to Chládková (ibid.), irrespective of sentence position, the manner adverbial carries alower degree of CD than anon- VLADISLAV SMOLKA 149 thematic subject, but ahigher degree of CD than the verb. She points out asimilarity between the relationship of an adverbial and the verb it complements and that holding between acontext-independent attribute and its head noun (ibid. 62), where the expression of quality (attribute) was found to be more dynamic than the noun by Svoboda (1968). On the other hand, referring to Bolinger’s pioneering work (Bolinger 1952), Firbas (1992: 53) points out that –ly adverbials are not totally unaffected by linear modification and tend to carry ahigher degree of CD when they occur after the verb than when they precede it. When these findings are confronted with examples of common sentence structures containing –ly adverbials, and, especially with minimal structures involving only the subject and the verb in addition to the adverbial, they often seem to run against empirical evidence. This is provided by prosodic prominence, the fourth factor of FSP operating in spoken language. In many cases, the intonation nucleus would naturally come not on the adverbial, but rather on the verb, as long as the subject was thematic. Similar clues are suggested by translation into an inflected language such as Czech, where an additional FSP factor, namely linear modification, ranks at the top of the non-prosodic factors. Anatural translation of most of the sentences in question would end up with the verb in the final position in the target language, suggesting its communicative importance. This is corroborated by the position of the intonation nucleus, which would also come on the final verb. 4. SCOPE OF RESEARCH For the sake of simplicity, the scope of research was restricted to short sentence structures constituting small distributional fields and consisting of only three constituents: the subject, the verb, and the –ly adverbial, because it is here that the competition between the adverbial and the verb is most straightforward. Such structures fall into two syntactic classes: verbs of copular nature followed by an obligatory adverbial, and intransitive verbs followed by an optional adverbial, which form the SVA and the SV(A) patterns, respectively. The former is of marginal interest because, as has been shown, the –ly adverbials are virtually always context-independent, and as they constitute an essential, compulsory complement of the intensive verb which cannot be omitted, they serve as aspecification and are inherently more dynamic than the verb. In the latter type, the adverbial is optional, which means that it can be left out without the sentence structure collapsing. They introduce into the sentence information that complements the semantic content of the verb and may vary in terms of relative importance. It can therefore be expected that it is in this sentence pattern that linear modification can assert itself most strongly as an indicator of CD. In addition to intransitive verbs, consideration must also be given to transitive verbs that are used without their usual complementing clause constituents, which have been left out and are only implicit.The very fact that they were left out suggests that they must be context-dependent and therefore carrying alow degree of CD. 150 LINGUISTICA PRAGENSIA 2/2019 An important role in the assessment of the FSP structure of the sentences explored is played by the subject. The present research focuses on instances where the subject is non-rhematic, typically context-dependent and, consequently, cannot be interpreted as anew phenomenon introduced onto the scene as the most dynamic element of apresentation sentence. Although the verbs included in this research are primarily verbs implementing the quality scale, with rhematic subjects the possibility of interpreting them as verbs of existence/appearance on the scene is not negligible. Instances of verbs that, under certain conditions, may imply presentation on the basis of semantic affinity with context-independent subjects are given by Firbas (1992: 60–61), and some more extreme examples are described by Adam (2014). The possible positions of an –ly adverbial in three-constituent sentences are restricted to three options demonstrated by the examples below. (1) Inwardly she smiled. (H97 618) (2) As she (sic) stared into her bemused eyes, he slowly smiled, and in that moment she would have committed murder if he’d asked her to. (H9V 2798) (3) She smiled proudly. (AC7 1244) In example (1) the adverbial is placed initially, before the subject, suggesting that its relationship to the verb, though not completely lost, is weakened, while the relationship to the subject is strengthened (Firbas 1992: 53). As aresult, the interpretation of the adverbial shifts away from specification towards the setting, and such interpretation is corroborated by the fact that initial adverbials are commonly separated from the sentence by acomma. The present paper focuses on examples (2) and (3), demonstrating the immediately preverbal and immediately postverbal (final) position, or SAV and SVA patterns, respectively. For the actual research, ten verbs were selected (agree, cry, inquire, laugh, nod, respond, sleep, smile, stare, and turn) and examples of their use along with –ly adverbials were retrieved from the British National Corpus (BNC), both from the written and spoken component. To exclude examples where the verb is complemented by other constituents, the search focused on instances where either the verb or the adverbial was the last component in the sentence. The search provided some interesting results. 5. SHORT –LY ADVERBIALS IN BNC Firstly, the SVA pattern invariably outnumbered the SAV pattern, in some cases significantly, as is clear from the following table. The disproportion between the preverbal and postverbal position would probably have been even more dramatic if all structural types of adverbials had been included in the research; the –ly adverbials constitute just afragment of the whole, and it is alogical assumption that adverbials in the form of, for instance, aPP, in spite of being much more frequent absolutely, are much less readily disposed to occur preverbally on account of their length and structural complexity. VLADISLAV SMOLKA 151 SAV SVA No. %No. % agree 118 45 142 55 cry 11 9 114 91 inquire 1 6 15 94 laugh 31 8 373 92 nod 20 5 403 95 respond 15 10 138 90 sleep 17 24 53 76 smile 44 5 833 95 stare 7 27 19 73 turn 11 21 41 79 Table 1.The proportion of preverbal and postverbal adverbials Secondly, although the respective verbs explored, expectably, tended to be complemented by different adverbials on the basis of semantic compatibility (cf. the semantic differences shown by, for example, agree— nod— smile— sleep— turn), the research proved that there was also adifference in the adverbials related to the position; the typical preverbal –ly adverbials were different from the typical postverbal adverbials. Classification of adverbials is anotoriously complex issue, but even so, the differences were sufficiently manifest. The SAV pattern characteristically included anarrower range of adverbials which occurred repeatedly and could be considered TrPro adverbs or, in Quirkian terms, disjuncts and subjuncts. The typical examples of this group (Type A) include actually, certainly, completely, entirely, finally, fully, hardly, merely, occasionally, only, really, totally, usually, etc. The prevalence of preverbal position is conspicuous, but probably cannot be explained solely by the fact that while most of them may occur both preverbally and postverbally, others (some downtoners and some focusing subjuncts) are restricted to medial position and, under normal circumstances, cannot occur finally (Quirk et al. 1985: 597–612). This is demonstrated by the following examples involving the adverbs entirely and merely, where only the former can occur in both positions. (4) Ientirely agree. (HHV 15474) (5) Iagree entirely. (F7A 451) (6) Travis merely laughed. (JY8 4476) Actually, contrary to what Quirk et al. say about positional restrictions of merely (ibid.) asearch in the BNC retrieved asingle final instance of this adverbial complementing, incidentally, one of the verbs explored, namely nod. The extreme scarcity of such astructure may however best be explained as aperformance anomaly. (7) ?He nodded merely. (ADY 272) Of those that fall in the category of adjuncts, the most commonly used adverbial was suddenly, which, according to Quirk et. al. (1985: 560), represents asemantic blend of 152 LINGUISTICA PRAGENSIA 2/2019 amanner and atime duration adverbial when used finally, whereas in the initial or medial position it is interpreted solely as atime adjunct. (8) He laughed suddenly. (CKF 1728) (9) Then he suddenly smiled. (J10 592) The SVA pattern was characterised by amuch broader range of different adverbials (Type B), mostly those that could be regarded as true manner adverbials or adverbials combining the expression of manner of action with adescription of the state or condition of the subject. They, too, occurred in the preverbal position, but significantly less frequently, suggesting that the final position is strongly preferred. (10) They all laughed heartily. (APW 3097) (11) The self-important politician smiled coldly. (AC2 2203) Other examples of adverbials of Type B expressing the manner of action include abruptly, austerely, briskly, feverishly, girlishly, gravely, quietly, loudly, obligingly, respectfully, slowly, softly, vigorously, and many others. Among those describing the state or condition of the subject are apologetically, cheerily, grimly, hysterically, nervously, politely, proudly, solemnly, thoughtfully, etc. It has to be pointed out that adistinct line between the two categories is hard to draw because there is aconsiderable amount of overlap between them. 6. DISCUSSION OF FINDINGS As has been shown above, not all adverbials retrieved from the corpus allow alternative placement in the sentence, and only those that do are relevant to the present research, because they open the way for linear modification to assert itself as an FSP factor. Admittedly, those typically occurring postverbally (Type B) are relatively rare in the preverbal position. Those that are restricted to asingle position are somewhat reminiscent of obligatory clause constituents; under special circumstances they may be fronted, but cannot migrate otherwise: (12) They went home. / *They home went. / Home they went. The essential question therefore is what difference, if any, there is between the communicative dynamism and the corresponding FSP function of an adverbial placed before the verb (in the NOT-position), and that placed after the verb (in the final position), as in the following examples (given here along with the minimal context in which they occurred in the corpus): (13) Luch considered, as she always did, and slowly nodded. (APW 3163) (Luch smiled, and Bridhe saw her. ‘Well, lassie! And are you happy that you’ve no need now to be traipsing away up the moor every noon-tide, eh?’ Luch con- VLADISLAV SMOLKA 153 sidered, as she always did, and slowly nodded. Bridhe patted her gently on the head.) (14) Wishart nodded slowly. (BMN 2451) (‘For all we know, your Lordship, it is possible that Queen Yolande was involved in her husband’s death. For her sake, for France’s sake, for Scotland’s sake, such suspicions must be cleared!’ Wishart nodded slowly. ‘Queen Yolande,’ he replied, ‘is to leave on tomorrow’s tide just after dawn.) If the meaning expressed by the verb + Adv structure is replaced by the verbonominal alternative, the difference in the position of the adverbial will be neutralised (Dušková 2017: 203–204, 2013: 36–37): (15) Luch considered, as she always did, and gave aslow nod. (16) Wishart gave aslow nod. According to Svoboda (1968), referred to by Dušková (2013: 37), in the verbonominal variant, the whole NP constitutes the rheme of the clause, and within the subfield constituted by the NP, the premodification exceeds the nominal head in the degree of CD. On the other hand, in the verbal construction the adverbial exceeds the verb in CD irrespective of the position, yet displays ahigher degree of CD when it follows the verb than when it precedes it. He also suggests that the relationship between the premodifying adjective and the NP is similar to that holding between the verb and amanner adjunct (Svoboda 1989: 131–132), in that the latter may also be treated as asingle unit, unless the adverbial is intensified. Still, this does not seem to solve the problem of prosodic prominence and translation into Czech. In natural oral presentation of all the examples presented by (13–16), the expected position of the intonation nucleus is on nod, no matter whether it is averb or anoun, and in semantically equivalent translation into Czech, all of the variants would employ alinear arrangement where the manner of the action precedes the action itself, and prosodic prominence would be given to the verb. It would require acontext where both the subject and the verb were context dependent for the manner adjunct to occur in the final position. However, following the line of thought suggested by Svoboda (1989), if the manner adverbial itself is intensified or modified, its CD increases accordingly, and the final position will most likely be associated with prosodic prominence, and, consequently, rhematicity. (17) He looked at the Lad and Taliesin thought the Lad nodded very slightly. (G10 282) This may be due not solely to the semantic intensification, but at least partly also to the end-weight principle, which favours, where possible, alinear arrangement of clause constituents from shorter to longer ones. It can be assumed that a similar effect may be achieved by coordination: For example, if you smell burning in the middle 154 LINGUISTICA PRAGENSIA 2/2019 of the night you will respond immediately and fully, and not bother to ‘consult’ any body clock to see if you should calm down and be asleep! (A75 750 ); or by employing astructurally heavier formal realisation of the adverbial, e.g. aPP: Sheila nodded in vigorous agreement. (A6N 2397). However, these structural options were not explored in the present research. In this respect, unmodified –ly adverbials are light elements, and their final placement is due to factors other than weight, for instance, to the default grammatical sequence of clause constituents. Another factor to consider is the fact that –ly adverbials typically occur as optional, non-obligatory constituents. This may prompt reasoning along two contradictory lines. Following the first, it may be reasoned that the possibility of leaving the adverbial out affects neither the essential meaning of the sentence, nor its grammatical completeness, as demonstrated by the following examples. Conversely, leaving out the verb results in the sentence structure breaking down. (18) Both my master and Iturned quickly. (HU0 652) (19) Both my master and Iturned. (20) *Both my master and Iquickly. Combined with intransitive verbs, which do not require any other clause constituent apart from the subject, -ly adverbials are rendered even less communicatively import - ant. And, conversely, the intrinsically high semantic salience of intransitive verbs increases even more when their subjects are context-dependent and therefore thematic, excluding the possibility of presentation scale implementation. On the other hand, it might be argued that the very fact that the adverbials are included in the sentence although this is, strictly speaking, unnecessary, suggests that they carry important information; this is probably why they are regarded as specifications in the theory of FSP. However, adistinction should be made between the A-type adverbials, that is, those that were commonly found in the preverbal position in BNC (e.g. completely, entirely, finally, fully, hardly, merely, etc., cf. Section 5.above) and the B-type ones, which were more frequent in the postverbal position (quietly, loudly, obligingly, cheerily, grimly, hysterically, etc.), in other words, true manner adverbials, which constitute amuch more open set. If the communicative importance of an adverbial is assumed to be inversely proportionate to its predictability, then it is clearly adverbials of the latter type that are less predictable and therefore inherently more dynamic; adverbials of the former type typically function as intensifiers, expressing different degrees on ascale of intensity, and although the choice of aparticular adverbial is not predictable, the range of options is limited and the kind of information they convey brings them close to the category of TMEs. Let us now look at FSP analysis of the sentence structures explored in this paper, starting with the simplest possible variant, consisting just of Sand V, and proceeding to the other linear arrangements. (21) He smiled. (A0F 619) (22) He merely smiled. (JXS 1026) VLADISLAV SMOLKA 155 (23) As she (sic) stared into her bemused eyes, he slowly smiled, and in that moment she would have committed murder if he’d asked her to. (H9V 2798) (24) He smiled shyly. (A7A 2400) In (21), as well as in all other examples above, the context-dependent Sserves as the theme. Since there are no postverbal complements, the Vcannot perform the function of mediating between the thematic and non-thematic part of the sentence as it typically does in atripartite FSP segmentation; instead, the categorial component of Vis the TrPr, and the notional component of Vis RhPr. In Firbasian interpretation of example (22) the Vconstitutes aTr, but it remains aquestion whether the adverb merely should be interpreted as Sp, although one carrying arelatively low degree of CD, or whether it would be regarded as aTrPro element, on apar with adverbials like naturally, usually, etc. (Firbas 1992: 78). It is worth noting that merely belongs to adverbials that practically invariably occur in the preverbal position (Type A). In (23) the verb is regarded as the Tr, and the adverbial, belonging to Group B, serves as Sp, exceeding the Vin terms of CD and therefore rhematic. The same is true of (24), with the slight difference that the final Ais more dynamic than the preverbal Ain (23). However, it is hard to identify how this difference in the degrees of CD is reflected in prosody or in translation into Czech. Svoboda’s observation (Svoboda 1989: 131–132) that, unless the adverbial is intensified, the Vand the manner adjunct may be treated as asingle unit (although it is not quite clear whether this is meant strictly as asubfield of CD distribution) opens the way for an alternative interpretation in which the adverbials in (23) and (24) would be considered as Tr-oriented elements, however, ones oriented to the notional component of the V, rather than to TrPr. In such acase, the notional component of the verb would carry the highest degree of CD and be rhematic, which would be in agreement with the indication provided by prosody and by translation into Czech. At the same time there would still be room for the difference in the degrees of CD carried by the preverbal and postverbal adverbial. Of course, this does not mean that the unintensified/unmodified -ly adverbial can never perform the rhematic function in the structures explored; however, this must be licensed by special circumstances in which the contextual factor is of key importance; the verb must be context dependent, or the context dependence must be sufficiently strongly implied, as shown in the following hypothetical examples, accompanied by probable translations into Czech; the likely position of the intonation nucleus is given in bold script. (25) He called her name and she turned. (Zavolal ji jménem aona se otočila.) (26) He called her name and she quickly turned. (Zavolal ji jménem aona se rychle otočila.) (27) He called her name and she turned quickly. (Zavolal ji jménem aona se rychle otočila.) (28) He called her name and she a) turned b) quickly, as if to prove her alertness. (Zavolal ji jménem aona se rychle a) otočila / otočila b) rychle, jako by chtěla prokázat svou bdělost.)