CC BY
5
ЯЗЫКОЗНАНИЕ
UDC 811.113.6 Damian Herda
Jagiellonian University in Krakow, Poland
FROM NOMINAL QUANTIFIERS TO ADVERBIAL MODIFIERS: A CORPUS INVESTIGATION WITH REFERENCE TO SWEDISH*
For citation: Herda D. From nominal quantifiers to adverbial modifiers: A corpus investigation with reference to Swedish. Scandinavian Philology, 2023, vol. 21, issue 2, pp. 199-217. https://doi.org/10.21638/11701/spbu21.2023.201
As has been noted in grammaticalization literature, partitives, i. e. nominals encoding parts, portions, and sets, tend to evolve into vague quantifiers, and then into degree adverbs. However, while this grammaticalization path has already received a fair amount of scholarly attention in English, the extension of such items beyond the nominal domain in Swedish remains an empirically unexplored territory. Thus, based on random samples of attestations extracted from selected Sprakbanken corpora, this paper offers a synchronic glimpse into the syntactic expansion of nine Swedish nominal quantifiers, namely droppe 'drop', nypa 'pinch', smula 'crumb', hop 'heap', hopar 'heaps', hog 'pile, hogar 'piles, massa 'mass, and massor 'masses'. The Swedish results largely coincide with those obtained for English, and demonstrate that in the verbal domain, most of the scrutinized elements reveal a preference for pronominal uses, in which they function as an argument of the verb rather than a genuine degree adverb, but which nonetheless give rise to secondary scalar inferences, whereas in the adjectival domain, a majority of the items exhibit a propensity to combine with the comparative forms of adjectives/adverbs. Both of these environments may therefore be assumed to constitute bridging contexts in the emergence of full-blown degree modifier uses of grammaticali-zed partitives. It is further shown that there exists a strong positive correlation between the items' respective degrees of grammaticalization in the quantifier function and their
* This research was funded by the National Science Center of Poland (Grant no. 2019/33/N/HS2/01695).
extents of adverbialization, which testifies to the importance of frequency of use in the scrutinized instance of grammaticalization.
Keywords: grammaticalization, adverbialization, nominal quantifiers, degree adverbs, Swedish.
1. INTRODUCTION
In grammaticalization theory, it is posited that partitive nouns, i. e. elements designating parts, portions, or sets of what their nominal collocates refer to, tend to develop into vague quantifiers, and then also into degree adverbs (cf. [Traugott, 2008]). Although this phenomenon, and especially the former phase thereof, has already been investigated quite extensively with reference to English (cf. [Herda, 2022a, b; Brems, 2003; 2007; 2011; 2015; De Clerck, Brems, 2016]), the expansion of such items beyond the nominal domain in Swedish remains an empirically unexplored territory (see [Herda, 2018]).
In view of the above, drawing on data derived from selected Sprak-banken corpora, this paper seeks to offer a synchronic glimpse into the syntactic extension of nine Swedish nominal quantifiers, namely droppe 'drop', nypa 'pinch', smula 'crumb', hop 'heap', hopar 'heaps', hog 'pile', hogar 'piles', massa 'mass', and massor 'masses'1, in an attempt to identify possible bridging contexts in the emergence of their adverbial uses in both the verbal and the adjectival domains. To this end, the obtained tokens for each item have been classified into six categories: (i) verbal inherent modification (e. g. alska massor 'love a lot'), (ii) verbal extent modification (e. g. vanta en smula 'wait a bit'), (iii) adverbial ambiguous (e. g. le ensmula 'smile a bit'), (iv) object-pronominal (e. g. kosta en massa 'cost a lot'), (v) adjectival modification of positives (e. g. en nypa trott 'a bit tired'), and (vi) adjectival modification of comparatives (e. g. en nypa lattare 'a bit easier'). Moreover, the degrees of adverbialization of the analyzed items, construed as proportionate frequencies of their adverbial uses, have been compared with their respective percentages of quantifier attestations, as determined by Herda [forthcoming], in order to establish whether these two grammaticalization parameters are positively correlated.
1 As can be seen, the plural variants of items invoking large quantities are analyzed separately from the singular forms, which is in keeping with the observation that the two may display distinct degrees of grammaticalization (cf. [Brems, 2003]).
The remainder of the paper is structured as follows. Section 2 outlines general facts related to the grammaticalization of partitive nouns into quantifiers and adverbs. Section 3 provides an account of the method used in the empirical investigation, while Section 4 presents the obtained results. Finally, Section 5 summarizes the main conclusions reached in the study and suggests prospects for future research on the topic.
2. PREVIOUS RESEARCH
As stated above, partitives nouns, whose primary function consists in "bounding or unitizing the entities expressed by the second constituent" [Verveckken, 2015, p. 48], and which incorporate "conception of [their] typical size" [Langacker, 1991, p. 88], display a tendency to gram-maticalize into vague quantifiers [Brems, 2003; 2007; 2011; De Clerck, Brems, 2016; Traugott, 2008], a development enabled by "the semanti-cization of quantifier meaning through repeated pragmatic inferencing of size or scalar implications" [Brems, 2011, p. 108]. It is further possible to distinguish between multal and paucal vague quantifiers, of which the former indicate a non-specific high quantity, and the latter point to a non-specific low quantity of what the concomitant nominal stands for [Huddleston, Pullum, 2002, p. 365, 366].
The aforementioned semantic generalization of partitives entails a number of distributional changes, among which are (i) collocational broadening (cf. [Brems, 2011, p. 103, 104, 105]), i. e. extension to, e. g., abstract nouns, (ii) loss of compatibility with other quantifiers (cf. [Kei-zer, 2007, p. 136]), (iii) restrictions in modification patterns, i. e. the possibility for such items to be pre-modified only by quantification-reinforcing adjectives [Brems, 2011, p. 201], e. g. hel 'whole' in en hel massa tid 'a whole lot (lit. 'mass') of time', and (iv) occasional omission of the indefinite article [Kinn, 2017], as in (1).
(1) Att ligga pa kokssoffan hos mormor vacker massa gamla minnen! [Sprakbanken: Twitter, 2015]
'Lying on the kitchen sofa at grandma's house evokes a lot (lit. 'mass') of old memories!'
Additionally, paucal nominal quantifiers are no longer susceptible to pluralization, whereas multal ones can in fact be pluralized, yet plural morphology serves an intensifying rather than unitizing function here,
i. e. it further strengthens the intrinsic scalar implications of a particular item of this kind (cf. [Brems, 2011, p. 203; Herda, 2019b]).
Crucially, the subsequent phase in the functional evolution of partitives nouns manifests itself in their syntactic expansion beyond the nominal domain (cf. [Doetjes, 1997; Traugott, 2008; Brems, 2011; De Clerck, Brems, 2016]). As De Clerck and Brems [De Clerck, Brems, 2016, p. 168] underline, the reason why the adverbialization of nominal quantifiers deserves special attention resides in the fact that "extension to increasingly more syntactic contexts is considered to be a symptom of further grammaticalization since it presupposes that reanalysis of the expression concerned has taken place." More specifically, adverbialized quantifiers may function as either inherent degree modifiers, combining with gradable adjectives/adverbs as well as verbal predicates encoding degree scales, e. g. psychological verbs, or extent modifiers, in which case they point to the temporal extension of the situations denoted by the accompanying eventive verbs (cf. [Bolinger, 1972]). However, the existence of verbs permitting both the degree and the extent interpretation, such as le 'to smile' or regna 'to rain', occasionally leads to ambiguities [Quirk et al., 1985, p. 603]. Even so, it is possible for both degree and extent modifiers to be subdivided in accordance with the type of quantifier they derive from. Thus, adverbs deriving from multal quantifiers are typically referred to as boosters, while those related to paucal quantifiers are traditionally labelled as diminishers or minimizers, the latter name being peculiar to adverbs used in non-assertive contexts (cf. [Quirk et al., 1985, p. 589; Claridge, Kyto, 2014, p. 29, 30]).
However, adverbial uses of nominal quantifiers ought to be distinguished from pronominal attestations, i. e. those in which the quantifier syntactically functions as the direct object in a sentence rather than a genuine degree modifier, as in (2)-(3). In such cases, the semantics of the verb implies a specific complement (cf. (2)), or it is possible to assume a highly general complement like saker 'things' (cf. (3))2.
(2) Kande mig bakis, aven om jag inte druckit en droppe [Sprakbanken: Bloggmix, 2011].
'I felt hungover, even though I hadn't drunk a drop.'
2 Pronominal occurrences, in turn, should be differentiated from clearly elliptical quantifier attestations, where the omitted quantified nominal can be easily retrieved from the immediate context of the utterance.
(3) Jag larde mig massor [Sprakbanken: Twitter, 2016].
'I learnt lots (lit. 'masses').'
As for the mechanisms underlying the expansion of grammaticali-zed partitives beyond the nominal domain, three scenarios have been suggested in the existing literature. According to one of them, the initial stage of the adverbialization of nominal quantifiers involves extent rather than inherent degree modification, as the former phenomenon is conceptually closer to vague quantification by dint of indicating numbers, i. e. frequencies, or amounts, i. e. durations (cf. [Claridge, Kyto, 2014; Herda, 2019a]). Following De Clerck and Brems [De Clerck, Brems, 2016], it is also pronominal occurrences that may facilitate the transition of vague quantifiers to adverbial settings by virtue of generating secondary degree/extent inferences, and thus attestations of this kind will likewise be taken into account in the present study. As for the adjectival domain, it has been proposed that the early adverbial uses of nominal quantifiers tend to involve comparatives rather than positives, especially forms analogous to Swedish mer 'more' and fler 'more (only for countable nouns)' (cf. [Norde, De Clerck, Colleman, 2014]), as in en massa mer/fler 'a lot (lit. 'mass') more', yet the possibility of a grammati-calized partitive functioning as a degree modifier still "calls for empirical investigations across larger data sets" [Norde, De Clerck, Colleman, 2014, p. 246].
In what follows, the above-discussed hypotheses are confronted with empirical Swedish data. Section 4 describes the method employed in the inquiry, whereas Section 5 reveals the obtained results, which are then summarized in Section 5.
3. METHOD
As already stated in the Introduction, the current investigation includes nine Swedish items, namely droppe 'drop', nypa 'pinch', smula 'crumb', hop 'heap', hopar 'heaps', hog 'pile', hogar 'piles', massa 'mass', and massor 'masses'. Since the extended uses of nominal quantifiers have been observed to occur predominantly in informal, web-based data (cf. [Norde, De Clerck, Colleman, 2014; De Clerck, Brems 2016]), the empirical material for this study was collected from the following Sprakbanken corpora: (i) Bloggmix 2011, (ii) Bloggmix 2012, (iii) Bloggmix 2013, (iv) Bloggmix 2014, (v) Bloggmix 2015, (vi) Bloggmix 2016, (vii) Blogg-
mix 2017, (viii) Familjeliv: Fritid och Hobby, (ix) Flashback: Livsstil, (x) Twitter 2015, (xi) Twitter 2016, and (xii) Twitter 2017. The entire dataset further encompasses slightly more than 2 billion tokens, which makes it both qualitatively and quantitatively close to that used in an analogous investigation conducted on English data (cf. [Herda, 2022b]).
The specific aim of the searches has been to obtain a random sample consisting of up to 250 relevant attestations per item (excluding occasional doublets), of which a maximum of 125 include instances representative of the verbal domain, and a maximum of 125 involve instances representative of the adjectival domain, so as to preclude the possibility of claiming a particular element to have undergone a substantial level of adverbialization if it predominantly occurs in one domain only. As mentioned before, attestations belonging to the verbal domain have been grouped into (i) adverbial degree modifier (DM), (ii) adverbial extent modifier (EM), (iii) adverbial ambiguous, i. e. allowing both the degree and extent modifier reading (AMB), and (iv) pronominal (PRON). Occurrences representative of the adjectival domain, in turn, have been categorized into those involving (i) positive (POS) and (ii) comparative (COM) forms of adjectives/adverbs, the latter also including uses with the inherently comparative adverb till 'more; additionally' (e. g. en massa till 'much (lit. 'mass') more') as well as the excess operator for 'too'.
Importantly, the current degree of each item's adverbialization is taken to be reflected in the number of all of its adverbial attestations in the sample divided by 250 (i. e. the highest possible number of tokens per item). Moreover, the adverbialization values (ADV) were juxtaposed with those related to the items' respective degrees of grammaticalization in the quantifier function (QUANT), operationalized as the proportionate frequency of an item's quantifier occurrences in a random sample, as reported on in Herda [forthcoming], in order to ascertain whether there exists a positive correlation between these two grammaticalization parameters3.
4. RESULTS
Table 1 shows the distribution of the scrutinized Swedish items in the verbal and the adjectival domains.
3 Notably, all of the obtained values have been rounded up to two decimal places.
Table 1. Adverbialization patterns of the analyzed Swedish items
Item Uses Total
Verbal domain Adjectival domain
DM No. (%) EM No. (%) AMB No. (%) PRON No. (%) POS No. (%) COM No. (%)
droppe 0 (0) 0 (0) 23 (14.02) 102 (62.20) 3 (1.83) 36 (21.95) 164 (100)
nypa 2 (12.50) 0 (0) 0 (0) 0 (0) 8 (50) 6 (37.50) 16 (100)
smula 107 (42.80) 6 (2.40) 7 (2.80) 5 (2) 98 (39.20) 27 (10.80) 250 (100)
hop 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
hopar 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
hög 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
högar 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
massa 16 (6.40) 17 (6.80) 10 (4) 82 (32.80) 3 (1.20) 122 (48.80) 250 (100)
massor 31 (12.40) 18 (7.20) 8 (3.20) 68 (27.20) 0 (0) 125 (50) 250 (100)
Total 156 (16.77) 41 (4.41) 48 (5.16) 259 (27.85) 112 (12.04) 316 (33.98) 930 (100)
As can be noted, smula 'crumb' exhibits the highest degree of adverbialization among the analyzed Swedish elements (245/250, i. e. 98 %), followed by massor 'masses' (182/250, i. e. 72.80 %) and massa 'mass' (168/250, i. e. 67.20 %), except that the difference between the latter two items in the proportion of their adverbial uses is not statistically signifi-
cant (p > .05). The items droppe 'drop' and nypa 'pinch', in turn, have undergone a limited level of adverbialization, the relevant values standing at 24.80 % (62/250) and 6.40 % (16/250), respectively. However, neither hop 'heap', hopar 'heaps' nor hög 'pile', högar 'piles' have been found to have developed any adverbial uses.
In the verbal domain, there is a predominance of pronominal uses, which suggests that pronominalization can indeed be seen as an intermediate phase between vague quantification in the nominal domain and adverbial degree modification. Nevertheless, the data do not provide any strong support for the importance of attestations involving eventive predicates in the expansion of vague quantifiers beyond the nominal domain. Even so, in the adjectival domain, uses involving comparative variants of adjectives/adverbs substantially outnumber those with positive forms, which, in turn, accords with the hypothesis assuming a facilitating role of such attestations in the change under scrutiny.
Importantly, the validity of the above findings is further strengthened by the fact that a methodologically analogous study concerned with the adverbialization of English nominal quantifiers likewise testifies to the significance of pronominal occurrences and those involving comparatives (cf. [Herda, 2022b]). Another noteworthy commonality between English and Swedish is that in both languages, the adverbialization frequency values of particular items are positively correlated with those pertaining to their respective degrees of grammaticalization in the quantifier function.
Table 2. The analyzed Swedish items as adverbs and quantifiers
NQ droppe nypa smula hop hopar hög högar massa massor
ADV 0.25 0.06 0.98 0 0 0 0 0.67 0.73
QUANT 0.32 0.37 0.89 0.54 0.01 0.25 0.19 0.99 1
Indeed, the values presented in Table 2 reveal a strong positive correlation between the Swedish items' levels of grammaticalization as adverbs and quantifiers (R = .88, p < .05). Compounded by the diachronic facts described in Traugott (2008) and Herda (2018), this observation points to a tendency for a high percentage of an item's quantifier uses to facilitate the emergence of additional adverbial occurrences.
In the subsequent parts of the paper, the items droppe 'drop', nypa 'pinch', smula 'crumb', massa 'mass', and massor 'masses' are dealt with separately through the prism of their adverbial attestations. In each case, a number of representative corpus examples are provided.
4.1. Droppe
Out of the 16,189 occurrences of the item droppe 'drop' in the analyzed material, only 164 instances have been recognized as relevant to the present study. Most of them, however, are pronominal attestations, in which droppe 'drop' combines with the verb dricka 'to drink, as already illustrated with (2).
Interestingly, even in its adverbial uses in the verbal domain, the element under discussion exhibits collocational constraints, in that it only modifies liquid-related verbs, such as regna 'to rain' (cf. (4)), svet-tas 'to sweat' (cf. (5)), and gräta 'to cry' (cf. (6)). Since such verbs allow both a degree and an extent reading, uses of this kind have been classified as adverbial ambiguous4. Another noteworthy fact is that similarly to its pronominal uses, droppe 'drop' modifying verbs functions as a negative polarity item, i. e. a minimizer, appearing predominantly in the context of negation, which can be explained with reference to a literal drop being commonly perceived as a minimal portion of a liquid substance.
(4) Är sä himla nöjd över att det inte regnade en droppe igär [Sprakbanken: Bloggmix, 2012].
'I'm so damn happy that it didn't rain at all (lit. 'drop') yesterday.'
(5) Räcker länge och du svettas inte en droppe [Sprakbanken: Flashback: Livsstil].
'It suffices for a long time and you don't sweat at all (lit. 'drop').'
(6) Har inte gratit en droppe [Sprakbanken: Twitter, 2015]. 'I haven't cried at all (lit. 'drop').'
In the adjectival domain, the adverbially used item droppe 'drop' most frequently modifies comparatives (cf. (7)-(9)), including the
4 Theoretically, it is also possible to treat verbs like regna 'to rain' as encoding an implicit liquid argument, in which case uses such as (4)-(6) could be classified as pronominal.
excess operator for 'too' (cf. (10)) and the inherently comparative form till 'more' (cf. (11)).
(7) Nar man fatt nog och aldrig vill dricka en droppe mer [Sprakbanken: Twitter, 2015].
'When you've had enough and don't ever want to drink even a bit (lit. 'drop') more.'
(8) Jag tycker dom fortjanar det finaste och basta som finns ju, inte en droppe mindre an det [Sprakbanken: Bloggmix, 2012].
'I think they deserve the finest and the best stuff that exists, not even a bit (lit. 'drop') less than that.'
(9) Nu en droppe mognare, bittrare och drygare... [Sprakbanken: Twitter, 2015]
'Now a bit more mature, more bitter, and tougher.'
(10) Detta var en droppe for mkt... [Sprakbanken: Twitter, 2017] 'That was a bit (lit. 'a drop') too much.'
(11) Gar barca vidare kommer jag ga ratt hem utan att dricka en droppe till [Sprakbanken: Twitter, 2017].
'If Barca goes further, I'll go home straight away without drinking even a bit (lit. 'drop') more.'
The adverbial attestations of droppe 'drop' involving positive forms of adjectives/adverbs, on the other hand, seem rather incidental due to their scarcity and the impossibility of identifying any semantic tendencies pertaining to the modified elements. In the scrutinized material, the degree modifier droppe 'drop' combines only with three relevant forms, namely motstroms 'against the current' (cf. (12))5, mork 'dark' (cf. (13)), and trakig 'boring/sad' (cf. (14)).
(12) Sa nu vet ni det ;-) En droppe motstroms. Hade velat skriva mer, men ovanstaende var redan over maxgransen [Sprakbanken: Bloggmix, 2015].
5 In truth, droppe 'drop' in (12) may also be interpreted more or less literally, i. e. as a drop which is against the current. However, since the limited corpus context does not permit disambiguation, the discussed example has been classified as a potential adverbial attestation. Similarly, (14) can potentially be seen as a quantifier use in which trakigt is a substantivized adjective ('boring/sad stuff'). Yet, again, given that no analogous uses have been recorded in the sample comprising the analyzed item's clearly adnominal occurrences, (14) has been categorized here as an adverbial use.
'So now you know it ;-) A bit (lit. 'drop') against the current... I'd like to have written more, but the above already exceeded the maximal boundary.'
(13) Hur ar det mojligt att en svart man och en vit kvinna far barn och barnet blir inte en droppe mork? [Sprakbanken: Twitter, 2016] 'How is it possible that a black man and a white woman have a child and the child is not even a bit (lit. 'drop') dark?'
(14) Och en droppe trakigt [Sprakbanken: Twitter, 2017]. 'And a bit (lit. 'drop') boring/sad.'
4.2. Nypa
Out of the 10,904 attestations of the noun nypa 'pinch' in the analyzed data, only two adverbial attestations in the verbal domain can be found: one involving the verb dofta 'to smell' (cf. (15)), which encodes a physical sensation, and the other involving the epistemic predicate over-driva 'to exaggerate' (cf. (16)). Yet, given the scarcity of the relevant uses, it is impossible to identify any semantic tendencies here.
(15) Daremot doftar det en nypa njure <...> [Sprakbanken: Bloggmix, 2012].
'However, it smells a bit (lit. 'pinch') of kidney.'
(16) Nej jag kanske overdrev en nypa <...> [Sprakbanken: Twitter, 2016]. 'No, maybe I exaggerated a bit (lit. 'pinch').'
As for the adjectival domain, on the other hand, the scrutinized material reveals 14 adverbial uses of the item under discussion. Notably, most of the positive forms of adjectives/adverbs modified by nypa 'pinch' are negatively colored (cf. (17)-(18)), which accords with the cross-linguistic tendency for diminishers to combine with neutrally and negatively, rather than positively, loaded elements. However, this apparent restriction does not seem to apply to comparatives (cf. (19)-(20)).
(17) Ar bara en nypa trott pa mig sjalv [Sprakbanken: Twitter, 2016]. 'I'm just a bit (lit. 'pinch') tired of myself.'
(18) En nypa osmakligt av ahlens :( [Sprakbanken: Twitter, 2017]. 'That's a bit (lit. 'pinch') distasteful on Ahlens's side :('
(19) Jag ar nojd, livet en nypa lattare och barnen vet vilka fingervantar av alla dessa rosa magiska vantar (...) ar just deras [Sprakbanken: Blogg-mix, 2013].
'I'm happy, life's a bit (lit. 'pinch') easier, and the children know which of all those pink magic gloves are theirs.'
(20) Just nar man tagit dem en nypa for mycket for givna... [Sprakbanken: Twitter, 2017]
'Just when you've taken them a bit (lit. 'pinch') too much for granted..'
4.3. Smula
Out of the analyzed items, smula 'crumb' is the only one which has undergone conventionalization in the adverbial function, its earliest pertinent attestations dating back to the 18th century (cf. [SAOB]). It should therefore come as no surprise that, similarly to English bit, Swedish smula 'crumb' exhibits a high degree of collocational openness. Another commonality between the two items is their notable preference for inherent degree modification (cf. [Herda, 2022b]). Moreover, the degree modifier smula 'crumb' reveals a predilection for change-of-state (cf. (21)-(22)) and psychological verbs (cf. (23)-
(24)).
(21) Efter tva langa timmar begav sig vara besokare ivag och Z lugnade ner sig en smula [Sprakbanken: Bloggmix, 2011].
'After two long hours, our visitors went away and Z calmed down a bit (lit. 'crumb').'
(22) Frekvensen har gatt upp en smula. [Sprakbanken: Twitter, 2017] 'The frequency has gone up a bit (lit. 'crumb').'
(23) Harlig stad, men langtar hem en smula nu [Sprakbanken: Twitter, 2015].
'Wonderful city, but I miss home a bit (lit. 'crumb') now.'
(24) Jag angrar mig en smula [Sprakbanken: Twitter, 2015]. 'I repent a bit (lit. 'crumb').'
When occurring in the adjectival domain, the adverbially used item smula 'crumb', again like English bit (cf. [Herda, 2022b]), combines predominantly with the positive forms of adjectives/adverbs. As is the case with nypa 'pinch, a large proportion of the positive forms of adjecti-
ves/adverbs modified by smula 'crumb' further display a negative coloring (cf. (25)-(26)), except that the latter is considerably more productive in the adverbial function (cf. Table 1). This tendency cannot, however, be extended to the modified comparatives (cf. (27)-(28)).
(25) Det är en smula löjeväckande och ytterst fânigt [Sprâkbanken: Flashback: Livsstil].
'That's a bit (lit. 'crumb') ridiculous and extremely silly.'
(26) Tillât mig säga att jag är bâde överraskad och en smula skeptisk [Sprâkbanken: Twitter, 2015].
'Let me say that I'm both surprised and a bit (lit. 'crumb') skeptical.'
(27) Men jag tror att jag faktiskt blivit bâde en smula försiktigare OCH klokare [Sprâkbanken: Twitter, 2017].
'But I think that I've actually become both a bit (lit. 'crumb') more careful and wiser.'
(28) Därför tog vi ungefär samma runda idag och lyckades faktiskt en smula bättre [Sprâkbanken: Bloggmix, 2012].
'So we went for approximately the same round today, and we actually did a bit (lit. 'crumb') better.'
4.4. Massa
While smula 'crumb' reveals a clear preference for inherent degree modifier uses over extent modifier ones, there is no statistically significant difference in the frequency of the two types of attestations in the case of massa 'mass' (cf. (29)-(32)). In its degree modifier occurrences, massa 'mass' further tends to combine with psychological verbs, especially sakna 'miss' (cf. (29)), as well as change-of-state predicates (cf. (30)).
(29) jag saknar dig massa! [Sprâkbanken: Twitter, 2015] 'I miss you a lot (lit. 'mass')!'
(30) Sen växte han en massa under sommaren [Sprâkbanken: Bloggmix, 2011].
'Then he grew a lot (lit. 'mass') during the summer.'
(31) Den här kommer jag använda massa [Sprâkbanken: Bloggmix, 2011]. 'This one I'm going to use a lot (lit. 'mass').'
(32) Min telefon krânglar en massa och jag har tänkt skaffa ny länge [...] [Sprâkbanken: Bloggmix, 2012].
'My phone breaks down a lot (lit. 'mass') and I've long considered buying a new one.'
In the adjectival domain, two out of the three adverbial uses of massa 'mass' with positive forms of adjectives/adverbs (cf. (33)-(35)) involve items denoting psychological (cf. (33) or physiological sensations (cf. (34)). As in the case of droppe 'drop', however, the paucity of the revenant uses makes it impossible to speak of any strong tendencies here.
(33) Dagen har varit massa mysig och varm [Sprakbanken: Bloggmix, 2011].
'The day has been very (lit. 'mass') nice and warm.'
(34) Häromdagen när jag lag i sängen och var trött (+ massa sugen pa godis) sa hörde jag glassbilen utanför [Sprakbanken: Bloggmix, 2011].
'The other day, when I was lying in bed tired (+ very (lit. 'mass') hungry for sweets), I heard an ice cream van outside.'
(35) Jag hade glömt att NK startade sin rea igar, men jag sprang förbi Ord-ning&Reda som egentligen är massa dyrt [Sprakbanken: Bloggmix, 2011].
'I'd forgotten that NK started its sales yesterday, but I ran past Ord-ning & Reda that is actually very (lit. 'mass') expensive.'
The remaining attestations of massa 'mass' representative of the adjectival domain involve comparatives. Although relatively frequent, such uses are predominantly made up of just two items, namely the comparative forms of the quantifiers mycket 'much' and mänga 'many', i. e. mer 'more' (cf. (36)) and fler 'more' (cf. (37)), respectively. In the data under scrutiny, there is only one attestation involving a form other than mer 'more' or fler 'more', namely the inherently comparative item till 'more; additionally' (cf. (38)).
(36) Haha jag kommer säkert pa en massa mer saker som jag borde köpa [Sprakbanken: Twitter, 2016].
'Haha, I'll surely think of a lot (lit. 'mass') more things that I should buy.'
(37) Vi behöver högre straff pa sexbrott och massa fler poliser [Sprakbanken: Twitter, 2017].
'We need stricter punishments for sex crimes and a lot (lit. 'mass') more police officers.'
(38) <...> sy klader + en massa till [Sprâkbanken: Bloggmix, 2011]. 'sewing clothes + a lot (lit. 'mass') more.'
Interestingly, what the above examples also demonstrate is that despite its lower percentage of adverbial attestations, massa 'mass' seems to have undergone a higher level of decategorialization than smula 'crumb', which manifests itself in the occasional omission of the indefinite article in the latter case (cf. (29), (31), (33)-(35), and (37)).
4.5. Massor
Whereas the adverbial uses of massa 'mass' in the verbal domain are almost equally distributed between inherent degree and extent modification, massor 'masses' displays a preference for the former type of contexts. Employed as a degree modifier, massor 'masses' further tends to combine with psychological verbs (cf. (39)-(42)), most of which exhibit a positive coloring.
(39) Ja absolut, vi mâste ses snart, saknar dig massor! [Sprâkbanken: Twitter, 2016]
'Yes, definitely, we must see each other soon, I miss you a lot (lit. 'masses')!'
(40) Vi alskar varandra massor och kanslan av att vi numera ar en familj ar fantastisk [Sprâkbanken: Bloggmix, 2013].
'We love each other a lot (lit. 'masses') and the feeling that we are a family now is fantastic.'
(41) Det ser vi fram emot massor [Sprâkbanken: Bloggmix, 2013]. 'We're looking forward to that a lot (lit. 'masses').'
(42) Och sâ mâste jag visa Dellas nya Bumbo stol som hon gillade massor! [Sprâkbanken: Bloggmix, 2012]
'And so I have to show Della's new Bumbo-table that she liked a lot (lit. 'masses')!'
In the dataset under analysis, there are no adverbial uses of massor 'masses' involving positive forms of adjectives/adverbs. However, in their investigation into the emergence of degree modifier uses of Dutch nominal quantifiers, Norde et al. [Norde, De Clerck, Colleman, 2014, p. 245]
mention, in passing, one authentic example where Swedish massor 'masses' functions adverbially in relation to an adjective in its positive form, namely Resan hem var massor trevlig 'The journey home was very (lit. 'masses') nice', which indicates that rather than being non-existent, adverbial occurrences of the analyzed Swedish item with positives are in fact conspicuously infrequent, especially when compared with those involving comparatives. Similarly to the case with massa 'mass', a vast majority of the degree modifier uses of massor 'masses' in the adjectival domain involve the comparative forms mer 'more' and fler 'more' (cf. (43)-(44)), attestations with other items being few and far between (cf. (45)).
(43) Det finns massor mer jag vill beratta, men jag har inte riktigt tid nu [Sprakbanken: Flashback: Livsstil].
'There's a lot (lit. 'masses') more I want to talk about but don't really have time now.'
(44) Samre bullerkrav ger inte massor fler bostader [Sprakbanken: Twitter, 2017].
'Worse noise limitations won't produce a lot (lit. 'masses') more dwellings.'
(45) Den har roda outfitten ar massor battre an det hon hade pa sig sist [Sprakbanken: Twitter, 2017].
'This red outfit is a lot (lit. 'masses') better than the one she was wearing last time.'
5. CONLUSION
Based on synchronic corpus data, this study has sought to examine the adverbialization patterns of nine Swedish nominal quantifiers, with a focus on the identification of possible bridging contexts. The obtained results suggest that in the verbal domain, pronominal attestations constitute an intermediate stage between nominal quantification and adverbial modification, while in the adjectival domain, it is uses involving comparatives that seem to drive the change. Moreover, drawing on additional data from Herda [forthcoming], it has been established that there exists a strong positive correlation between the percentages of the analyzed items' quantifier and adverbial uses, thus indicating that frequency plays an important role in the grammaticalization of Swedish partitive nouns. Notably, all of these findings concur with those arrived at in an analogous investigation devoted to English (cf. [Herda,
2022b]), which lends further substance to their cross-linguistic validity. Still, future research on the topic should be extended to other, especially non-Germanic, languages in order to confront the above-discussed observations with novel empirical data.
REFERENCES
Bolinger D. Degree Words. The Hague: Mouton, 1972. 324 p.
Brems L. Measure noun constructions: An instance of semantically-driven grammaticalization. International Journal of Corpus Linguistics, 8 (2), 2003. P. 283-312.
Brems L. The grammaticalization of small size nouns: Reconsidering frequency and analogy. Journal of English Linguistics, 35, 2007. P. 235-293.
Brems L. The Layering of Size Noun and Type Noun Constructions in English. Berlin: Mouton de Gruyter, 2011. 409 p.
Brems L. Some binominal size noun constructions in English and French: A con-trastive corpus-based perspective. Nordic Journal of English Studies, 14 (1), 2015. P. 90-115.
Claridge C., Kyto M. I had lost sight of them then for a bit, but I went on pretty fast: Two degree modifiers in the Old Bailey Corpus. Diachronic Corpus Pragmatics. Ed. by I. Taavitsainen, A. Jucker, J. Tuominen. Amsterdam, Philadelphia: John Benjamins, 2014. P. 29-52.
De Clerck B., Brems L. Size nouns matter: A closer look at mass(es) of and extended uses of SNs. Language Sciences, 53, 2016. P. 160-176.
Doetjes J. S. Quantifiers and Selection: On the Distribution of Quantifying Expressions in French, Dutch and English. The Hague: Holland Academic Graphics, 1997. 307 p.
Herda D. On the adverbialization of Polish indefinite quantifiers of nominal origin: A diachronic study of troche 'a bit, odrobinf 'a bit', and mas$ 'a lot. Studies in Polish Linguistics, 14 (1), 2019a. P. 19-42.
Herda D. On the effect of pluralization on the numeralization of nouns in English and Polish: A contrastive corpus-based study. Linguistica Silesiana, 40, 2019b. P. 139-155.
Herda D. On the gradience of English size nouns: Frequency, productivity, and expansion. Baltic Journal of English Language, Literature and Culture, 12, 2022a. P. 30-47.
Herda D. On the grammaticalization of partitives into degree adverbs: The case of Polish odrobina 'crumb; a little' and Swedish smula 'crumb; a little'. Zeszyty Naukowe Towarzystwa Doktorantow Uniwersytetu Jagiellonskiego. Nauki Hu-manistyczne, 23 (4), 2018. P. 19-32.
Herda D. To be "a boat load healthier" and not to "care a single scrap": On the ad-verbialization of English size nouns. GEMA Online Journal of Language Studies, 22 (2), 2022b. P. 111-127.
Herda D. Swedish nominal quantifiers and their degrees of grammaticalization: A corpus inquiry. Filologiskt smorgasbord 5. När forskning och undervisning möts — en 50-ars jubileumsskrift fran nordistiken i Kraków. Ed. by E. Data-Bukowska, M. Wasilewska-Chmura. Kraków: Jagiellonian University Press [forthcoming]. Huddleston R., Pullum G. The Cambridge Grammar of the English Language.
Cambridge: Cambridge University Press, 2002. 1842 p. Keizer E. The English Noun Phrase: The Nature of Linguistic Categorization. Cambridge: Cambridge University Press, 2007. 394 p. Kinn T. Norwegian masse: from measure noun to quantifier. The Very Model of a Modern Linguist. BeLLS, vol. 8. Ed. by V. Rosén, K. De Smedt. Bergen: University of Bergen, 2017. P. 143-166. Langacker R. Foundations of Cognitive Grammar. Vol. 2: Descriptive Application.
Stanford: Stanford University Press, 1991. 628 p. Norde M., De Clerck B., Colleman T. The emergence of non-canonical degree modifiers in non-standard varieties of Dutch: A constructionalization perspective. Extending the Scope of Construction Grammar. Ed. by R. Boogaart, T. Colleman, G. Rutten. Berlin, New York: De Gruyter Mouton, 2014. P. 207250.
Qurik R., Greenbaum S., Leech G., Svartvik J. A Comprehensive Grammar of the
English Language. London; New York: Longman, 1985. 1779 p. SAOB — Svenska Akademiens ordbok. Available at: https://www.saob.se (accessed: 09.03.2023).
Sprakbanken. Available at: https://spraakbanken.gu.se/korp/ (accessed: 09.03.2023)
Traugott E. C. Grammaticalization, constructions and the incremental development of language: Suggestions from the development of degree modifiers in English. Variation, Selection, Development — Probing the Evolutionary Model of Language Change. Ed. by Eckardt R., Jäger G., Veenstra T. Berlin; New York: De Gruyter Mouton, 2008. P. 219-250. Verveckken K. D. Binominal Quantifiers in Spanish: Conceptually-driven Analogy in Diachrony and Synchrony. Berlin; Boston: Mouton de Gruyter, 2015. 516 p.
Дамиан Херда
Ягеллонский университет, Польша
ОТ ИМЕННЫХ КВАНТИФИКАТОРОВ К НАРЕЧНЫМ МОДИФИКАТОРАМ: КОРПУСНОЕ ИССЛЕДОВАНИЕ НА МАТЕРИАЛЕ ШВЕДСКОГО ЯЗЫКА*
Для цитирования: Herda D. From nominal quantifiers to adverbial modifiers: A corpus investigation with reference to Swedish // Скандинавская филология. 2023. Т. 21. Вып. 2. С. 199-217. https://doi.org/10.21638/11701/spbu21.2023.201
* Исследование проведено при поддержке Национального научного центра Польши (грант № 2019/33/М/Ш2/01695).
Как отмечается в литературе по грамматикализации, партитивные именные части речи, обозначающие части, порции и множества, имеют тенденцию развиваться в неопределенные квантификаторы, а затем в наречия степени. Однако, несмотря на то что этот процесс уже был предметом многих исследований, основанных на данных английского языка, распространение этого типа единиц на наречия остается неисследованным в шведском языке. В данной статье предлагается синхронный анализ адвербиализации девяти шведских партитивных существительных, а именно droppe 'капля', nypa 'щепотка', smula 'крошка', hop 'куча', hopar 'кучи', hög 'груда', högar 'груды, massa 'масса' и massor 'массы', на основе случайных выборок, взятых из отобранных корпусов Sprâkbanken. Полученные результаты свидетельствуют о том, что в глагольной области большинство проанализированных примеров относятся к местоименным употреблениям, в то время как в адъективной области отмечается преобладание сочетаний со сравнительными степенями прилагательных/наречий. Таким образом, оба этих типа окружений можно рассматривать как промежуточные контексты, способствующие конвенционализации наречной функции у изучаемых существительных. Более того, было обнаружено, что существует сильная положительная корреляция между уровнем грамматикализации изучаемых единиц и степенью их адвербиализации, что доказывает значительную роль частоты употребления в анализируемом явлении.
Ключевые слова: грамматикализация, адвербиализация, именные квантификаторы, наречия степени, шведский язык.
Damian Herda
PhD in Linguistics,
MA in English Studies and Swedish Philology, Research Assistant,
Jagiellonian University,
9a, al. Adama Mickiewicza, Kraków, 31-120, Poland
E-mail: damian.herda@uj.edu.pl
Херда Дамиан
доктор лингвистики,
магистр английских исследований и шведской филологии,
Ягеллонский университет,
Польша, 31-120, Краков, ул. Адама Мицкевича, 9а
E-mail: damian.herda@uj.edu.pl
Received: March 10, 2023 Accepted: July 1, 2023
