We have so far looked more closely into maximizers (Chapter 5) and boosters (Chapter 6), which can be found at one end of the intensification scale. We now turn to downtoners, which can be found towards the opposite end of the intensification scale, scaling down the force of the target of modification. We start by introducing our classification of our downtoners and recapping our research questions (Section 7.1). We then survey the types and tokens comprising our data (Section 7.2.1) and discuss the suffixed and suffixless forms (Section 7.2.2). In Section 7.3 we look into the semantics of our downtoners and then focus on their targets of modification and collocational features in Section 7.4. Our findings are summarized in Section 7.5.
7.1 Introduction
As explained in Chapter 3 (Section 3.2.1), downtoners scale ‘downward from an assumed norm’ (Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. 1985: 590) and comprise a variety of items classified in various ways in the literature. They have ‘a generally lowering effect on the force of the verb or predication’, and they can display various semantic distinctions (Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. 1985: 597). When looking for potential downtoner candidates for our searches, we consulted the HTOED (path: the external world > relative properties > quantity or amount > moderateness of quantity, amount, or degree / smallness of quantity, amount, or degree / insufficiency), searched the OBC for adverbs ending in -ly, and consulted standard reference grammars (Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. 1985; Reference Huddleston and PullumHuddleston and Pullum 2002) and recent literature (notably Reference BolingerBolinger 1972, Reference ParadisParadis 1997, Reference MargerieMargerie 2008, and the items referred to under example (19)).
We have opted to focus on three subgroups of downtoners, namely moderators, diminishers, and minimizers, which can be arranged on the intensification scale in the order mentioned, ranging from the mildest to the strongest (for justification for our definitions and classification, see Section 3.2). More specifically, we consider eight moderators, that is, comparatively, fairly, moderately, partially, passably, reasonably, relatively, somewhat; five diminishers, that is, a little, faintly, lightly, slightly, sparingly; and six minimizers, that is, barely, hardly, scantily, scarcely, (in) the least, (very) little. One example of each downtoner in context is given below, examples (1)–(8) being moderators, (9)–(13) diminishers and (14)–(19) minimizers; the moderators in examples (5)–(7) and the diminisher in (13) constitute the only instantiation of each respective item.
Moderators
(1) the wound was deep and comparatively dangerous (t18900908-689, witness, m higher)
(2) she writes letters in English, and, judging from them, she can read and write fairly well (t19060108-169, witness, f)
(3) the liver was moderately healthy (t18700404-333, witness, m, higher)
(4) my opinion was that she was an angry woman partially excited by drink, (t18820109-202, witness, m, higher)
(5) when I am sober he is passablyFootnote 1 kind (t18421128-56, witness, f)
(6) in a reasonably healthy child the skin would heal more quickly than with older people (t19000212-185, witness, m, higher)
(7) I came to the conclusion that the fatal dose was taken within two days of the death because of the relatively large amount found in the stomach and intestines. (t19120227-48, witness, m, higher)
(8) I am instructed by the prisoner to state, that he was somewhat in embarrassed circumstances [italics in the original] (t18380129-625, lawyer, m, higher)
Diminishers
(9) It made me a little giddy. (t17551204-29, victim, f, lower)
(10) I looked, and I saw faintly the figure of a man with something white over his shoulders; (t18100110-39, f, lower)
(11) I said, “Yes,” and handed him the envelope, which was lightly sealed down. (t19111205-16, defendant, m, lower)
(12) I was slightly wounded in my breast. (t17721021-23, victim, f, higher)
(13) this is unlike the medical men,Footnote 2 who very sparingly take themselves, or administer to their families, the drugs they send in such plenitude to their patients. (t18360404-906, defendant, m, higher)
Minimizers
(14) Gentlemen, it is still more unpleasant for me to say a word on the conduct of the prisoner, I shall therefore barely state it, (t17830726-1, lawyer, m, higher)
(15) it was very dark indeed, I could hardly observe any thing before me (t18261026-34, witness, f)
(16) the room was scantily furnished, but clean (t19050626-547, witness, m, higher)
(17) I had scarce finished doing so when she returned, (t18120916-52, defendant, m, lower)
(18) I got no share in the plunder – that did not annoy me in the least; (t18350511-1164, witness, f, lower)
(19) the brain was healthy and very little ensanguined, less than usual; (t18640411-442, witness, m, higher)
As has appeared in our discussion of the history of intensifier usage, most research has targeted amplifiers, while only relatively little has been done on downtoners. Diminishers and minimizers, in particular, have been neglected (Reference RissanenRissanen 2008a: 345), with only a few exceptions (e.g., Reference Claridge, Kytö and HundtClaridge and Kytö 2014a, Reference Claridge, Kytö, Taavitsainen, Jucker and Tuominen2014b; Reference BrintonBrinton 2021; Reference Claridge, Jonsson and KytöClaridge, Jonsson, and Kytö 2021). However, as intensifiers have emerged as an area of ongoing change and displayed sociolinguistic patterning (e.g., Reference MacaulayMacaulay 2002), we can expect downtoners to show interesting trends of change. To report on previous findings, Borst in his early study (Reference Borst1902) found out that of the some 240 amplifiersFootnote 3 and some 30 downtoners he investigated, 58 per cent and 61 per cent, respectively, emerge in the (early) modern period.Footnote 4 Of the amplifiers, 20 per cent are no longer in use today, and some others have survived only in dialects; yet no more than two downtoners have been lost (Reference BorstBorst 1902: 157–9, with no specification as to which items were in question; see also Reference Claridge, Kytö, Taavitsainen, Jucker and TuominenClaridge and Kytö 2014b: 31). Our previous research (Reference Claridge, Jonsson and KytöClaridge, Jonsson, and Kytö 2021) on the OBC has shown downtoners to be overall less common than amplifiers in the late modern courtroom discourse, a result that tallies with the findings presented by Reference D’ArcyD’Arcy (2015: 462–6, 467) on spoken New Zealand English used by speakers born in 1851–1980. Not many single-word downtoners have been investigated historically, among them rather (Reference Rissanen, Hasselgård and OksefjellRissanen 1999b, Reference Rissanen2008a), fairly versus rather (Reference MargerieMargerie 2008), fairly and pretty (Reference Nevalainen and RissanenNevalainen and Rissanen 2002), nigh(ly), near and next (to) (Reference BrintonBrinton 2021), and (a) little (Reference Claridge, Jonsson and KytöClaridge, Jonsson, and Kytö 2021). In the last-mentioned study, we found that (a) little dominated clearly in our group of downtoners but declined across our period (1720–1913).
In terms of pragmatics, downtoners are generally treated as mitigation phenomena such as hedges (Reference LakoffLakoff 1973; Reference Fraser, Kaltenböck, Mihatsch and SchneiderFraser 2010; cf. also Reference GrundGrund 2021: 142) or hedges, bushes, or shields (Reference CaffiCaffi 1999). For Reference LakoffLakoff (1973: 471), hedges are words ‘whose job is to make things fuzzier or less fuzzy’, as, for instance, sort of in sort of tall and intensifiers such as pretty in pretty much. Reference CaffiCaffi (1999: 883), extending Lakoff’s metaphor, presents a three-fold classification of mitigating mechanisms, that is, bushes, hedges, and shields, ‘based on the three components of the utterance on which mitigation can operate: the proposition, the illocution, and the utterance source’.Footnote 5 In its narrow sense (Reference FraserFraser 1980), mitigation relates to face-threatening acts (Reference Brown and LevinsonBrown and Levinson 1987) and in its broad sense to the notions of weakening, downgrading, and downtoning. The latter sense involves interactional factors beyond mere face concerns, among them the weakening of one of the interactional parameters and the effects of cognitive, emotive, and other psychological factors (Reference CaffiCaffi 1999: 884).
From our perspective, it is of interest to consider the distinction Reference Fraser, Kaltenböck, Mihatsch and SchneiderFraser (2010: 15) makes between propositional hedging, as in He’s sort of nice and speech act hedging as in I must ask you to stop doing that (cf. Reference HüblerHübler 1983). In the former we have a rhetorical strategy that attenuates the full semantic value of the proposition (cf. bushes in Reference CaffiCaffi 1999: 888–9); in the latter it is the full force of the speech act that is affected (cf. hedges affecting the illocutionary force in Reference CaffiCaffi 1999: 888–9). In propositional hedging, it is the truth value of the proposition that is affected, and in speech act hedging, it is the commitment of the speaker to the illocutionary force of the utterance that is concerned. In usage, these go together to some extent, as weakening the propositional content will be based on or affect speaker commitment. Consequently, we have not coded our data systematically for this distinction. In our approach, we take as propositional those types which have scope over a specific constituent of an utterance, for example, in example (20), about six minutes (NP, adverbial) and somewhat more (adverbial) instead of having scope over the complete utterance/clause; note that there are several hedging expressions in the context, given in italics. Propositional hedging seems to be the dominant, almost exclusive form with downtoners, although instances of speech act hedging also occur as in (21); the items included in our study are underlined.
(20) Q. What distance of time might it be between the firing the two pistols?
W. Bridger. It might be about six minutes, it was but a very little time. Then they went away all together, I saw them go towards Mayhew’s house; this place was about a quarter of a mile, or somewhat more, from their house, I did not see them go into the house. (t17510417-57, witness, m, higher)
(21) Q. I need hardly ask you if he is acquainted with your Christian name? (t18000917-46, m, lawyer)
Our research into the downtoners attempts to answer four questions. Firstly, what is the distribution across the LModE period and what can be stated about the trends of development on a preliminary basis (we will return to the developments across time in a systematic way in Chapter 9)? Secondly, what targets do the downtoners investigated modify? Thirdly, what patterns of collocational preferences can be revealed by a semantic classification of the targets of modification, and fourthly, what is the distribution across speakers with regard to their speaker roles, gender, and social class? The first three questions will be addressed in this chapter and the last question in Chapters 10 and 11.
7.2 An Inventory of Forms
7.2.1 Types and Tokens: An Overview
We outlined the steps we took when looking for potential downtoner candidates to be included in our study in Section 7.1. As with maximizers and boosters, we also went through the output list of a search of all -ly forms in the OBC. In contrast to amplifiers, only moderately, slightly, and scarcely were found to have the corresponding zero forms moderate, slight, and scarce in our data. Besides the items presented in the previous section, we searched for instances of marginally, mildly, thinly, meagrely, and partly, finding no relevant instances of the first four; we list our attested and non-attested types of downtoners in Table 7.1. Partly presented a special problem, so we ultimately excluded the form from our investigation due to severe doubts regarding its actual degree usage (see Section 3.2.3). It is a very frequent form but seems to have a fairly literal sense in most cases, indicating, for example, a distinct part, segment, or component, such as he was partly dressed; he had no coat or waistcoat on (t18820327-431). This was most clear in paired uses, such as partly this and partly that but also obvious enough in single occurrences.
Table 7.1 Downtoners searched for in the Old Bailey Corpus
| Attested types: | 19 | a little, barely, comparatively, faintly, fairly, hardly, lightly, moderately, [partly], partially, passably, reasonably, relatively, scantily, scarcely, slightly, somewhat, sparingly, (in) the least, (very) little |
| Non-attested types: | 4 | marginally, mildly, thinly, meagrely |
Out of the total of more than 35,000 hits of our search forms (see Appendix A), our screening resulted in a total of 7,874 relevant examples of downtoners; of these, 5,186 were instances of a little and 2,688 instances of other downtoners. In addition, some fifty instances were classified as indeterminate uses and set aside. It is noteworthy that downtoners overall are clearly less frequent and diverse than amplifiers; this is mirrored by, for example, Grund’s Salem degree modifier data (N = 981) only containing 4–5 per cent downtoners (Reference GrundGrund 2021: 118–19; see also Reference D’ArcyD’Arcy 2015). There also seems to be far less innovation and change in the area of downtoners than is the case with amplifiers.
Figure 7.1 gives an overview of the development of the downtoner categories moderators, diminishers, and minimizers as compared to the distribution of the boosters and maximizers investigated in the previous chapters. While there is a decline in the downtoners and the boosters, the downtoners displaying a less prominent downhill curve than the boosters, the development for the maximizers is the reverse; as seen in Chapter 5, maximizers were on the rise.
Figure 7.1 Intensifiers per 100,000 words in the Old Bailey Corpus, by category
Figure 7.2 shows the distribution of the full inventory of the downtoners studied. While the dominant type a little constitutes 66 per cent of the occurrences and is followed by hardly, comprising 13 per cent, the rarest types each constitute less than half a per cent of the downtoners data (for consistency, the percentages are rounded to 0 per cent in Figure 7.2).
Figure 7.2 Distribution of the downtoners per 100,000 words, shown as bars. Raw frequencies and the proportion of a type among all downtoners are listed in the right margin
The line graph in Figure 7.3 presents the diachronic distribution of the five most frequent downtoners of our study, a little and slightly being diminishers and the other three, hardly, scarcely, and (in) the least being minimizers. Taken together, these top five items constitute as much as 91 per cent of our downtoner data (7,167 out of 7,874 tokens). While there is a decline in the diminisher a little, the diminisher slightly is on the rise, as well as the moderators fairly, partially, and somewhat (not shown in Figure 7.3). Slightly thus appears to encroach somewhat on the space of a little, potentially taking over some of its functions. Among the minimizers, hardly declines slightly, while scarcely and (in) the least fluctuate across the five subperiods.
Figure 7.3 Distribution of the five most frequent downtoners across the period studied (1720–1913) in normalized frequencies per 100,000 words
From the present-day perspective, Reference Biber, Johansson, Leech, Conrad and FineganBiber et al. (1999: 566–7) note that non-amplifiers display a wider range of items in academic prose than in conversation. They give corpus findings for four of our moderators, relatively, fairly, somewhat, and moderately, and for one of our diminishers, slightly. None of these occur in British or American English conversation above the 5 instances per 100,000 words adopted for listing the results (or in their terms, 50 instances per one million words). In academic writing, slightly, fairly, and relatively occur 10 to 20 times and somewhat 5 times per 100,000 words; moderately only occurs once per 100,000 words (no register information given). The corresponding figures for the occurrence in our speech-related data in the last subperiod (1880–1913) are 3.6 for slightly, 2.5 for fairly, 1.7 for somewhat, and 0.2 for moderately (per 100,000 words); relatively only occurs once in our full time span. Comparing the rates of occurrence of our downtoners with those presented by Reference Hessner and GawlitzekHessner and Gawlitzek (2017) in their study of intensifiers in the Spoken BNC2014 corpus, we find that the rising trend recorded for slightly in the OBC data (subperiod figures: 0.1, 0.3, 0.3, 2.2, 3.6) has continued up until the present day (8.7 occ. per 100,000 words); see Table 7.2 and Figure 7.4. Another item displaying increase in use is fairly, which emerged in our data in the fourth subperiod (1840–79), occurred 2.5 times per 100,000 words in the last subperiod and is used 5.7 times per 100,000 words in the Spoken BNC2014 data. Hardly presents a declining trend up until the 1800s and, after some minor fluctuation, reaches the rate of 4.0 occurrences per 100,000 words in the Spoken BNC2014. Barely, a relatively rare item in our data, also presents a rising trend, albeit with a relatively low figure (1.4) in the Spoken BNC2014 data.
Table 7.2 The twelve most common downtoners in our Old Bailey Corpus data and the frequency in BNC2014S of those documented by Reference Hessner and GawlitzekHessner and Gawlitzek (2017); frequencies per 100,000 words
| 1720−59 | 1760−99 | 1800−39 | 1840−79 | 1880−1913 | BNC2014S | |
|---|---|---|---|---|---|---|
| a little | 30.2 | 26.2 | 17.7 | 20.4 | 15.1 | |
| hardly | 5.5 | 4.5 | 3.3 | 4.5 | 3.5 | 4.0 |
| scarcely | 0.8 | 0.9 | 1.2 | 3.0 | 1.1 | |
| slightly | 0.1 | 0.3 | 0.3 | 2.2 | 3.6 | 8.7 |
| (in) the least | 0.8 | 2.0 | 0.9 | 0.9 | 0.5 | |
| (very) little | 1.3 | 1.2 | 0.8 | 0.9 | 0.7 | |
| somewhat | 0.4 | 0.3 | 0.2 | 0.9 | 1.7 | |
| fairly | – | – | – | 0.02 | 2.5 | 5.7 |
| partially | – | – | 0.2 | 0.7 | 0.8 | |
| comparatively | – | 0.02 | 0.1 | 0.2 | 0.3 | |
| barely | 0.1 | 0.1 | 0.04 | 0.1 | 0.3 | 1.4 |
| moderately | – | 0.02 | – | 0.2 | 0.2 |
Figure 7.4 Diachronic distribution of the four downtoners in our Old Bailey Corpus data that are also found in Reference Hessner and GawlitzekHessner and Gawlitzek’s (2017) study of BNC2014S (normalized frequencies)
Taking into account the genre or register factor, with our data deriving from trial records and the present-day data from other spoken contexts, we looked into the 13 trial texts of the BNC1994, totalling 126,256 words; see Table 7.3 and Figure 7.5. The downtoners in our study occur with an overall incidence of 32.2 per 100,000 words (Tables 4.1a–c on pp. 74–5), while the figure obtained for the shared items in the BNC1994 trials data is almost three-fold, 110 per 100,000 words; of our 19 items, only 8 are represented in the BNC1994 trials data. Even though the incidence figure for the BNC1994 is based on only fifty-six occurrences, it is interesting when considering the register effect attested in Reference Biber, Johansson, Leech, Conrad and FineganBiber et al. (1999: 566–7), that is, that our items recorded in their present-day data were more prone to occur in academic prose than in spoken language. What the courtroom and academic discourse have in common is the formality of the register, as well as the need to hedge one’s statements where there are no guarantees for the precision of the information communicated. Also, and perhaps even more interestingly, slightly (11.9 occ. per 100,000 words) and fairly (9.5) oust little (7.9) in the present-day data. This may be due to the small data set, or possibly to stylistic issues, slightly and fairly being a modicum higher stylistically than little. It may also go back to semantic nuances, slightly and fairly covering narrower and thus more precise areas than little. Note that it was also the case that maximizers proved much more frequent in the present-day courtroom discourse than in the late modern data (see Section 5.2.1, Figure 5.4); among the boosters (Chapter 6), very displayed practically similar figures in the BNC1994 trials as in our courtroom data, while so and greatly have lost ground.
Table 7.3 Downtoners attested in the British National Corpus trials (normalized frequencies per 100,000 words)
| Downtoner | Raw frequency | Normalized frequency |
|---|---|---|
| slightly | 15 | 11.9 |
| fairly | 12 | 9.5 |
| little | 10 | 7.9 |
| reasonably | 7 | 5.5 |
| relatively | 4 | 3.2 |
| somewhat | 4 | 3.2 |
| partially | 3 | 2.4 |
| least | 1 | 0.8 |
| Total | 56 | 110.0 |
Figure 7.5 The top five Old Bailey Corpus downtoners and their occurrence in the British National Corpus trials (further OBC downtoners in the BNC trials: fairly 9.5, reasonably 5.5, relatively 3.2, somewhat 3.2)
7.2.2 Dual-Form Downtoners: Suffixed and Suffixless Forms
As with maximizers (Chapter 5) and boosters (Chapter 6), downtoners occur in suffixed and suffixless forms. However, as mentioned earlier, of the nineteen items included in the study, only three present dual forms: scarcely, moderately, and slightly. These uses are illustrated in examples (22) to (24); for the frequency information, see Table 7.4.
(22) I could scarce perceive any Part of his Body free. (t17320906-69, witness, f)
(23) what I want to convey is that in moderate temperate conversation with me, other people being in the room, she would suddenly conceive some thought and express herself excitably to those people (t18620707-745, witness, m, higher)
(24) in mixing it so as to make manure there is very little that is unpleasant or disagreeable; if there is, it is a slight acidulousFootnote 6 odour, but nothing noxious (t18570706-817, witness, m, higher)
Table 7.4 Suffixed versus suffixless downtoner forms (raw frequencies and percentages of the suffixless forms)
| Suffixed | Suffixless | Total | Percentage of suffixless | |
|---|---|---|---|---|
| scarcely | 315 | 49 | 364 | 13% |
| moderately | 19 | 1 | 20 | 5% |
| slightly | 337 | 5 | 342 | 1% |
Interestingly, as many as twenty of the twenty-six instances of scarcely found in the first subperiod occur in zero form. In diachronic terms, the zero forms dwindle down rapidly across the subsequent periods (21/7/1/0 out of the total of 41/67/177/53). The handful of zero instances of slightly and moderately all occur in the two final subperiods. For the overall development of downtoners, see Section 4.3.1, Figure 4.7.
7.3 The Semantic Inventory of Downtoners
The distribution of our downtoners within the three subgroups we have distinguished on semantic grounds is displayed in Table 7.5. As mentioned earlier, all downtoners have a lowering effect on the target of modification (Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. 1985: 597), and under this umbrella effect, it is possible to distinguish semantic shades in the force expressed by the various items. We discuss the rather varied ways of defining and classifying downtoner groups in Chapter 3 (see Section 3.2). Our three subgroups of downtoners draw mostly on Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. (1985) and Reference Huddleston and PullumHuddleston and Pullum (2002), with affinities to Reference BolingerBolinger (1972) and Reference ParadisParadis (1997); for illustrations, see examples (1)–(19). Thus, moderators are found at one end of the downtoning scale, conveying moderating, or slightly lowering, effect denoted by the target of modification (Reference ParadisParadis 1997: 86–9; Reference Huddleston and PullumHuddleston and Pullum 2002: 722–3). Diminishers are found further down on the scale, taken to convey the meaning ‘to a small extent’. Minimizers then target the lowest end of the scale and appear as ‘negative maximizers’, which convey the meaning ‘(not) to any extent’ (Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. 1985: 597).
Table 7.5 Downtoner subcategories (raw frequencies)
| Moderators | Diminishers | Minimizers | |||
|---|---|---|---|---|---|
| somewhat | 169 | a little | 5,186 | hardly | 1,023 |
| fairly | 124 | slightly | 342 | scarcely | 364 |
| partially | 90 | faintly | 5 | (in) the least | 252 |
| comparatively | 31 | lightly | 2 | (very) little | 229 |
| moderately | 20 | sparingly | 1 | barely | 29 |
| passably | 1 | Total | 5,536 | scantily | 4 |
| reasonably | 1 | Total | 1,901 | ||
| relatively | 1 | ||||
| Total | 437 | ||||
The distribution of our downtoner occurrences across the three semantic subcategories is given in Table 7.5 (raw frequencies); the diachronic development across the subperiods appears in Table 7.6 and in Figure 7.6. Of the three groups, the diminishers are the best represented, mostly owing to the top-frequency item a little. Their use decreases across the period, with a temporary rise towards the end of the nineteenth century. Minimizers are less well represented, with hardly as their top item, and also show a declining trend with a temporary rise in the same period as diminishers. Moderators are the least-represented category in our data, but their use starts increasing from the 1800s onward; the top-frequency items are somewhat, fairly, and partially. The same distribution is shown by Reference GrundGrund’s (2021: 119) Salem data, with twenty-seven, ten, and three occurrences for paucal (= diminisher), minimal (= minimizer) and moderate modifiers, respectively. We now turn to the semantic development of our downtoners to gain some insights into the kind of source terms from which they originated.
Figure 7.6 Downtoners by semantic category (normalized frequencies)
Table 7.6 Downtoners by semantic category (normalized frequencies per 100,000 words)
| 1720−59 | 1760−99 | 1800−39 | 1840−79 | 1880−1913 | |
|---|---|---|---|---|---|
| Diminishers | 30.4 | 26.5 | 18.1 | 22.7 | 18.7 |
| Minimizers | 8.4 | 8.8 | 6.3 | 9.4 | 6.1 |
| Moderators | 0.4 | 0.3 | 0.4 | 2.0 | 5.5 |
The source terms of our downtoners give the impression that the spectrum of semantic shades they display is more varied than is the case with maximizers and boosters. Among such shades are appearance and/or fairness/rightfulness (e.g., fairly, faintly, lightly, reasonably) and evaluation regarding the quality in question (e.g., moderately, sparingly, scarcely), on the basis of the definitions and paths of change given for the items in the OED. We illustrate the quantitative and qualitative aspects of the source terms involved in Figure 7.7; for maximizers and boosters, see Figures 5.6 (Section 5.3) and 6.4 (Section 6.3), respectively.
Figure 7.7 Semantic input domains of downtonersFootnote 7, Footnote 8
Figure 7.7 shows that in terms of types, the quantitative and qualitative source domains are evenly represented in our data, with ten and nine types, respectively. Yet in terms of the tokens, the category of quantitative origin is represented by over 83 per cent of the data, as a result of strong dominance of a little (see Table 7.7, on p. 178). Within the category of qualitative origin, the subcategory of force, difficulty, comprising hardly, constitutes over 78 per cent of the data. Overall, speakers in the Old Bailey courtroom seem to resort to the more neutral downtoners (a little, hardly) at the expense of the more colourful items. These two items are rather bleached items and hardly can probably be taken to be already exclusively scalar. However, little (in spite of its frequency) is probably somewhat less bleached, while items like passably or sparingly may retain more of their original meanings. Overall, many items in Figure 7.7 have not completely bleached nor reached a pure scalar reading (e.g., moderately, comparatively, lightly, partially, reasonably).
Table 7.7 Downtoners by semantic input domain (raw frequencies)
| Quantitative origin | Qualitative origin | ||
|---|---|---|---|
| (limited/restricted) amount/size | comparison | ||
| comparatively | 31 | ||
| a little | 5,186 | relatively | 1 |
| scarcely | 364 | appearance | |
| slightly | 342 | barely | 29 |
| (in) the least | 252 | lightly | 2 |
| (very) little | 229 | force, difficulty | |
| somewhat | 169 | hardly | 1,023 |
| moderately | 20 | incompleteness | |
| scantily | 4 | partially | 90 |
| passably | 1 | rightfulness, evaluation | |
| sparingly | 1 | fairly | 124 |
| Total | 6,568 | faintly | 5 |
| reasonably | 1 | ||
| Total | 1,306 | ||
The category of quantitative origin and, within the category of qualitative origin, the subcategory of force, difficulty, contribute most to the declining use of downtoners across our period; see Table 7.8. The former displays an overall decline of some 30 per cent from the 1720s to 1913, and the latter a decline of some 40 per cent. Yet within the category of qualitative origin, we can also see a sharp rise in the subcategory of rightfulness, evaluation, occasioned by the increase in the use of fairly during the last subperiod.
Table 7.8 Distribution of downtoners by semantic input domain (normalized frequencies); lowest frequencies retain two decimal places to be distinguishable
| 1720−59 | 1760−99 | 1800−39 | 1840−79 | 1880−1913 | |
|---|---|---|---|---|---|
| Quantitative origin | |||||
| (limited/restricted) amount/size | 33.6 | 30.8 | 21.2 | 28.5 | 22.9 |
| Qualitative origin | |||||
| comparison | 0.0 | 0.02 | 0.1 | 0.2 | 0.3 |
| appearance | 0.1 | 0.1 | 0.04 | 0.1 | 0.3 |
| force, difficulty | 5.5 | 4.5 | 3.3 | 4.5 | 3.5 |
| incompleteness | 0.0 | 0.0 | 0.2 | 0.7 | 0.8 |
| rightfulness, evaluation | 0.0 | 0.02 | 0.02 | 0.05 | 2.53 |
Further insights to the semantics of our downtoners can be gained from the developments accounted for in the OED. The first degree meanings recorded in the OED for our downtoners stretch across the history of English, mostly from the ME period onwards. The degree meanings recorded only after the 1500s include, in EModE, for example, moderately, passably, relatively, faintly, slightly, and barely; those recorded only from 1700s onward include, for example, comparatively, fairly, and scantily. Starting with our best and better represented and lexically related types, their source terms go back to items conveying meanings referring to small size or amount, that is, a little as a diminisher, and (very) little and the superlative form in the least as minimizers (OED, s.v. little), or to senses conveying qualitative aspects of appearance, rightfulness, or evaluation, among them various nuances of scantiness, difficulty, or other restrictive or restraining circumstances; for instance, the frequent minimizer hardly, originally denoted, among other things, ‘with energy, force, or strenuous exertion; vigorously; violently’ and ‘with severity or rigour; harshly, cruelly, sorely; strictly’ in OE (OED, s.v. hardly, 1 and 3, respectively) and ‘not easily or readily; with difficulty’ in late ME (5.b); the first example with the Modern English meaning ‘to an insignificant degree; scarcely, barely; not quite; almost not at all’ in the OED is from the early sixteenth century, dated ?1532, and the next from 1553 (8.a).
Of our better-represented diminishers, slightly is attested in EModE, as mentioned earlier, with, for example, the meanings ‘slimly, slenderly; flimsily, unsubstantially’ (the first example from 1521 in the OED, s.v. slightly, 1.a), ‘with little respect or ceremony; disparagingly, slightingly’ (the first example from 1601, under meaning 4), and with the present-day sense ‘in a slight or small degree; to a slight extent’ (the first example from 1594, under 5.a). Of the better-represented minimizers, scarcely is recorded in the sense ‘scantily, in small quantities; inadequately, sparingly, niggardly, parsimoniously’ in ME (the first examples in the OED from 1340 and a1400, s.v. scarcely, 1); it was also ‘originally used to express a restrictive qualification, = ‘barely’, ‘only just’; hence also, = ‘barely, or not quite’, ‘only just, if at all’’ from the thirteenth century onward (under 2.a). Our best-represented moderator, somewhat, has been recorded in adverbial uses with the meaning ‘in a certain degree or measure; to some (slight or small) extent; slightly, a little; rather’ from the twelfth century onward (OED, s.v. somewhat, adverb, 1.a). Fairly, our second-most frequent moderator, is given two main meanings in the OED, ‘in a fair manner, so as to be fair’ and ‘senses relating to amount, extent, or degree’ (OED, s.v. fairly, I and II, respectively), with undertones of ‘so as to present an attractive appearance; with grandeur; beautifully, finely, splendidly’ (I.1.a) and ‘in accordance with what is right or just; equitably; without bias, impartially. Also: with good reason, rightfully’ (I.4.a) (see also our note 8 in this chapter). Interestingly, the first recorded instance of the use ‘modifying an adjective or adverb: to a moderate degree or extent; reasonably, pretty, quite’ in the OED (II.8.a) appears relatively late, in 1805; our earliest occurrences of the 124 instances of this use attested in the OBC are from 1880. As for partially, our third-most frequent moderator, the meaning ‘to some extent; incompletely, restrictedly; partly’ was already there in late ME, the first instance given in the OED from a1475 (s.v. partially, I.1). Regarding the very rare items in our data, our only example of the diminisher sparingly is from 1836, which tallies with the first instance in the OED with the meaning ‘to a slight extent; in a small degree’ recorded for 1796 (OED, s.v. sparingly, 2.d); the original meaning of ‘in a sparing or saving manner; frugally, economically’ is from c1450 (under meaning 1).
As for the role played by foreign influence, both of our top items, a little and hardly, are of Germanic origin. Latin and Romance languages are thus left only poorly represented among the rest of the items: only eight out of the nineteen items are of Latin or Romance origin, and of the eight items, only scarcely, from Old Northern French, stands out with some 5 per cent (356 occurrences) of all the downtoners.
Considering the nature of the source terms of our downtoners and the related degree meanings already adopted or on their way to the language at the time of the Old Bailey trials, it is easy to appreciate why speakers in these trials found them useful for their communicative purposes as tools lending themselves to hedging or attenuating their statements.
7.4 Targets of Intensification and Collocational Features
7.4.1 A General Survey of Word-Class Targets
As with maximizers and boosters, we are interested in the targets of modification of our downtoners. Figure 7.8 details the distribution of the word-class categories of targets, and Table 7.9 gives the frequencies for the items in the five best-represented categories. These five most common categories are prepositional phrases, verbs (including verb phrases), adjectives, adverbs, and pronouns; see examples (25)–(30) with targets in underlining. Beyond these five, our downtoners also modify participles, nominal determiners, noun phrases, and numerals; see examples (31)–(34). Regarding the verbal area, we have coded both verbs, as in (26), and verb phrases, as in (27), both subsumed under V in Figure 7.8. In (26) slightly modifies the verb touch directly, indicating a low degree of force characterizing the action. In (27) we regard the whole verb phrase [be an expert] as the target. While Reference Huddleston and PullumHuddleston and Pullum (2002: 331) would probably interpret hardly here as an external modifier of the noun phrase an expert, we align with Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. (1985: 451) in seeing this ‘as not specifically modifying the noun phrase, since they are mobile’, where they refers to adverbs like hardly and unlike quite or rather, which do modify noun phrases. Thus, (27) indicates the degree of being an expert (cf. also example (41) for a similar case).
(25) Prepositional phrase
they are worth 3d., scarcely of any value (t18430130-739, victim, m)
(26) Verb
I took my hand carelessly, and touched him slightly in the mouth, but not to hurt him (t18360704-1743, defendant, m)
(27) Verb phrase
I am hardly an expert – looking at it rather more carefully now I should accept it as his signature. (t18880319-406, witness, m)
(28) Adjective
I asked her her name, she was somewhat slow in telling me; (t17500711-53, victim, m)
(29) Adverb
after he was in bed, he turned the cloaths a little back (t17520408-53, f)
(30) Pronoun
she had little more on her bones, set aside skin; (t17670115-24, witness, f)
(31) Participle
she was very scantily fed while with us – I have given the children food because they had not sufficient (t18800803-444, witness, f)
(32) Determiner
Q. There was hardly any time for any length of conversation to take place between Tyler and Stevenson? (t18001203-65, lawyer, m, higher)
(33) Noun phrase
Q. How long after you received it was your attention called to it? A. Scarcely a moment (t18430703-2103, victim, m, higher)
(34) Numeral
there are three different formations of the “d” in that letter, and they are all represented here; out of the four “d’s” there are scarcely two alike; (t18710109-107, witness, m)
Figure 7.8 Downtoners by target of modification; proportional distribution of target categories
Table 7.9 Targets modified by individual downtoners (raw frequencies); the downtoners are listed in alphabetical order
| Prepositional phrases | Verbs/Verb phrases | Adjectives | Adverbs | Pronouns | |
|---|---|---|---|---|---|
| a little | 2,794 | 484 | 990 | 748 | 89 |
| barely | 1 | 13 | 5 | ||
| comparatively | 26 | 2 | 3 | ||
| faintly | 4 | ||||
| fairly | 87 | 36 | |||
| hardly | 7 | 746 | 99 | 65 | 20 |
| lightly | 1 | ||||
| moderately | 2 | 16 | 2 | ||
| partially | 1 | 34 | 43 | 5 | |
| passably | 1 | ||||
| reasonably | 1 | ||||
| relatively | 1 | ||||
| scantily | 1 | 2 | |||
| scarcely | 6 | 235 | 38 | 17 | 15 |
| slightly | 23 | 141 | 117 | 11 | |
| somewhat | 15 | 25 | 98 | 19 | 3 |
| sparingly | 1 | ||||
| (in) the least | 21 | 151 | 67 | 1 | |
| (very) little | 27 | 102 | 69 | 20 | 7 |
| Total | 2,895 | 1,939 | 1,661 | 925 | 138 |
There is a greater variety of target categories for downtoners than for the boosters we investigated in the previous chapter; in addition to the categories modified by boosters, the downtoners in our data (as well as the maximizers) also intensify noun phrases and numerals. While the most frequent targets of boosters and maximizers are adjectives, the most frequent targets of downtoners are prepositional phrases and verbs/verb phrases. Considering that a vast majority of the prepositional phrases occur with a little (see Table 7.9), it is the verbal targets (examples (26) and (27)) that stand out as the specialty of downtoners compared to all other intensifiers. Among the items that are particularly prone to verbs/verb phrases instead of the other word-classes are hardly, scarcely, slightly, (in) the least, and (very) little.
As with maximizers (Chapter 5) and boosters (Chapter 6), we have carried out a semantic classification of the target adjectives and verbs of our downtoners. However, before looking into these uses, the abundance of prepositional phrases as targets of downtoning modification deserves some attention. What we have here are mainly uses with indications of time, for example, after/before/past ten o’clock, or other references to manner, place, or other circumstances relevant to the context of the crime under discussion; instead of giving exact information, speakers find it preferable to hedge their statements. Further typical expressions attested include in liquor, off colour, out of order, over a mile, and under price (for further discussion, see Reference Claridge, Jonsson and KytöClaridge, Jonsson, and Kytö 2021: 73).
We start with the classification of verbs, given the great importance of verbs as targets of downtoner modification. As was the case with maximizers and boosters, we used the process types of systemic-functional grammar for our classification (Reference Halliday and ChristianHalliday and Matthiessen 2004). As explained in previous chapters (notably Chapters 3 and 5), the system comprises the major classes material – mental – relational – behavioural – verbal – existential, the first three with subtypes. Links can be made with the adjective classification, for example, mental and behavioural are included as subtypes in human propensity (cf. Reference DixonDixon 1977, Reference Dixon, Dixon and Alexandra2004; see Chapter 3 in the present volume).
Figure 7.9 shows the proportional distribution of the target verbs of the downtoners. As with the maximizers, the downtoners predominantly modify verbs of the material and mental process types, whereas the second largest process type modified by boosters was behavioural. Verbal, behavioural, and relational verbs show a fairly even distribution for all three intensifier groups, except for maximizers, for which behavioural verbs are rare and relational verbs more frequent. While the target verbs of maximizers and boosters represent only five process types, the target verbs of downtoners represent all six process types, including the existential type, typically be + hardly as in there is hardly a worse woman living (t17520219-24, m).
Figure 7.9 Semantic process types of downtoned verbs/verb phrases (based on Reference Halliday and ChristianHalliday and Matthiessen 2004); proportional distribution
We next present some classified examples of target verbs. While material types typically describe the circumstances of the crime in question (see (35) and (36)), mental, behavioural, and relational types can apply both to description of the crime and to the situation and topics in court; here (37) and (38) refer to the context of the crime and (39) to further detail coming up in court. The verbal class seems to work mostly within the courtroom situation, as in (40); this example is also an instance of speech act hedging. An example of an existential use is given in (41).
(35) Material happening
I picked it up, and then I seized him – he resisted slightly (t18360815-1867, victim, m)
(36) I can only explain the three incisions in the liver by supposing that the stabber partially withdrew the knife and then restabbed (t19070108-29, witness, m, higher)
(37) Mental thinking
I got out of bed, and scarcely believed her at first (t18620707-754, witness, m, higher)
(38) Behavioural
she somewhat revived, and told me, with tears in her eyes, she should be starved to death, for she could not swallow. (t17611021-34, witness, f)
(39) Relational: Having attribute
he wears a suit of black clothes and earrings – he does not resemble you in the least (t18680406-375, witness, m, higher)
(40) Verbal saying
Q. I need hardly ask you if he is acquainted with your Christian name? (t18000917-46, lawyer, m)
(41) Existential
There was hardly a day in the year when I was in town but I went to it. (t17720109-48, victim, m)
We now turn to adjectives and start by presenting the syntactic distribution of adjectives modified by the top five downtoners, that is, a little, hardly, scarcely, slightly, and in the least (for a little we went through a random sample of 200 adjectives); see Table 7.10 and Figure 7.10. For an explanation of the categories ‘attributive’, ‘predicative’, and ‘other’, see Section 5.4.1, including Footnote note 5. The dominance of predicative uses attested in previous research (see Section 5.4) also holds for downtoners: these uses dominate in over 70 or 80 per cent of uses recorded for the five items. Attributive uses appear only with slightly and in the least, in a handful of instances. The results align with the trend presented in previous research, that is, predicative uses clearly dominate in the data. These results align with those obtained for our top-frequency maximizers and boosters, which, apart from one or two items, appear in predicative uses (for details, see Tables 5.7 (Section 5.4.1) and 6.8 (Section 6.4.1, respectively).
Figure 7.10 Syntactic distribution of adjectives modified by the top five downtoners (for a little pertaining to a random sample of 200 adjectives)
Table 7.10 Syntactic distribution of adjectives modified by the top five downtoners
| Predicative | Attributive | Other | |
|---|---|---|---|
| a little | 172 (86%) | 1 (1%) | 27 (14%) |
| hardly | 84 (85%) | 15 (15%) | |
| scarcely | 33 (87%) | 5 (13%) | |
| slightly | 85 (73%) | 12 (10%) | 20 (17%) |
| (in) the least | 51 (76%) | 4 (6%) | 12 (18%) |
As with maximizers and boosters, we classified our adjectives according to Reference DixonDixon’s (1977, Reference Dixon, Dixon and Alexandra2004) semantic categories. The proportional distribution of these is displayed in Figure 7.11, and the frequencies and examples from the data for each semantic class are found in Table 7.11. The top position is occupied by human propensity (43 per cent of the instances), with physical property in the next position (with 26 per cent); these figures come close to those obtained for little in our previous study (41 per cent and 20 per cent, respectively; see Reference Claridge, Jonsson and KytöClaridge, Jonsson, and Kytö 2021: 74). The clear dominance of human propensity across downtoners shows how important it seems to be to consider the (criminal) intentions and actions of those involved in a crime and to downtone certain behaviours in the context of a crime. Note the abundance of adjectives denoting drunkenness (Table 7.11; see also Reference Claridge, Fanego and Rodríguez-PuenteClaridge 2019). The importance of physical property points to the material evidence for a crime that is being discussed in court; similar circumstantial detail is found, for example, in the categories dimension, quantification, position, colour, age, and speed (Table 7.11).
Figure 7.11 Semantic classes of downtoned adjectives (based on Reference DixonDixon 1977, Reference Dixon, Dixon and Alexandra2004); proportional distribution. The classes of the unlabelled bars on the right are position, similarity, colour, age, difficulty, and speed.
Table 7.11 Semantic classification of downtoned adjectives (cf. Reference DixonDixon 1977, Reference Dixon, Dixon and Alexandra2004)
| Types | Tokens | Most frequent types | |
|---|---|---|---|
| Human propensity | 237 | 706 | intoxicated (46), the worse for liquor (42), excited (27), able (22), elevated (19), the worse for drink (19) |
| Physical property | 188 | 438 | open (36), burnt (18), bent (14), lame (12), bloody (10) |
| Value | 36 | 141 | good/better (57), worth (26), odd (7), strange (6) |
| Dimension | 24 | 98 | tall/(the) taller (24), large/larger (23), thick/thicker (9) |
| Quantification | 20 | 88 | better (45), enough (7), half (6), large (6), extra (4) |
| Qualification | 29 | 57 | possible (10), able (7), doubtful (5), necessary (5) |
| Position | 22 | 52 | open (14), higher (9), ajar (4), related (3) |
| Similarity | 7 | 26 | different (15), similar (5), like (2) |
| Colour | 11 | 20 | black/blacker (7), discoloured (4) |
| Age | 7 | 15 | late/later (6), older (4), recent (2) |
| Difficulty | 4 | 7 | easy/easier (4), altered (2) |
| Speed | 5 | 5 | quick/quicker (2), slow (1) |
Comparing the semantic classifications of the target adjectives across our maximizers, boosters, and downtoners is of interest, as it will show which types of target content are being strategically amplified or toned down. Human propensity stands out as the top category for our three types of intensifiers: the figures for the proportional distributions are surprisingly similar (44 per cent for the maximizers, 39 per cent for the boosters and 43 per cent for the downtoners). People involved in the crime discussed or being examined about it are at the centre of discourse. However, while value is the next-best-represented category for the amplifiers (for the maximizers with 18 per cent and the boosters with 25 per cent), it is physical property that occupies this position with downtoners (with 26 per cent of the total of instances). These are properties such as open (thirty-six occ.), burnt (eighteen occ.), bent (fourteen occ.), lame (twelve occ.), and bloody (ten occ.) (see Table 7.11), which the speakers are prone to hedge or tone down in their statements. In fact, the value of the target is toned down to a considerably lesser extent than amplified (with downtoners in 9 per cent of the instances vs. the maximizers with 18 per cent and the boosters with 25 per cent, as just stated). This could reflect the speakers’ reluctance to voice their thoughts on value issues, perhaps owing to possible negative repercussions. As for human propensity, maximizers are used to amplify adjectives such as sober (492 occ.), innocent (226 occ.), satisfied (122 occ.), and sure (103 occ.) (Table 5.8), while downtoners are used to play down the effects of adjectives such as intoxicated (forty-six occ.), the worse for liquor (forty-two occ.), excited (twenty-seven occ.), able (twenty-two occ.), elevated (nineteen occ.), and the worse for drink (nineteen occ.) (Table 7.11). Interestingly, the question of potential intoxication is amplified with sober but frequently downtoned for the opposite meanings.
7.4.2 Targets of Dual-Form Downtoners
The targets modified by dual and zero-form maximizers and boosters were discussed in Chapters 5 and 6, respectively. For an overall view, Table 7.12 summarizes the target categories of the downtoners for which a suffixed/suffixless distinction applies in our data: moderate, scarce, and slight (see Table 7.12). Although most suffixed forms occur with verbs (cf. Reference Nevalainen, Rissanen, Tyrkkö, Timofeeva and SaleniusNevalainen and Rissanen 2013), it is also with verbs/verb phrases that most suffixless forms occur, that is, thirty-eight (or 69 per cent) out of the total of fifty-five occurrences. Of these, thirty-seven instances occur with scarce, as in, for example, They had scarce left the shop and I could scarce get it off. Only stray instances of adjectives as targets were recorded for suffixless downtoners (moderate (one occ.), scarce (two occ.), and slight (four occ.); see Table 7.13). The least-represented targets of our suffixless downtoners are determiners (example (42)), numerals (example (43)), and adverbs (example (44) being the only instance), which all occur with scarce.
(42) It was not very full of people; there were scarce any gentlemen there at the time. (t17611021-33, witness, m, lower)
(43) I staid with them there scarce five minutes, and I left them there. (t18131027-48, witness, m)
(44) The Prisoner Rogers (his Master being my Packer) has done Business for me these five Years, and has had scarce over less than the value of 1000 l. of mine in his Custody, (t17311208-40, witness, m)
Table 7.12 Targets of modification and suffixed versus suffixless forms of downtoners
| Target of modification | Suffixed | Suffixless | Total |
|---|---|---|---|
| Verb/verb phrase | 340 | 38 | 378 |
| Adjective | 164 | 7 | 171 |
| Participle | 52 | 0 | 52 |
| Adverb | 29 | 1 | 30 |
| Prepositional phrase | 29 | 0 | 29 |
| Determiner | 15 | 6 | 21 |
| Noun phrase | 19 | 0 | 19 |
| Pronoun | 15 | 0 | 15 |
| Numeral | 8 | 3 | 11 |
Table 7.13 Adjectives modified by zero-form downtoners (raw frequencies)
| Zero form | Adjective types | Adjective tokens | Adjectives |
|---|---|---|---|
| moderate | 1 | 1 | temperate (1) |
| scarce | 2 | 2 | able (1), well (1) |
| slight | 4 | 4 | acidulous (1), aperients (1), black (1), hollow (1) |
7.4.3 Individual Downtoners: Collocates and Semantic Prosodies
As with our maximizers and boosters, after observing the syntactic and semantic patterns in their use across the different parts of speech, we now turn to the lexical items collocating with the individual downtoners. Here, too, we could reasonably expect the speech situation in the courtroom to prime the use of certain context-specific collocations. For consistency with the discussion of maximizers and boosters, we will focus on the five most common categories of targets of modification and seven of the best-represented downtoners. The top seven downtoners to be included in discussion comprise two diminishers (a little, slightly), four minimizers (hardly, scarcely, (in) the least, (very) little), and one moderator (somewhat).
Table 7.14 summarizes the most frequent target lexemes of the top seven downtoners, that is, a little, hardly, scarcely, slightly, (in) the least, (very) little, and somewhat. As mentioned earlier, prepositional phrases are the leading category, with a little dominating in collocations with after and before (in close to 2,800 instances). These expressions convey the speakers’ wish or need to hedge the information given on the sequence of events and other circumstances relating to the criminal case under discussion and so do collocations with adverbs and the pronoun more, for example, a little further/more/farther. A little also frequently occurs in prepositional phrases with in liquor, and in other expressions conveying the state of intoxication with adjectives (the worse for beer/drink/liquor and intoxicated). These are situation-specific uses, primed by the topics discussed in the courtroom. Hardly and scarcely occur mostly with forms of be able to (close to 480 instances), conveying limitations or restrictions regarding one’s ability, an important component in a witnesses or defendant’s account of the events. Hardly ever (thirty-four occ.) as well as scarcely ever (fourteen occ.) stand out as relatively frequent collocations, conveying hedging tones regarding the circumstantial information given. In the least and have doubt co-occur in collocations such as I have no doubt in the least about it or Yes, I have no doubt about it in the least (twenty-one occ.). Such constructions along with hardly and scarcely occurring with be able to are flexible to some extent, with varying verb forms and word order, while other instances are quite fixed, for example, a little collocating with after/before/better/further/more.
Table 7.14 Most frequent target lexemes of the top seven downtoners
| Prepositional phrase | VerbFootnote a | Adjective | Adverb | Pronoun | |
|---|---|---|---|---|---|
| a little | after … (1,186), before … (868), in liquor (242), over … (85) | be able to (35), hesitate (20) | better (73), the worse for beer/drink/liquor (56), open (36), intoxicated (32), taller (21) | further (131), more (100), farther (94) | more (86), less (3) |
| hardly | at the request (2) | be able to (368) | able (18), worth (14), possible (6) | ever (34), so (17) | any (11) |
| scarcely | of any value (2) | be able to (108) | able (9), light (3) | ever (14) | anything (6) |
| slightly | under the influence of drink (7) | know (20) | different (6), open (6), swollen (6) | upwards (2) | – |
| (in) the least | like … (12) | have doubt (21) | intoxicated (6), the worse for liquor (5) | – | little (1) |
| (very) little | after … (7), before … (4) | think (25) | acquainted (12), worth (10), better (7) | more (14) | more (7) |
| somewhat | like … (4), in liquor (3) | resemble (3) | confused (5), similar (5) | less (2) |
a Verbs are given in their base forms.
Looking into the most frequent target lexemes of downtoners by downtoner category, and including all the types of downtoners in the figures (Table 7.15), moderators tend to occur with positive adjectives and adverbs such as good and well. Diminishers occur most frequently in the earlier-mentioned peak collocations with after/before/better/further/more/farther and in expressions conveying aspects of the state of drunkenness. With the minimizers, be able to and the verb know along with the adverb ever stand out frequency-wise.
Table 7.15 Most frequent target lexemes of downtoners, by downtoner category
| Prepositional phrase | VerbFootnote a | Adjective | Adverb | Pronoun | |
|---|---|---|---|---|---|
| Moderators (tot. 437) | like … (4), in liquor (3) | recover (7) | good (18), large (12), burnt (8), open (8) | well (27), regularly (5), more (4) | less (2) |
| Diminishers (tot. 5,536) | after … (1,186), before … (868), in liquor (243), over … (87) | be able to (36), know (25) | better (75), the worse for beer/drink/liquor (58), open (42), intoxicated (37) | further (131), more (101), farther (94) | more (86), less (3) |
| Minimizers (tot. 1,901) | like … (14), in liquor (8), after … (7), before … (5) | be able to (482), know (134) | able (29), worth (26), acquainted (12), sensible (11), possible (9) | ever (47), so (18), more (15) | any (16), anything (8) |
a Verbs are given in their base forms.
As appears in Chapter 6 (see Section 6.4.3), boosters can be intensified by other boosters, and in Chapter 5 we saw that even maximizers could be intensified by other maximizers although the instances recorded in our data were only a handful (see Section 5.4.3). However, no downtoners modified by other downtoners could be found in our data. Downtoners can nevertheless intensify and be intensified by boosters and maximizers, for example, the downtoner a little modifies the maximizer too in our data in twenty-four cases, and little is modified by so in thirty-nine cases (see Section 6.4.3).
7.5 Chapter Summary
The present chapter has described the representation of downtoners across the three subcategories we have distinguished, that is, moderators, diminishers, and minimizers, and their modificational and collocational patterns in terms of syntax and semantics. Based on information drawn from the HTOED, Reference Quirk, Greenbaum, Leech and SvartvikQuirk et al. (1985), and Reference Huddleston and PullumHuddleston and Pullum (2002), and other sources, altogether 19 downtoners were included in the investigation, totalling 7,874 relevant occurrences, with the incidence figure of 32.2 per 100,000 words. Compared with boosters (194.8) and maximizers (38.8), downtoners remain the least-represented category of intensifiers used in the Old Bailey courtroom. Of the three subcategories, the diminishers and minimizers showed a declining trend, while the moderators were on the rise. Moderators and, as shown in Chapter 5, maximizers are the two categories of intensifiers bucking the declining trend displayed by boosters, diminishers, and minimizers. This points to their serving the particular needs of the communicative situations where language users found themselves in the courtroom. It may also reflect stylistic changes in the recording practices adopted by the scribes across the period (see Section 9.2.1, on the time variable). As for PDE, downtoners occur more in academic writing than in the spoken trial register (at least as regards those of our items included in Biber et al.’s investigation (Reference Biber, Johansson, Leech, Conrad and Finegan1999: 556–7)), which we take to point to the different need felt for the use of hedging devices in the two registers, both at the formal end of the formality continuum of each respective mode. At any rate, the relatively little amount of hedginess attested means that the weak communicative style proposed by Reference O’BarrO’Barr (1982) and discussed in Section 3.5 is not a characteristic of the Old Bailey discourse.
The top five items, a little, hardly, scarcely, slightly, and (in) the least comprised 91 per cent of our downtoners data. Even the two top-frequency downtoners, a little and hardly, represented 79 per cent of the data, the former being five times more frequent that the latter. While these two items were in decline across the period, a number of items, notably slightly and fairly, were on the rise, and they still seem to be on rise in spoken PDE. Overall, narrowing down to courtroom discourse, our PDE data pointed to a rise in the trials register for the items shared by the OBC and BNC1994, but the relatively scanty representation of downtoners in the latter leaves little room for generalizations.
Considering forms of downtoners, suffixless forms were only meagrely represented. Of the nineteen downtoners included in the study, only three occurred in suffixed and suffixless forms, that is, scarcely, moderately, and slightly. No more than fifty-five suffixless occurrences of these were attested in the data; the best-represented form was scarce with forty-nine instances, the majority of them found for the first two subperiods. Maximizers and boosters displayed a considerably more varied use of the suffixless forms than downtoners.
Turning to the targets of modification, downtoners presented a profile different from maximizers and boosters, their most frequent targets being prepositional phrases and verbs. The major bulk of the prepositional phrases occurred with a little, and the items occurring particularly with verbs included hardly, scarcely, slightly, (in) the least, and (very) little. The semantic classification of our targets of modification (Reference DixonDixon 1977, Reference Dixon, Dixon and Alexandra2004) showed that target adjectives mostly conveyed aspects of human propensity and physical property, highlighting the importance of the human being and circumstantial detail in the discourse around the criminal cases and in the mitigation of the whole process. As for the semantic classification of target verbs (Reference Halliday and ChristianHalliday and Matthiessen 2004), the circumstances of the crime under discussion were typically conveyed by material types, while the description of the crime and topics treated were conveyed by mental, behavioural, and relational types.
The emphasis on the participants and material circumstances of a given crime also characterizes the frequent lexical collocations of downtoners and their targets of modification. Among the very frequent collocations were a little before/after and a little better/further/more/farther. Expressions conveying the state of drunkenness were also common, for example, a little in liquor/drink/intoxicated. Hardly and scarcely occurred frequently with forms of be able to and the adverb ever. Both fixed and flexible collocations were represented in the data.
Having focused on the inventories of our maximizer, booster and downtoner forms and their structural and semantic characteristics in descriptive terms in Chapters 5, 6, and 7, we now turn to the possible influence of our speakers’ background on intensifier use in the Old Bailey courtroom. Accordingly, Chapter 8 approaches the use of intensifiers from a bird’s-eye perspective across the diachrony and the groups of speakers distinguished in the OBC.
