Intensifiers in competition: The picture from early English medical writing (pre-print)

Submitted. Published in Taavitsainen, Irma & Päivi Pahta (eds.) 2010. Early Modern English Medical

Texts: Corpus Description and Studies. Amsterdam / Philadelphia: John Benjamins.

Intensifiers in competition: The picture from early English medical writing

Belén Méndez-Naya and Päivi Pahta

1. Introduction

Intensifiers are words like very or really.1 They are degree adverbs, scaling upwards the

degree of the quality expressed by the item they modify, as in very interesting or really

hot.2 They are versatile in their expressivity and communicative functions: speakers can

use them to foreground and emphasize the points they make, adjust the strength of their

claims, convey their attitude, values and emotions, and try to influence their audience in

1 A version of this paper was presented in the 31st ICAME Conference in Giessen, May 26-30, 2010. We

thank the participants for their comments and suggestions on the study. We are also grateful to María José

López-Couso for her perceptive and helpful comments on an earlier draft of this study. Belén Méndez-

Naya is grateful to the following institutions for generous financial support: Spanish Ministry for Science

and Innovation and European Regional Development Fund (grant HUM2007-60706), and the Autonomous

Government of Galicia (grants 2008-047, 2009-047 and INCITE-08PXIB204016PR).

2 We use the term “intensifier” here only for those elements which scale an established norm upwards (see

e.g. Huddleston and Pullum et al. 2002: 585). These items are also called “amplifiers” (see e.g. Quirk et al.

1985: 589; Biber et al. 1999: 554, who use “intensifier” as a cover term for all degree adverbs), and

comprise boosters (e.g. very) and maximizers (e.g. completely). For a discussion of the uses of the term, see

Méndez-Naya (2003: 373-374).

various ways (see e.g. Labov 1984; Partington 1993; Lorenz 1999). They also index

social identities and group-memberships, and are a strong indicator of shifting norms and

practices in a speech community (see Ito and Tagliamonte 2003: 257).

According to Brinton and Arnovik (2006: 441), intensifiers comprise one of the

major areas of grammatical change and renewal in English, especially from the Early

Modern English period onwards (Peters 1993). Their tendency to constant competition

and change (Bolinger 1972: 18) has received considerable attention in the literature, and

has been shown to be sensitive to both linguistic (e.g. type of head, adjective class) and

extralinguistic determinants (e.g. age, sex, social class, type of interaction) (see Macaulay

2002; Hopper and Traugott 2003: 122-123; Ito and Tagliamonte 2003; Méndez-Naya

2008; Nevalainen 2008; Tagliamonte 2008). Moreover, the use of intensifiers in Present-

day English is register-specific: frequent use of intensifiers is typically associated with

informal contexts of discourse, such as conversation, while more formal types of

communication, especially academic writing, do not favour these devices particularly

(Biber et al. 1999: 565). However, even in registers where intensifiers in Present-day

English are relatively rare, their analysis may reveal interesting patterns of variation and

change over time and across genres. Earlier studies have shown, for example, that

academic genres in Present-day English differ in their intensification practices both in the

repertoire and frequency of intensifiers (see Swales and Burke 2003; Pahta 2006b).

Research on intensification practices in Middle English medical texts indicates similar

variation between different varieties of medieval scientific writing (Pahta 2006a).

This case study examines intensifier use in Late Middle and Early Modern

English medical writing, covering the period c.1375 to 1700. Focusing on the medical

register, we investigate the competition between full, well and right, which earlier

research has established as the most salient intensifiers in Middle English, and very,

which became the dominant intensifier from the sixteenth century onwards (Fettig 1934:

2; Mustanoja 1960: 319, 323-324, 326-327; Nevalainen 1997: 174-175; Adamson and

González-Díaz 2004; Méndez-Naya 2004). In addition to these four competing items, we

have included the item highly in the analysis, which has been shown to play a prominent

role in Present-day English medical writing and whose emergence thus merits inclusion

here (Pahta 2006b). Furthermore, we have paid attention to dual form adverbs3 in the

analysis, and for this reason the forms fully and high are also included in the study;

rightly and verily, though, are excluded since they do not occur as intensifiers in the data

(cf. OED s. vv. rightly,4 verily; Nevalainen 1997: 177-178 for similar observations).

Intensifiers can occur with different types of head, modifying adjectives, adverbs,

participles and verbs (Bolinger 1972: 15). Several earlier studies indicate that the

majority of intensifiers in Present-day English occur as adjectival modifiers; according to

some scholars, this is their prototypical syntactic function (see Bäcklund 1973: 279;

Allerton 1987: 16; Lorenz 2002: 144). For this study, in addition to adjectival modifiers,

we have also analysed intensifiers co-occurring with adverbial heads.

At a general level, we chart the chronology of intensifier shifts in early English

medical writing. At a more specific level, we analyse the syntactic and semantic

3 We refer here to adverbs which show variation between suffixed (-ly) and suffixless (zero) forms, e.g.

pairs such as slow/slowly (see Nevalainen 1994: 139).

4 OED Online (draft revision June 2010) does include a degree reading for rightly in dialects, usually in

negative constructions, providing examples from 1741 onwards (OED s.v. rightly adv. 4.a).

behaviour (e.g. syntactic function, syntactic structure) and communicative functions of

the items at issue in order to disclose the factors underlying and affecting the patterns of

development in their competition. We also explore whether intensifier use can be

employed as a tool to characterize different varieties of medical writing, and comment on

practices in individual texts.

The outline of the study is as follows. Section 2 introduces the material used for

the present study, the Corpus of Early English Medical Writing, and the varieties of

medical writing which we have selected for analysis from it. Section 3 provides the big

picture, with the general trends in intensifier shifts in the time-span covered. The

diffusion of the intensifiers under study across the different semantic classes of adjectives

is presented in section 4. In section 5, we focus on the emergence of very, the intensifier

par excellence from the sixteenth century onwards, and the retreat of right, its only true

competitor from 1500. Section 6 deals with the most prominent communicative contexts

featuring our intensifiers, and finally the main conclusions are presented in section 7.

2. Material

The material for our study comes from the first two components of the Corpus of Early

English Medical Writing, i.e. Early Modern English Medical Texts (EMEMT, 1500-

1700) and its medieval counterpart, Middle English Medical Texts (MEMT, 1375-1500).

We analyse material from two different varieties of medical writing, representing broad

text categories that we call “treatises” and “recipes”. Treatises are expository texts that

provide an opportunity to compare the findings with present-day scientific writing. Our

material for treatises includes the category of SPECIALIZED TEXTS in MEMT and its

equivalents in EMEMT, the categories of GENERAL TREATISES (category 1) and SPECIFIC

TREATISES (category 2). The medieval treatises represent the early emergence of scientific

medical writing in English rather than Latin, and are mostly translations from Latin. They

include learned theoretical texts on physiology and natural philosophy, as well as tracts

focusing on a specific illness or a field of specialization, or particular method of

prognosis or treatment. A wide range of specific topics are included, such as

ophthalmology, reproduction, gynaecology and obstetrics, urinoscopy, phlebotomy,

epilepsy, syphilis and the plague. The EMEMT samples from GENERAL TREATISES come

from texts that are intended to give a systematic account of the whole field of medicine.

SPECIFIC TREATISES cover a wide range of topics in a variety of styles, the common

denominator being a focus on an individual disease or method. This corpus category is

divided into five subcategories based on the focus of the text (see pp. 000-000). The

target audiences of treatises in both MEMT and EMEMT range from academic specialists

to the widest popular readership.

Our second variety, “recipes”, is a key genre in the history of medicine and medical

writing, containing texts that provide instructions on how to prepare and administer

medicines, or inform the readers of the uses and effects of medicines. In EMEMT, we

have analysed the category of RECIPE COLLECTIONS, also including texts like herbals that

provide information on the therapeutic properties of substances. In MEMT, the analysis is

based on the category of REMEDYBOOKS AND MATERIA MEDICA, which also contains

various types of health guides.

All in all, our corpus material amounts to over 1.4 million words; 61.5% of the

material comes from treatises and 38.5% from recipes (see Table 1). In the analysis we

have divided the material into six subperiods (MEMT 1-2, EMEMT 1-4), 50 years each,

apart from the earliest subperiod, which contains all early MEMT material up to 1449. In

EMEMT the division of the material into the subperiods is straightforward, as the year of

printing indicated in the books can be used for periodization. In the MEMT material the

period division is only tentative, since in many cases it is impossible to date very

precisely the manuscripts from which the texts have been edited; furthermore, a late

manuscript may in fact reflect the linguistic practices of an earlier source copy, which

complicates periodization further.

Table 1. Material: word counts of treatises and recipes per subperiod in MEMT and

EMEMT.

Word count (WC)

Subperiod

WC

treatises

WC recipes Total

MEMT 1 1375-1449 20,982 148,338 169,320

MEMT 2 1450-1499 66,277 67,932 134,209

EMEMT 1 1500-1549 59,012 46,581 105,593

EMEMT 2 1550-1599 181,338 92,047 273,385

EMEMT 3 1600-1649 237,643 70,432 308,075

EMEMT 4 1650-1700 314,621 125,323 439,944

Total 879,873 550,653 1,430,526

3. General trends

The corpus material contains 2,162 instances of full/fully, high/highly, right, very and

well, when used as adjectival and adverbial modifiers; 64.8% of these are found in

treatises and 35.2% in recipes. The raw frequencies in treatises and recipes in each

subperiod are shown in Table 2. The relative frequency of intensifiers in the whole

dataset is 15.17 items per 10,000 words, with a higher frequency in treatises (15.98) than

in recipes (13.87).

Table 2. Frequency (N) of intensifiers as adjectival and adverbial modifiers in treatises

and recipes per subperiod in MEMT and EMEMT.

N intensifiers

(N/10,000)

Treatises Recipes Total

Subperiod

MEMT1 1375-1449 19 (9.06) 122 (8.22) 141 (8.33)

MEMT2 1450-1499 40 (6.04) 26 (3.83) 66 (4.92)

EMEMT1 1500-1549 95 (16.15) 37 (8.06) 132 (12.60)

EMEMT2 1550-1599 345 (19.00) 152 (16.56) 497 (18.18)

EMEMT3 1600-1649 352 (14.94) 150 (21.34) 502 (16.41)

EMEMT4 1650-1700 550 (17.56) 274 (22.00) 824 (18.82)

Total (N/10,000) 1,401

(15.98)

764 (13.87) 2,162 (15.17)

From a diachronic perspective there is a clear rise in the frequency of intensifiers

in the transition to the early modern period. This rapid increase in intensification in the

first half of the sixteenth century is clearly seen in Figure 1. The rise attested in our data

is in line with a general increase in intensification strategies in the Early Modern English

period that has been noted e.g. by Peters (1994: 271-273), who also reports a

simultaneous increase in the inventory of intensifiers. Figure 1 also shows that whereas

the increase is observed in both varieties we examined here, the patterns of increase are

different. The rise is earlier in treatises, where the frequency of intensifiers grows from

roughly 6 per 10,000 words in the second half of the fifteenth century to over 16 in the

first half of the sixteenth century, and continues to rise to 19 in the second half of the

century. In the seventeenth century the frequencies in fact become somewhat lower. A

similar rise in the volume of intensification happens later in recipes, which lag behind in

this shift: there is a clear increase in intensification already during the first half of the

sixteenth century in recipes too, but in the second half of the century this increase gains

momentum. The frequency of intensifiers continues to rise in recipes even in the early

seventeenth-century data, when it reaches its highest peak at 22 intensifiers per 10,000

words.

Figure 1. Intensifiers in treatises and recipes in MEMT and EMEMT (N/10,000 words)

Table 3 contains the raw figures of individual intensifiers in the whole dataset,

divided into the six subperiods. The results show that some of the items included in the

analysis occur with a very low frequency in the material; high and highly, for example,

are rare in all subperiods.

Table 3. Individual intensifiers in MEMT and EMEMT; raw figures.

Subperiod

Intensifier

1375-

1449

MEMT1

1450-

1499

MEMT2

1500-

1549

EMEMT1

1550-

1599

EMEMT2

1600-

1649

EMEMT3

1650-

1700

EMEMT4

Total

Full 73 26 5 12 1 1 118

Fully 8 2 3 4 0 0 17

High 0 1 0 0 8 0 9

Highly 0 0 0 0 0 5 5

Right 36 24 16 25 3 3 107

Very 1 3 107 443 484 807 1,845

Well 23 10 1 13 6 8 61

Total 141 66 132 497 502 824 2,162

The chronological shifts in the competition of the intensifiers begin to emerge in

the comparison of normalized frequencies. Figure 2 shows the frequency of individual

intensifiers in the data per 10,000 words in each subperiod. Full decreases quite steeply

from the medieval to the early modern period, while right and well decline steadily. The

first instances of very appear in treatises in the medieval period, and in the first half of the

sixteenth century its frequency increases rapidly.

Figure 2. Intensifiers in MEMT and EMEMT (N/10,000).

The findings in medical writing can be compared with results from earlier corpus-

based research. Figure 3 illustrates intensifier shifts in the data from the Helsinki Corpus

of English Texts, a general-purpose multi-genre corpus of c. 1.5 million words extending

from Old English to Early Modern English.5 The figure is based on an earlier study on

the competition between full, well and right (Méndez-Naya 2004); very was analysed for

this study (cf. also data from Nevalainen 1997: 174).

5 For information on the Helsinki Corpus data see e.g. Rissanen et al. (1993), or the Corpus Resource

Database (CoRD) at http://www.helsinki.fi/varieng/CoRD/.

Figure 3. Intensifiers in the Helsinki Corpus of English Texts.

When the findings in the Helsinki Corpus data from the Late Middle and Early

Modern English periods are compared with those from medical writing in MEMT and

EMEMT, the big picture that emerges is similar: full, well and right decline towards and

beyond the transition from Middle to Early Modern English, while very emerges and

rapidly becomes the most frequent intensifier in both datasets in the early sixteenth

century. However, there are some discrepancies. For example, full is significantly less

frequent in Late Middle English medical writing than it is in the Helsinki Corpus. It

seems likely that the nature of the data provides the answer here, since the Helsinki

Corpus as a multi-genre corpus contains for example some verse texts, in which

intensifiers, full in particular, are frequently used as fillers to the meter.6 Right is also

more frequent in the Helsinki Corpus, and this, we think, may be due to genres like

letters, which contain conventionalized address forms making use of right, e.g. honorifics

like the right excellent (Peters 1994: 275, 278; Raumolin-Brunberg 1996: 168; Sánchez-

Roura 2000: 79-80). In medical writing these are rare.

When the diachronic development of very in medical writing, seen in Figure 2, is

compared with the overall developments in the frequency of intensifiers shown in Figure

1, we can see clearly that it is precisely the rise of very that causes the dramatic rise in

intensification. The curve of very is similar in shape to that of the whole dataset.

When the frequency of very in the whole dataset is compared with its frequency in

the two texts categories, treatises and recipes, it becomes evident that the increase is

slower in recipes, but eventually very becomes even more frequent in the latter than in

treatises (Figure 4). This slow onset may reflect the fact that many early printed recipe

collections were actually based on materials that had already circulated in manuscripts

and therefore may be more conservative or even archaic in their discursive practices.

During the period examined in this study, however, competition in the medical market

became heavy and the question of selling books became crucial, so it is possible that this

6 Set combinations like full glad and full well survive in poetry until Present-day English (see Bolinger

1972: 22). Note also the peak in the line corresponding to well in M2 (1250-1350) in Figure 3, which is

also probably due to the prevalence of verse texts in this subperiod of the Helsinki Corpus. Monosyllabic

intensifiers like full, eall and well have also been reported to be rather frequent in Old English poetry

(Peltola 1971: 656).

need to persuade buyers and the need to sell is also reflected in intensification practices in

later recipe texts.

Figure 4. Very in MEMT and EMEMT: treatises vs. recipes (N/10,000)

4. Exposing diffusion

Previous research has established a correlation between the growing delexicalization and

grammaticalization of a given intensifier and its width of collocation (cf. Partington

1993: 183; Ito and Tagliamonte 2003: 268). Wider diffusion of intensifiers tends to

correlate with an increase in token frequency, a concomitant of grammaticalization (e.g.

Brinton and Traugott 2005:26). Contrariwise, receding intensifiers tend to occur with an

ever decreasing range of collocates, until eventually they are found only in restricted

collocations which may become fossilized over time (e.g. Méndez-Naya 2003: 377 and

references therein).

In order to study intensifier shifts in more detail, we have studied the distribution

of the intensifiers according to their head, and given that adjectival heads are prevalent in

the corpus we have focused on the diffusion of the intensifiers under analysis across

adjective classes.

Before we turn to the corpus data, a few words are in order concerning the

distinction between adjectival and adverbial heads, since this is not always as

straightforward as it may seem. Consider examples such as those illustrated in (1a-d).

(1) a. stop the bottle very close (Grey, Choice manual, 1653: 10;

recipes)

b. of each take half a pound; shred them very small (Woolley,

Compleat servantmaid, 1677: 39; recipes)

c. Take two ounces of Chene-roots, first slit very thin (Grey, Choice

manual, 1653: 11; recipes)

d. cover him very warm to sweat an hour (Elkes, Approved medicines,

1651: 12; recipes)

In these and similar examples, two interpretations are possible: close, small, thin

and warm7 can be analysed as adjectives functioning as predicative complements of the

respective objects in a complex transitive construction, or, alternatively, as zero adverbs

realizing the function of manner adjuncts. This indeterminacy is also noted by the OED,

in which, for instance, the entry for small adv. mentions that such instances may be

understood as illustrating “perh[aps] the adj[ective] used predicatively”. Similar

comments are found for the other forms as well. Ambiguity between adjectives and

adverbs is also increased by the acceptability of adverbs in predicative structures with

verbs of appearance and perception such as look, seem, sound, smell or appear in earlier

English (e.g. it smells sweet vs. it smells sweetly; cf. Nevalainen 1994: 145; see also

Gisborne 2000, who terms this “the attributary construction”; and Killie 2000, who

argues for the adjective analysis). Even though we acknowledge the ambiguity, we

decided to count examples such as those in (1a-d) (82 in all) as cases of adjectival heads,

realizing that a different interpretation of their nature would affect our results

considerably.

Adjectival heads in our corpus have been classified according to the taxonomy

proposed by Dixon (1977), and further refined in Dixon (2004: 3-5), where a total of 13

adjectival classes are proposed: DIMENSION (‘big’, ‘small’, ‘short’, ‘wide’, ‘deep’); AGE

(‘new’, ‘young’, ‘old’); VALUE (‘good’, ‘bad’, ‘lovely’, etc.); COLOUR (‘black’, ‘white’,

etc.); PHYSICAL PROPERTY (‘hard’, ‘soft’, ‘heavy’, ‘wet’, ‘well’, ‘sick’, ‘tired’); HUMAN

7 Other ambiguous forms include clean, fine, hard, strong and thick.

PROPENSITY (‘jealous’, ‘happy’, ‘kind’); SPEED (‘fast’, ‘quick’, ‘slow’); DIFFICULTY

(‘easy’, ‘difficult’, ‘hard’); SIMILARITY (‘like’, ‘unlike’, ‘similar’, ‘different’);

QUALIFICATION (‘definite’, ‘true’, ‘probable’, ‘likely’, ‘common’); QUANTIFICATION

(‘all’, ‘many’, ‘some’, ‘few’); POSITION (‘high’, ‘low’, ‘near’); cardinal NUMBERS and

‘first’, ‘last’.

We now move to the results the corpus analysis for each of the intensifiers in

focus. Special attention will be given to the four major intensifiers full, well, right and

very. The data on diffusion across adjective classes for these four items are given in

Figures 5-8 below and discussed individually.

4.1 Full/fully

Full 1375-

1449

MEMT1

1450-

1499

MEMT2

1500-

1549

EMEMT1

1550-

1599

EMEMT2

1600-

1649

EMEMT3

1650-

1700

EMEMT4

Dimension

Age

Value

Colour

Phys.

property

Human

propensity

Difficulty

Qualification

Quantification

Figure 5. Diffusion of full across different semantic types of adjectival heads in MEMT

and EMEMT

There are 118 tokens of the intensifier full in the corpus, with a clear prevalence of

adjectival over adverbial heads (103 examples vs. 15).8

As can be seen in Figure 2, above, the normalized frequency of full decreases by

over 50% from MEMT1 (4.37) to MEMT2 (1.94), although, as also observed in other

studies (Fettig 1934: 93-94; Méndez-Naya 2004), full remains the most common

8 Earlier studies have reported that full shows a preference for adverbial heads in Old (Ingersoll 1978: 158)

and Middle English (Peters 1993: 159, 163). The Helsinki Corpus data analysed in Méndez-Naya (2004)

show the dominance of adverbial heads with full until M3 (1350-1420), when adjectives take over. The

data from the corpus of medical writing (1375 onwards) are thus consistent with the pattern found in the

Helsinki Corpus material.

intensifier in our Middle English data. This decrease runs parallel with the restriction of

full to less and less adjectival types, from nine different classes in MEMT1 to just two

from 1500 onwards.

After 1552, the only adjective occurring with full in the data is ripe, as in (2). Six

out of the eight examples of full ripe derive from one and the same text, i.e. Walter

Bailey’s Three kindes of peppers (1588). With adjectives like ripe, which denote a

precise value of the property involved,9 full is to be interpreted as a maximizer (‘totally,

completely’).

(2) and that black pepper was full ripe (Bailey, Three kindes of peppers, 1588:

B2v; treatises)

The restriction of full to fewer and fewer collocate-types is also corroborated by evidence

from adverbial heads. The only combination that occurs more than once is full well

(EMEMT2).

Fully is found in 17 examples, 6 with an adjectival head, and 11 with adverbs. The

adjectives involved are bounded (ripe 2 examples, open, cretyk), as in example (3), with

9 Such adjectives, which may be construed in terms of either-or (e.g. true, ripe) or may represent the

ultimate point on a scale (e.g. brilliant) have been labelled bounded adjectives (Paradis 2001: 50-53; 2008:

322-326).

just one exception (fully thick). From a semantic point of view, they all denote a physical

property.

(3) when the compound is fully rype (Bailey, Mithridatium, 1585: B5v; treatises)

The most common adverb collocating with fully is so (9 examples), in structures of

double intensification (e.g. fully so round, fully so long).

4.2 High/highly

High is only found with adjectival heads in our material (9 examples) In eight of these, it

is associated with the colour adjective red as in (4):10

10 These examples could allow for a different interpretation: high can be analysed an adjective in a

premodifying string (cf. OED s.v. high adj. 10a, “Of qualities etc., of great amount, degree, force or value;

great, intense, extreme, strong, forcible, violent”). This interpretation is supported by the existence of

predicative examples in which high and red are coordinated (e.g. as also the Water, being high and red; is

likewise red and high of colour Brian, Pisseprophet, 1637; treatises) and premodifying strings in which

high is not the leftmost element in the string (of an intense, high, darke, red colour; red high colour Brian,

Pisseprophet, 1637; treatises). The presence of the adjective high in the leftmost position of premodifying

strings may have prompted the intensifier reading (cf. Adamson 2000 on the reanalysis of the adjective

lovely as an intensifier). Following the OED (s.v. high adv. III.10.b, “with an adj[ective] = Highly, to a

great degree”, which gives precisely an example of the combination high red (1663 BOYLE Colours (J.), A

high-red tincture), we have opted for the intensifier analysis.

(4) Urine is for the most part of an high red colour (Brian, Pisse-prophet, 1637:

35; treatises)

The remaining example involves the physical property adjective mighty (hihe

myghty, referring to wine).

Highly appears with adjectival collocates rather late in the material analysed.11

The five examples in which highly occurs with adjectival heads are contained in texts

from the last subperiod (EMEMT4, 1650-1700); the earliest one, reproduced as (5), is

dated 1694. Therefore, the spread of highly, which has a prominent role as adjective

modifier in Present-day English medical writing in MEDICOR, a corpus of English

medical writing of the 1990s (Pahta 2006b), must have taken place after 1700. Most of

the examples of highly, with all types of head, occur in treatises, which ties in well with

Pahta’s (2006b) finding that highly is more common in professional genres like research

articles and textbooks than in writings aimed at a more general readership.

(5) Whether it is not highly reasonable to believe, that Choler is the cause of the

Disease (Chamberlen, Practice of physick, 1694: 82; treatises)

11 In the corpus of medical writing highly first appears with verbs and participles (cf. also Fettig 1934: 114).

Similarly, the MED (s.v. heighlī 3.a) only gives examples of highly with verbs. Fettig (1934: 115) provides

an early example (Orm 4612) with an adverbial collocate. Examples with adjectival heads are first found in

the fifteenth century (see also Peters 1993: 120).

4.3 Well

well 1375-

1449

MEMT1

1450-

1499

MEMT2

1500-

1549

EMEMT1

1550-

1599

EMEMT2

1600-

1649

EMEMT3

1650-

1700

EMEMT4

Dimension

Value

Phys. property

Human

propensity

Quantification

Figure 6. Diffusion of well across different semantic types of adjectival heads in MEMT

and EMEMT

There are 61 examples of the intensifier well in the corpus; 49 of them (79%) occur with

adjectival collocates. It is interesting to note that well can modify adjectives in the

comparative form (see Fettig 1934: 186; OED s.v. well adv. IV.19).12

The MEMT

material yields two relevant instances of comparative heads, one of which is given in

example (6).

12 Full could also modify a head inflected for the comparative (Fettig 1934: 94, 101), but it did not appear

as frequently as well in this environment. The corpus yielded no such examples.

(6) latynge o þe same veyne for þe same seknes, bod wele better & mare sykyr

undyr þe arm (Daniel, Liber uricrisiarum 1, 1375-1449: 155; treatises)

As was the case with full, the decrease in frequency observed for well in the Early

Modern English data (from 0.75 in MEMT2 to 0.1 in EMEMT1, see Figure 2) is

accompanied by a restriction in the range of adjective collocates. The adjective class that

is most stable over time with well in the data is human propensity. In the EMEMT

material, when intensifier well was simply residual, we can observe the consolidation of

some common collocations like well worth, illustrated in example (7) (3 examples) and,

with adverbs, such as well near, as in (8) (8 examples; 6 of these in the same text, by

Monardes from 1580) (see OED s.v. well adv. IV.16.b). Near is the only adverb

occurring with well in the EMEMT corpus.

(7) And these two properties are well worth the observing (Packe, Mineralogia,

1693: 3; treatises)

(8) as also well neere all the dogmatistes (Galen, Methodus medendi, 1586: 48r;

treatises)

4.4 Right

Right is recorded in 107 corpus instances. As was the case with the intensifiers discussed

so far, the percentage of adjectival collocates is very high (81.3%; 87 examples vs. 20

examples of adverbial collocates). Example (9) below illustrates right as an adjective

modifier.

(9) for they be ryght fearse & angry with them/ that they bere them away (Lytell

boke of the .xxiiii. stones, 1528: A3r; recipes)

Regarding frequency and diffusion, right shows a different picture from the one

described for full and well. While the decrease in the frequency of full and well is rather

dramatic in the corpus of medical writing, right remains more or less stable until the first

half of the sixteenth century, when it experiences a slight drop. As mentioned above,

there is a close relationship between frequency and diffusion, thus the fact that right

keeps its place would account for the stability in its range of collocates until 1550, seen in

Figure 7. In EMEMT2 (1550-1599) we witness a restriction in the adjective classes

collocating with right, since only value and physical property are available. Moreover,

after 1550 value adjectives, and more specifically, adjectives expressing “strongly

appreciable qualities” (Bolinger 1972: 52; see also Méndez-Naya 2008: 47), like

excellent, good and honourable, become the default collocates of right (see examples 17

and 26). Only 3 of the 21 examples of right with an adjectival head in EMEMT2

collocate with adjectives other than value.

Again, restriction is also to be perceived in the case of adverbial heads; in the

EMEMT material, six out of the nine examples involve the combination right well.

Right 1375-

1449

MEMT1

1450-

1499

MEMT2

1500-

1549

EMEMT1

1550-

1599

EMEMT2

1600-

1649

EMEMT3

1650-

1700

EMEMT4

Dimension

Value

Colour

Phys. property

Human

propensity

Quantification

Figure 7. Diffusion of right across different semantic types of adjectival heads in MEMT

and EMEMT

The data from the corpus of medical writing offer a different picture in the

development of right from that obtained with the Helsinki Corpus of English Texts (see

Figure 3). In the Helsinki Corpus, even though there is a peak in frequency in M4 (1420-

1500), roughly corresponding to MEMT2 (1450-1499) in the corpus of medical writing,

c. 40% of examples of adjectival heads correspond to stereotyped honorific expressions

such as right honourable, right singular, right excellent high and myghty prynce (cf.

Méndez-Naya 2006: 156, 2007: 204), which may suggest that right was no longer an

“active” intensifier after the second half of the fifteenth century. This is followed by a

marked decrease in frequency in E1 (1500-1570). In the corpus of medical writing, by

contrast, the decrease in the frequency of right takes place later, and, more revealingly,

right still enjoys some variability until 1550, which is manifest in the token/type ratio of

its adjectival collocates (12 tokens/10 types), and the survival of negative adjectives (e.g.

fearse & angry, example (9) above, EMEMT1). Honorifics do not abound in the corpus

data: 1 example in MEMT2 (right dere); 1 example in EMEMT2 (right reuerend); 1

example in EMEMT3 and 2 examples in EMEMT4 (right honourable).13

As seen in section 3, the discrepancy between the Helsinki Corpus and the early

English medical writing data can be attributed to the different make-up of the two

corpora, the Helsinki Corpus being a multi-genre collection.

4.5. Very

13 The stereotyped honorifics right honourable and right reverend are still used in Present-day English (see,

e.g., LDOCE Online s.vv. Right Honourable, Right Reverend).

very 1375-

1449

MEMT1

1450-

1499

MEMT2

1500-

1549

EMEMT1

1550-

1599

EMEMT2

1600-

1649

EMEMT3

1650-

1700

EMEMT4

Dimension

Age

Value

Colour

Phys. property

Human

propensity

Difficulty

Similarity

Qualification

Quantification

Number

Figure 8. Diffusion of very across different semantic types of adjectival heads in MEMT

and EMEMT

Very is by and large the most frequent intensifier in the corpus, with 1,845

attestations of which 1,422 (77%) correspond to adjectival heads.

In previous studies it has been observed that very is first found in its new function

as a modifier of adjectives in the second half of the fourteenth century (e.g. Fettig 1934:

25). This is also the date suggested by the Helsinki Corpus data, with the two earliest

examples occurring in M3 (1350-1420). The dating of the first example of very

modifying an adjective in the corpus of medical writing is consistent with the findings

reported in earlier studies. The first occurrence is found in MEMT1 (1375-1449), and is

reproduced as (10) below:

(10) sette hit ouer a softe fire til þe vin aigre wexe verry rede. (Rupescissa,

Quintessence, 1375-1449: 191; recipes)

It has been surmised in the literature (e.g. Mustanoja 1960: 326-7) that the

intensifier use of very originates in examples where it was an attributive adjective in a

complex premodifying string (e.g. the verray parfit blisfulnesse, Chaucer Boethius

?a1425 (c1380); Helsinki Corpus). Once reanalysed as an intensifier, very spread from

attributive adjectives to predicative adjectives, as in (10) above, and adverbs, as in (11)

below, which reproduces the first instance of very as an adverbial modifier in the corpus

of medical writing. Again, this is consistent with the data extracted from the Helsinki

Corpus, where the first example of very with an adverbial head occurs in M4 (1420-

1500).

(11) he shuld enter in to a dry stew & ther with mesurable swetyng with owte

anguishe and after verrey whele dryed with clene lynen clothis (Torrella,

Tretece of the pokkis, 1450-1499: 470; treatises)

As seen in Figures 2 and 4 above, the frequency of very rises steeply in

EMEMT1, more specifically in texts dated after 1540, that is, more than one century after

the appearance of the earliest examples of the intensifier. This would illustrate what Mair

(2004: 129) terms the “delayed-increase pattern” in grammaticalization, according to

which a marked increase in frequency should be seen as “a delayed symptom of earlier

grammaticalisation” (2004: 138). The spectacular rise in the frequency of very comes

together with its rapid diffusion across adjective classes: from one adjectival class in

MEMT1 and MEMT2, to ten different classes in EMEMT1. The diffusion is also marked

in the case of adverbial collocates, as reflected in the type/token ratio: there are 23

examples of adverbial heads in this period, featuring 18 different adverbs.

5. Close-up on very vs. right

Given that it is precisely the rise of very that causes the general increase in intensification

in the material, it is worth taking a look at this in more detail. We therefore divided the

sixteenth-century material in EMEMT into twenty-year slices. In doing so we have also

been able to detect some text-characteristic tendencies and have identified frequent users

of the incoming item. We compare the development with one of the outgoing variants,

right, which holds its position in the competition longer than the other intensifiers.

As shown in Figure 9, the dramatic rise in the frequency of very takes place in the

third twenty-year period, corresponding to 1540-1559, and is particularly noticeable in

treatises. More specifically, this increase is to be put down to three texts, Eucharius

Roesslin’s Byrth of mankynde (1540), with a normalized frequency of 3.5 per 10,000

words, illustrated in example (12), Boke of chyldren (1546) by Thomas Phayer

(N/10,000: 2.6), exemplified in (13), and especially Bartholomeus Cocles’s Epitomye of

phisiognomie (1556), with over 5 examples every 10,000 words, illustrated in (14).

(12) these be tokens that it sholde be verye weake (Roesslin, Byrth of mankynde,

1540: 15v; treatises)

(13) If the childe be not very young, the mawe of a leueret, dronke with water

(Phayer, Boke of chyldren, 1546: T4r; treatises)

(14) That head which is very voluble, or lightly turning here and there (Cocles,

Epitomye of phisiognomie, 1556: A5v; treatises)

Recipes lag somewhat behind in the spread of very, but experience a marked

increase in the period from 1560 to 1579. Again, one text is mostly responsible for this

rise: Girolamo Ruscelli’s Secretes of Alexis of Piemount (1562), with 29 tokens of the

intensifier (2.8/10,000). Of the items under analysis, only very features in this text,

illustrated in (15).

(15) A worke well approued, verye profitable and necessarie for euerye man.

(Ruscelli, Secretes of Alexis of Piemount, 1562: title page; recipes)

As mentioned above, remedies are characterized by a slower adoption of very,

while they still show relatively frequent use of the receding intensifier right. They only

catch up with treatises in the use of very after 1580, and this is concomitant with the

decrease of right. Paying attention to the first twenty-year periods, we have identified two

conservative texts, in which the frequency of right is still relatively high. One of them is

Roger Bacon’s Waters artyfyciales (1550) (7 examples of right and no instances of very),

illustrated in (16), and Thomas Gale’s Antidotarie (1563), exemplified in (17). In

Antidotarie we find a similar distribution of right (10 examples) and very (8 examples).

However, even though right is slightly more common, its collocates are less varied. Thus,

it is only found with adjectival heads. There are 10 examples, but only 3 different

adjectives, the positively loaded comfortable, excellent and good. This is characteristic of

a receding intensifier, while the incoming very modifies both adjectives and adverbs (8

tokens, and 7 types).

(16) This water stronge is of ryghte greate and meruaylous vertue (Bacon, Waters

artyfycialles, 1550: C1v; recipes)

(17) Thys is a ryght excellent medicine in woundes of the head. (Gale, Antidotarie,

1563: 14r; recipes)

Figure 9. Very and right in EMEMT in 1500-1599: twenty-year periods.

6. Communicative contexts in focus

The analysis of intensifiers in their microlevel communicative context in the data

provides information about the domain-specific communicative functions of intensifiers

in medical discourse across time, while a comparison between the different corpus

categories reveals possible differences in intensification practices in different subgenres.

In a qualitative analysis of the communicative uses of intensifiers in the MEMT and

EMEMT data reported here, we have paid special attention to three types of contexts that

have been identified in earlier research on medieval and present-day medical writing, i.e.

descriptive, instructive and evaluative contexts (see Pahta 2006a, 2006b). Our analysis

indicates that intensifiers are prominent in these uses in the present material as well. In

these contexts, intensifiers are used in partly overlapping functions for scaling up or

accentuating various descriptive, instructive or evaluative meanings. There are obvious

large-scale diachronic shifts in the use of specific items here, as very replaces the other

items in all these contexts during the first half of the sixteenth century, but the actual

communicative uses of intensifiers do not change dramatically.

Descriptions and categorizations are central in the organization of knowledge in

any field of science, and in medicine, precise descriptions and clear distinctions are also

crucial for successful medical practice. In our material, intensifiers are frequently used in

contexts providing descriptive information, underlining specific qualities and distinctions.

This practice is common in both treatises and recipes. Intensifiers can be used for

highlighting particular characteristics of, for example, body parts, as in example (18), or

physiological processes, sicknesses or symptoms, as in (19). They are also used for

emphasizing specific qualities of substances, for example, for diagnostic purposes, as in

(20), or in introducing new substances to the readers by comparison with familiar

substances, underlining distinctions or similarities, as in (21).

(18) she may not lo~ge sytte styll/ and her eyes be very small (Seynge of uryns,

1525: C4r; treatises)

(19) þe i3en swellen but litil, ne þei ben not ful hoot ne ful rede (Gilbertus

Anglicus, Compendium medicinae, 1375-1449: 43; recipes)

(20) Uryne that shineth raw and right brighte, if the skyn in the bottome shine not

(Partridge, Treasurie of commodious conceits, 1573: F7r; recipes)

(21) Our people doe vse it there to pitche theyr Shippes withall, for it is well neere

lyke vnto Tarre (Monardes, Ioyfull newes, 1580: 6r; treatises)

Intensifiers are also frequently found in instructive passages that give advice on how to

prepare or administer medicines, treat patients or perform medical procedures. This

practice is more common in recipes than in treatises; examples are provided in (22)-(24).

In some instructive passages intensifiers crop up frequently, as in example (24), where

the recurring use of right emphasizes the correct procedure.

(22) then receaue the feate as they come forth/ & binde them with some fayre

lynnen clothe/ and so tenderly and very softly loose out the byrth tyll all be

come forth (Roesslin, Byrth of mankynde, 1540: 23r; treatises)

(23) Take notice that in drawing forth Acid red Juices … have your hands very

clean, and take an especial care that no Iron come near it (Charas, Royal

pharmacopoea, 1678: 53; recipes)

(24) Take an iren pricke or a poyntel that is right sharpe… put hit into a glasyn

vaisshel and close hit right faste… Receipue hit into a potte of glas þat is

right clene… Thenne take hit oute and grynde hit right smalle on a marble

stone (Rupescissa, Quintessence, 1375-1449: 55; recipes)

Perhaps the most common type of communicative context for intensification in our data

is evaluation, where intensifiers are used in argumentative and persuasive functions. Our

qualitative assessment suggest that the use of intensifiers in various types of evaluative

functions increases considerably in the early modern period, and it seems that it is

precisely this increase in argumentative, persuasive and promotional discourse that

causes the overall rise in the volume of intensification in early modern medical writing.

Intensifiers frequently co-occur with evaluative adjectives like profitable, excellent and

good, illustrated in examples (25)-(29), for emphasizing the importance or usefulness of

particular measures, treatments or substances for the patient’s health, for a specific group

of patients, or for curing particular ailments.

(25) holdyng in of thi breth helpith moch and is ful profitable (Gouernayle of

helthe, 1450-1499: 48; recipes)

(26) who so shuld make of the water brennynge of wyne of the decocktion with

these floures y~ shuld be a ryght excellente thynge for the syckenesses

abouesayde (Bacon, Waters artyfycialles, 1550: A4r; recipes)

(27) Thys vnguent is verye good to drye vp vlceracions (Gale, Antidotarie, 1563:

27r; recipes)

(28) The soote of butter is very effectuall against watering eyes (Bright,

Sufficiencie of English medicines, 1580: 31; recipes)

(29) A very safe Clister to be used by either Man or Woman (Woolley,

Gentlewomans companion, 1673:179; recipes)

Intensifiers also occur with heads denoting more negative qualities. In such cases they

typically serve to emphasize criticism or warnings against the use of particular substances

or medical procedures, as in examples (30)-(32).

(30) giving Alkalies in this case must be at least superfluous, if not highly

pernicious (Colbatch, Physico-medical essay concerning alkaly and acid,

1698: 103; treatises)

(31) it prove not only injurious to them, but very prejudicial to such as might have

received benefit from them (Chamberlen, Practice of physick, 1694: 12;

treatises)

(32) Purging in this Disease (when it is simple) seldom or never doth any good, as

also Blood-letting, but both are very injurious, as many have found to their

great prejudice. (Couch, Praxis catholica, 1680: 51; treatises)

More hedged warnings or critical comments can also make use of intensifiers combined

with a positive adjective in sentences containing negation, as in (33).

(33) Tho’ it may cleanse and dry Ulcers, and these in the Mouth also, yet I think it

not a very convenient Medicament to be applied often to the Teeth, because all

sorts of Vitriol are apt to blacken, discolour, or foul the Teeth, &c. (Salmon,

Pharmacopoeia Bateana, 1700: 6; recipes)

7. Conclusion

The big picture that emerges from medical writing in terms of intensifiers in competition

in the transition from the medieval to the early modern period is in many ways similar to

that described in earlier research (e.g. Méndez-Naya 2004). The most frequent

intensifiers of the Middle English period, full, right and well decline and give way to an

entirely new variant, very, which first appears in the late fourteenth century and rapidly

replaces the earlier variants as the most popular intensifier in the course of the sixteenth

century. Highly, a commonly used intensifier in Present-day English medical writing,

occurs rarely in the material.

On a more focused view, some new details become visible. For example, the

frequency of the intensifiers full, right and well in the Middle English period is somewhat

lower in the single-register data from the medical corpus than in the multi-genre Helsinki

Corpus. This seems to suggest that already in this early period, the vernacular registers of

medical and scientific writing had their own distinctive character in intensification

practices, and in comparison with some other types of writing, such as letters or poetry,

for instance, did not particularly favour the use of intensifiers. The decrease in the

frequency of full, right and well in the transition period from Middle to Early Modern

English is also paralleled by a gradual restriction in their diffusion across different types

of collocates. In our Early Modern English data, full and well occur only in very few

restricted collocations, such as full ripe and well worth, while adjectives expressing

strongly appreciable qualities become the default collocates of right, as in right profitable

and right excellent.

The dispersion of very in the sixteenth century, the force and speed with which it

emerges and spreads to a wide range of syntactic and semantic functions, is striking, and

contributes to a general rise in intensification in medical writing. Its diffusion in

adjectival heads across different semantic classes is surprisingly rapid. Its emergence is

also interesting in terms of textual diffusion: it first appears in the texts of a few “heavy

users” and later on becomes more evenly distributed across the material, although there is

great variation in intensification practices between individual texts throughout the whole

period. In the sixteenth century, treatises, many of them new translations or new

compositions in English, lead the change, while recipes, mostly going back to originals

already circulating in English manuscripts, lag behind and still show relatively high

frequencies of the outgoing right. The frequency of very in treatises is at its highest in the

late sixteenth century, whereas in recipes the use of very only reaches its peak later, in the

seventeenth century.

The communicative uses of intensifiers remain relatively stable in early English

medical writing over time. The analysis confirms three partly overlapping contexts of

discourse where intensification is frequent. Intensifiers are commonly used in various

descriptive contexts, emphasizing specific features of objects and concepts in focus and

highlighting distinctions between them. They are also frequent in instructive contexts,

foregrounding and underlining advice on medical practice. Perhaps the most common

context of intensification in early medical writing is that of evaluation and assessment.

Here intensifiers frequently co-occur with evaluative adjectives in persuasive and

argumentative functions, emphasizing the importance or usefulness of particular medical

treatments or substances, or advocating against the use of harmful or dangerous ones. In

such functions intensifiers become an important rhetorical device in the marketing

discourses promoting useful advice literature in the medical marketplace.

Our study has shown that the analysis of register-specific patterns of variation and

change in the competition of intensifiers using the corpora of Middle English Medical

Texts and Early Modern English Medical Texts yields an interesting picture which may

help qualify the general findings of studies drawn from general-purpose, multi-genre

corpora like the Helsinki Corpus of English Texts. The results show that even within this

one register, which in the extensive period examined in the present study was also

affected by large-scale language-external changes, involving dynamic vernacularization,

fundamental changes in domain-specific theory and practice and a dramatic shift in the

medium and circumstances of disseminating texts, the pace of change in intensification

practices can be significantly different in different genres of writing. Several earlier

studies making use of the corpora analysed here suggest that innovations in medical

writing first take place in the more learned layers of writing. In the case of intensifiers the

picture seems more complicated: the first texts making frequent use of the incoming

variant very, although representing the category of treatises, do not fall within the most

learned texts in the material. This, we believe, reflects the nature of the linguistic practice

examined, since intensifiers, as stated in our introduction, are a particularly expressive

means, with their use governed by a complex variety of linguistic and extralinguistic

factors. The study indicates that the communicative choices of a few “very innovative”

individuals are crucial in the rise of very in sixteenth-century medical writing, and it

would certainly be interesting to move further to a more fine-grained level of analysis,

with a detailed study of the role of individuals in intensifier shifts. Another topic to be

addressed in our future research is intensification practices in other varieties of

contemporary medical writing, including the Philosophical Transactions and its new

cutting-edge genres of early scientific writing.