Degree and Related Phenomena in the History of English: Evidence of Usage and Pathways of Change

1. Degree Phenomena: Previous Research, Classifications, and Analytical Approaches

Degree constitutes an area that is very much prone to change in its linguistic realizations, and this makes it a very interesting object of historical and diachronic research. The chronological coverage of long-term and short-term developments in previous research is uneven; there are more studies on Early Modern English and on Present-Day English change in progress than on other periods. In fact, previous research has so far only scratched the surface of the wide area of degree, with a distinct focus on intensifiers. Degree items and constructions form a group with fuzzy boundaries, as many linguistic realizations are possible. Adverbial intensifiers (e.g., hardly, somewhat, very, entirely) and (gradable) adjectives (e.g., the finest silk) are a prominent area, but these are joined by any word with suitable semantics (a perfect match) or contrast to other words (disaster versus debacle or failure), intensificatory word-formation types (e.g., arch-pretender, megabitch, ice-cold), semantic or lexical repetition (e.g., tiny little, happen over and over and over again), individual forms from minor lexical categories (e.g., this big, such a fool), special constructions (e.g., exclamatory what a night!), as well as idioms/fixed expressions (e.g., dead as a dodo, before you can say Jack Robinson), and the rhetorical devices of conventional or creative hyperbole (cf. Claridge 2011) and litotes (e.g., wait for ages, not unreasonable).

The present special issue, which is divided into two issues of the Journal of English Linguistics, aims to highlight the wide and varied repertoire of expressions conveying intensification in the history of English. In this introduction, we first survey previous work in the field, summarize some of the central classifications of intensifiers, and outline the analytical frameworks used in recent work on the topic. We then describe the contents of the special issue in overall terms before introducing the individual studies and their significance to the study of intensification in the history of English.

General as well as theoretical treatments of intensification and degree phenomena over the past few decades have included the seminal monograph by Bolinger (1972), classifications presented in reference grammars such as Quirk, Greenbaum, Leech, and Svartvik (1985) and Huddleston and Pullum (2002), and more recent cognitive explorations, such as Paradis (2008). The by now vast literature on grammaticalization has also provided important input, of which one might mention, by way of exemplification, Traugott (2008), also touching on construction grammar, and Traugott (2010) on subjectification. Attention has also been paid to contexts and functions, albeit not yet sufficiently so in historical studies. Degree phenomena are sensitive to speaker needs in different contexts—that is, their specific use and distribution depends to a large extent on the medium and genre. Spoken interactive contexts have been highlighted as an important locus for intensity and are therefore also commonly the focus of modern intensifier research. In Biber’s (1988:102-108) text typology, for example, the lexical degree types of amplifiers (e.g., completely, highly), general emphatics (e.g., so, really), and hedges (e.g., almost, more or less) stand out as characteristic uses on the dimension “involved versus informational production.” Thus, they are generally more frequent in conversation, but also in written interaction in letters. Degree indicators of various kinds are also found in polite interactions and the models that attempt to theorize such interactions. Brown and Levinson’s model (1987:104-106, 145, 176-177) refers to them explicitly in the positive politeness strategies 2 (exaggerate interest, approval, sympathy with H[earer]) and 3 (intensify interest to H), and in the negative strategies 2 (question, hedge) and 4 (minimize the imposition). Leech’s (2007:182, 195) model encapsulates degree/quantity already by its paired high/low value constraints and by listing hedges (a tiny bit) and intensifiers (really, terribly) as prominent linguistic resources. Given Biber’s (1988) “involved” characterization, it is not surprising that degree indicators have also been treated in the context of emotive communication. Caffi and Janney (1994:344, 358, and passim), for example, identify the psychological dimension Activity/Arousal with the linguistic category Quantity/Intensity and list in their typology of emotive devices the category of Quantity (distinction more/less), comprising devices like intensifying wh-pronouns (what a day!), emphasizing adjectives (a real/complete/total catastrophe), degree adverbs (hardly, very, absolutely), and repetition (we’re happy, really happy). All emotional or generally highly subjective contexts thus also prominently feature degree devices. A further notable context is that involving persuasion. Martin and White’s Appraisal Theory (2005:135-153) may be mentioned here, which includes a degree category labeled “graduation.” This category allows speakers to express the intensity of their attitudes towards people, objects, or situations and their degree of commitment to the utterance. Their force-indicating devices include all types of intensifiers, lexical degree items, and repetition. The classical proponent of persuasion is of course ancient rhetoric, which includes many degree-indicating devices in its repertoire, all under the umbrella of attention-attracting emphasis for the sake of aiding audience comprehension and especially conviction (Lauerbach 2010). Claridge (2006) shows that intensifying and heightening, rather than comparative, uses of the superlative are indeed more common in both involved and persuasive nineteenth-century texts (e.g., drama, fiction, letters; debates, trials). Interestingly, there are no systematic empirical (synchronic) studies focusing on intensification across genres or registers, or on its occurrence in particular involved or persuasive contexts.

There are various historical treatments of intensifiers, the one with the widest range being Peters (1993), who treats the large class of boosters in the whole Helsinki Corpus (Old English-Early Modern English) as well as in parts of the London-Lund Corpus of Spoken English and Miller and Brown’s Scottish data collection for Modern English. There is considerable research focusing on both the Early Modern period (e.g., Rissanen 2008) and recent or Present-Day English, also including sociolinguistic aspects (e.g., Macaulay 2002). Late Modern English, by now a linguistic period in its own right, has been neglected with regard to intensifiers; we have found few papers that are directly relevant (mostly dealing with one specific author), namely Brorström (1987), Clifford-Amos (1995), Cacchiani (2006), and, partly, Nevalainen and Rissanen (2013). Méndez-Naya and Pahta (2010) also consider a group of intensifiers in the specialized register of historical medicine. What the last two and virtually all research on historical intensification have in common is their focus on written discourse. However, access to the Old Bailey Corpus (OBC), a recent resource for speech-related language from the Late Modern English period, has already generated some work, e.g., Bernaisch (2014) and Claridge, Jonsson, and Kytö (2020, forthcoming).

Various studies have looked at individual or, at most, pairs of intensifiers, e.g., the papers collected in Méndez-Naya (2008), or Margerie (2011). Margerie (2011) is further of interest as she considers the multi-word intensifier to death, bringing in the constructional dimension, while most other historical work restricts itself to single-word items. A thorough view of constructions in the degree area is offered by Brems (2011), who investigates the whole synchronic area of “N of N” constructions, i.e., a lot of X, a bit of X, and many similar forms.

Unsurprisingly, much empirical work on intensifiers in English has been based on definitions and classifications presented in the above-mentioned treatments by Bolinger (1972), Quirk, Greenbaum, Leech, and Svartvik (1985), and Paradis (1997, 2008). Quirk, Greenbaum, Leech, and Svartvik (1985), drawing on Bolinger (1972), distinguish amplifiers, which are subdivided into maximizers (perfectly, entirely), denoting an upper extreme of the scale, and boosters (very, so), marking a high degree. In addition to amplifiers, Quirk, Greenbaum, Leech, and Svartvik (1985) distinguish downtoners, which fall into diminishers (a little, moderately), denoting a low/small extent on the scale, and minimizers (little, barely), marking an extremely low level. Further, Quirk, Greenbaum, Leech, and Svartvik (1985) identify approximators (almost, nearly) and compromisers (quite, rather): both lower the degree slightly, but, more importantly, they also work as hedges. Refinements to this classification have also been presented in individual studies, such as King (2016) who, in addition to boosters and maximizers, introduces a category for excessivizers (overly). Paradis (2008), while adhering to Quirk, Greenbaum, Leech, and Svartvik (1985) to a great extent, distinguishes two major categories, totality modifiers and scalar modifiers, the former comprising maximizers (absolutely, completely) and approximators (almost, nearly), and the latter boosters (extremely, highly), moderators (fairly, pretty), and diminishers (a little, slightly). This typology is based on Paradis’ (2008) “boundedness hypothesis,” whereby totality modifiers combine with bounded meanings (e.g., completely/almost full) and indicate an absolute boundary, while scalar modifiers combine with unbounded meanings (e.g., extremely/fairly/a bit odd) indicating the scalarity of the expression in question (Paradis 2008:326). One and the same item may indicate either bounded or unbounded meaning depending on the context (e.g., dead right/easy), which may reflect change in progress from bounded to unbounded degree modifier. Furthermore, intensifiers of a certain type may coerce a corresponding reading when combining with an adjective of another type, e.g., a bounded one in completely good (Paradis 2008:326). The boundedness notion will be referred to or relied on also in a number of papers in this special issue.

Considering the empirical aims of most recent work on intensifiers, the influential analytical frameworks applied have included the corpus linguistic framework and approaches such as historical sociolinguistics, historical pragmatics, as well as language variation and change. Many studies have been informed by a pragmatically based theory of semantic change (e.g., Traugott & Dasher 2001), grammaticalization theory, and approaches to constructional change (e.g., Hilpert 2013), in order to produce a comprehensive picture of intensifier development.

In terms of methodology, English intensifiers lend themselves to item-based corpus searches when they appear as lexical items and in frame constructions. Indeed, the considerable number of English multi-genre and specialized historical corpora available to researchers has been made use of in most studies over the past few decades. The output files containing raw data usually require manual screening to enable further quantitative and qualitative analyses aimed at revealing syntactic and lexico-semantic patterns, and changes in them across the centuries.

Historical sociolinguistic approaches (e.g., Nevalainen 2006), paying attention to speaker variables in linguistic change, have proved especially well-suited for being combined with corpus linguistic methodology. They have been successfully applied to the study of linguistic developments in, e.g., the Corpus of Early English Correspondence (CEEC), A Corpus of English Dialogues 1560-1760 (CED), and the Old Bailey Corpus (OBC). Such approaches assume that principles from modern sociolinguistics can be transferred to historical contexts. In the case of intensifiers, the social variables considered have included predictors such as genre and speakers’ sociolinguistic properties (gender, social status, and speaker roles). For instance, past research has indicated that women, lower-ranking, and non-standard speakers may have leading roles in various types of change (e.g., Ito & Tagliamonte 2003). Modern sociolinguistic research on intensifiers has favored the variable approach, investigating all intensifiable adjectives for the presence or absence of an intensifier (e.g., Tagliamonte 2008). Some historical research, such as Pahta (2006a, 2006b) and Bernaisch (2014), has equally concentrated on the adjective context as the most common environment, although not necessarily with a sociolinguistic aim.

Another influential framework has been historical pragmatics (e.g., Culpeper & Kytö 2010), which allows one to examine sociohistorical and pragmatic aspects of historical texts in their sociocultural communicative contexts, including attention paid to, for instance, conversational principles, speech acts, politeness strategies, face concerns, and speakers’ need to hedge or assert their utterances (e.g., Pahta 2006a, 2006b; González-Díaz 2014; Claridge, Jonsson & Kytö forthcoming). Processes of change that are relevant to the development of intensifiers such as grammaticalization and lexicalization are at the heart of historical pragmatics. Overall, the mental processes of language users have thus been important in explaining changes in the area of intensifiers, such as subjectification and metaphorization, and their effects are usually visible in the lexical and syntactic combinations found in language use (cf. Bybee 2006). Frequency, in other words repeated occurrence or co-occurrence of items, will lead to entrenchment in the mental inventory of language users and in time to new lexical and grammatical units (cf. Hilpert 2013). Quantitative data can therefore give us indirect clues about trajectories of change in intensifier usage.

In sum, the development of individual intensifiers and intensifiers as a class raises fundamental issues in language change. What changes are there in frequencies across time, genres, and language user groups, how are new items recruited, how are earlier items or uses of items revived, and how do items and their use transform across the centuries? Of equal interest is how formerly common intensifiers such as most, right, and full gradually fall out of use or persist in certain fixed expressions. Some intensifiers lie dormant for some time or at least pass beneath speakers’ awareness, only to reappear at some point: an example is intensifying all (as in “John is all wet”; Buchstaller & Traugott 2006:345), which has been around since Old English but seems “new” to many observers in modern American English, as pointed out by Buchstaller and Traugott (2006:346). Beyond these questions lies even more uncharted territory regarding the diachronic aspects of the manifold degree phenomena mentioned above.

2. The Contents of the Present Special Issue: A Bird’s-eye View

The present special issue highlights the various faces of intensifier usage and change in the history of English. Most research on English intensifiers has so far tended to focus on items belonging to the word class adverb. While intensifying adverbs appear prominently in this double special issue (Issue 1: Aijmer; Brinton; Claridge, Jonsson & Kytö; Hiltunen; Stange), we cast the net wider than usual by including in Issue 2 studies which target intensifying adjectives (González-Díaz), determiners (Vartiainen), synthetic forms (Méndez-Naya), and syntactic constructions (Ghesquière & Troughton; Ghesquière; Vartiainen).

Further, as indicated in section 1, the targets modified by degree items can be very diverse, going far beyond the prototypical case of intensified adjectives. This is where the present special issue (Issue 2) also covers a broader range of issues than has usually been the case: various papers focus on the noun phrase (González-Díaz; Ghesquière) or on whole constructions (Ghesquière & Troughton; Vartiainen). Even in prototypical adjectival contexts, the difference between predicative or attributive uses may play a role (Vartiainen; Aijmer).

The “degree” domain also intersects to some extent with “focus,” “intensity,” and “emphasis,” both with regard to identical items used (e.g., really) and with regard to conceptual overlap. The nature of this overlap may possibly be elucidated by “close-ups” of such multi-functional items, where the network of uses and the sequence in which the different meanings arose may give valuable clues as to the pathways of and borderlines between these semantic areas. One of the contributions probes the degree nature and development of exclamative what-phrases (Ghesquière & Troughton), while intensifying this and that are highlighted in another (Vartiainen). The modification of non-gradable items, as happens with the so-called “GenX so” (Stange), is also of interest in this context.

Questions of language change are paramount in this field, as was evident in section 1. Is there really as much change as assumed or is there rather long-lasting stability with only some ebb and flow? In the case of change, it is the sources that are of interest, but also the pathways taken from source to a newly emerging form, which often crucially include the changing of collocational patterns. This is thus an important focus of research also in several papers in the present special issue (Aijmer; Brinton; Méndez-Naya; Vartiainen). Exactly when and where a new form/construction arises and how (fast) it subsequently spreads (or not) remains under-researched for many forms. The reasons why forms may lie dormant for a long time (Vartiainen) or wane and re-emerge (Aijmer) are thus of great interest. The impact of generic contexts is another factor in the rise and fall of degree items (González-Díaz; Hiltunen; Vartiainen), as are sociolinguistic aspects (Aijmer; Claridge, Jonsson & Kytö; Stange). The mechanisms and processes of grammaticalization, delexicalization, and semantic bleaching and subjectification have played a prominent role in researching this field. These mechanisms are also in focus here, with, e.g., incipient and static grammaticalization (Brinton) being highlighted.

A wide coverage of data is necessary to allow a multi-faceted perspective on degree types and uses. The contributions span the whole history from Old English (Méndez-Naya) to Present-Day English, including its varieties (Stange). Corpus data are taken both from general corpora (e.g., the Penn Parsed Corpora of Historical English, the Corpus of Historical American English [COHA]) and specialized corpora (e.g., Late Modern English Medical Texts [LMEMT], the Corpus of American Soap Operas). Such a spread is important for understanding the mechanisms of language change, as degree items may arise, spread, and wane both in the language as a whole as well as in specific situational or generic contexts depending on the particular needs of users in different environments.

To highlight the multi-faceted nature of degree phenomena and at the same time to enhance the coherence of the special issue, the papers have been organized into two groups comprising (i) those on individual or groups of intensifying adverbs (Aijmer; Brinton; Claridge, Jonsson & Kytö; Hiltunen; Stange), and (ii) those highlighting wider perspectives on degree and intensification (Ghesquière & Troughton; Ghesquière; González-Díaz; Méndez-Naya; Vartiainen).

3. Issue 1: Individual or Groups of Intensifying Adverbs

Issue 1 of this double special issue comprises five articles highlighting developments in individual intensifying adverbs or groups of them. In her investigation of well in its intensifying uses in current British English, Karin Aijmer looks into the intriguing history of an item which is used as an intensifier in early English, mainly in fixed combinations such as well aware, well open, and well pleased, but which after a long period of hibernation starts to gain in popularity in the speech of adolescents in corpus data from the early 2000s, appearing with adjectives such as good and happy. After examining some anecdotal evidence, pointing to the re-emergence of the use in the 1980s, Aijmer turns to what she defines as “diachronic comparable corpus linguistics,” i.e., comparisons between the data found in the British National Corpus (BNC) from the 1990s (BNC 1994) and its recent counterpart BNC 2014, compiled some twenty years later. The author shows that well is used increasingly over time, by a wider range of speakers and combining with more adjective types. Most adjectives convey the speaker’s affect or evaluation of something as good, bad, surprising, or expected. The corpus evidence indicates that adolescents use the new intensifying well as a fad or a vogue word for its newness value. Aijmer ascribes the rise of the use to a grammaticalization process.

In the second study, Laurel Brinton turns to nearly, the approximator downtoner, which means ‘almost, all but, virtually,’ and the earlier variants based on the same root, nigh, nighly, near, and next (to). The pathways of these spatial adverbs are interesting in that they follow the reverse pathway postulated for other degree adverbs, as the degree modifier uses with adjectives and participles precede the degree adjunct uses with verbs. For tracing the developments, Brinton uses evidence from the Corpus of Contemporary American English (COCA), the Corpus of Historical American English (COHA), A Corpus of Early American Literature, and the Corpus of Late Modern English Texts (CLMET3.0). The results of the study show, for instance, that while nigh functions as a downtoner in Old English and continues doing so across the subsequent centuries, its use drastically declines at the end of the Late Modern English period, persisting longer in American English than in British English, to the first decade of the twentieth century. The decline in frequency is accompanied by the use of nigh becoming restricted to modifying adjectives, impossible, in particular. The zero form near is replaced by nearly, which, contrary to the received view of the development of related zero and -ly forms, specializes in downtoning uses, near functioning as a local adverb or preposition.

Downtoners have been a rather neglected area in previous literature. In addition to Brinton’s study of nearly and its related forms, the degree and non-degree uses of the adverbial little are in focus in the contribution by Claudia Claridge, Ewa Jonsson, and Merja Kytö. In its degree uses, little appears as a diminishing and minimizing intensifier and in its non-degree uses as a durative (e.g., “I waited a little”). The data are drawn from the socio-pragmatically annotated Old Bailey Corpus (OBC, version 2.0), which enables researchers to trace patterns of usage across the speaker groups. In addition, attention is paid to the semantics of the item and the items it is used to modify. In terms of frequency, little is shown to decline across the period covered by the OBC (1720-1913), and the item is used much more as a downtoner than as a durative, and primarily as a diminisher. Most instances occur with the indefinite article (a little), leaving zero article or otherwise modified instances (e.g., very little) in a minority. The items modified mostly comprise adjectives and prepositional phrases, although other types are also attested. The use made of little by the different speaker groups is analyzed by applying a negative binomial regression model to the data, which enables the authors to estimate the contribution of each predictor (time, gender, social class, and speaker role) while holding the other predictors constant. It turns out that while little is used more in the courtroom by lower social ranks and non-legal, lay speakers for hedging purposes, it is also used to lend precision and assertiveness to utterances among men, in particular.

The Late Modern English period, notably the eighteenth century, receives further attention in the study by Turo Hiltunen of three high-frequency amplifiers in medical writing. Medical writing undergoes important changes in the Late Modern English period, and intensifiers deserve attention as a feature characteristic of the genre. The data are drawn from the Corpus of Late Modern English Medical Texts (LMEMT), which contains seven text categories of medical writing, allowing insights into sub-register variation. The study aims at exploring the frequencies and co-occurrence patterns of the three amplifiers, and at linking them to discourse functions and meanings characteristic of eighteenth century medical texts. Two types of syntactic structures are included in the analysis, i.e., the instances where so, very, and too precede adjectival heads (very good) or adverbial heads (too much). The frequency analyses are complemented by analyses of co-occurrence preferences of words and grammatical constructions (colligations or collostructions) by measuring the attraction and reliance values in question. As in present-day medical writing, the frequency analyses highlight the relevance of the situational characteristics and communicative purposes of the registers and sub-registers in usage patterns attested in the data. Despite the fact that the field of medicine experiences major disciplinary and institutional changes in the period, stability rather than dramatic change characterizes the development of the three amplifiers.

The final article in issue 1, by Ulrike Stange, is devoted to a study of GenX so as a modifier of verb phrases as in I was so hoping you’d be here. These instances are a fairly recent addition to the repertoire of intensifier uses, and their emergence has been traced to informal American English. The investigation aims at providing robust evidence to meet the claims presented in previous introspective and small-scale corpus-based investigations. The data are drawn from the Corpus of American Soap Operas, which yields over 1350 relevant instances. A number of the claims presented in previous empirical studies can be corroborated. As for sociolinguistic predictors, GenX so is shown to be used mostly by (younger) women, and regarding linguistic factors, it is found in all types of declarative sentences, before or (mostly) after auxiliaries and modals; it combines with non-auxiliary verbs much more frequently than with modals; apart from the going to-future, which favors the use of second person subjects, all other verb phrases prefer first person subjects; characteristic uses include positive expressions such as I’m so looking forward, I so appreciate, I so love, and I was so hoping, and affirmative uses overall have a wider collocational range than negative uses. Not all claims presented in previous introspective studies can be corroborated by the corpus findings, however.

4. Issue 2: Wider Perspectives on Degree and Intensification

The second issue assembles five papers that go beyond the usual intensifier class and focus on degree-indicating deictics, adjectives, pronouns, and affixes as well as on constructional issues. In contrast to all other contributions in this special issue, the first article concerns items that are of the grammatical/functional type, namely wh-forms. Subjectivity and emotivity play a prominent role in Lobke Ghesquière and Faye Troughton’s contribution on the origins of what-exclamatives as constructions conveying a degree meaning. The degree meaning of what is that of a booster, scaling towards a very high degree on an open-ended or unbounded scale. Degree what typically modifies evaluative qualities or size and quantity measures, and expresses surprise or unexpectedness together with further related emotions. The involved nature of the construction is also shown by its common co-occurrence with direct speech, interjections, and terms of address. Taking their cue from Siemund’s (2017) hypothesis that so-called elliptical wh-exclamatives like what a mess are in fact an independent construction labeled “interrogative degree modification,” from which full exclamative clauses are syntactically expanded, they proceed to investigate this suggestion across the whole history of British English with the help of the three Penn-Helsinki corpora, CLMET 3.0, and WordbanksOnline. They find the expansion hypothesis impossible to uphold, as ellipted exclamatives are not found earlier than full ones and are even much rarer than full ones in the written data up to the present. However, they cannot rule out a possibly parallel development of exclamative clauses on the one hand and an independent degree modification construction on the other. Their own tentative suggestions for the origin of the degree exclamative are rhetorical questions and/or French influence. The suggestion of rhetorical questions with what as the origin is based on Bolinger’s (1972) identification-to-intensification pathway and the facts that rhetorical questions appear earlier in the data, carry undertones of unmet expectations, are thus emotive, and themselves invite a degree reading. This reading is downtoning rather than upscaling, however, which presents a problem. It is possible to hypothesize French influence since both full and elliptic exclamatives are attested from the early twelfth century onwards as potential models for English speakers, but more research is clearly needed to substantiate this hypothesis.

If what is not a prototypical degree item, neither is good, which is investigated in the contribution by Lobke Ghesquière. Whereas Paradis (2000) argued that only bounded and extreme adjectives are used for nominal intensification, good is clearly neither but has nevertheless developed degree uses. This study analyzes prenominal uses of good from Old English to Present-Day English on the basis of the quotations database of the OED. It finds four relevant uses of good, namely as an adjectival intensifier (good strong teeth—albeit only semi-bleached in these cases), two types of degree modifier of nouns—a quantitative (a good handful) as well as a qualitative (a good beating) type—and a quantity modifier (a good few quid). In all of these uses good retains its unbounded nature and therefore functions as a booster, indicating a (fairly) high point on an open-ended scale. As an adjective modifier, good retains part of its descriptive meaning (paraphrasable as ‘very + approval’), combines with adjectives with a positive interpretation, and is thus not completely bleached. The quantitative nominal type also receives a positive interpretation of ‘substantial’ in the sense of ‘bigger is better.’ In contrast, the qualitative noun degree modifier is largely restricted to negative or violent action nouns; here it upgrades the force inherent in the meaning of the noun. As a quantity modifier it can only accompany absolute quantifiers like many (versus relative ones like half) and generally has a restricted collocational range. Diachronically, good has expanded from descriptive modification to degree modification and finally to quantity modification. The degree uses start out from the quantitative nominal type and expand from there into qualitative nominal and adjectival modification. The process involves both syntactic change from modification to submodification (thus secondary grammaticalization) and subjectification.

Victorina González-Díaz’s study also deals with degree-modifying adjectives, but, unlike good, not on their own, but as pairs in “(tautological) size-clusters” or “size-adjective clusters,” e.g., big huge house, tiny little bird. A large data collection, comprising the three Penn corpora, ARCHER, BNC, and EEBO, is used to investigate the diachronic development of such clusters from Middle English onwards. Old English is excluded, as only the gradual syntactic and semantic restructuring of the noun phrase, in particular of the prenominal slots, makes such pairs possible in later language periods. Apart from a pronounced dip in Late Modern English, the tautological size cluster usage increases slowly but consistently from the 1400s onwards and also becomes more diversified over time with regard to types used in the pattern (while little and great nevertheless remain the most common items). The expansion in Early Modern English is helped by borrowing and lexico-semantic changes, which provide new candidates for the construction. While witnessing further general frequency expansion, Present-Day English also exhibits new functions. Continuing the noun modifier (denoting physical dimensions) inherited from earlier periods, it also shows emphasizer uses (as in the evaluative description of an adult as “great big baby”) and a very rare intensifier subjunct use, e.g., “great big little lumps.” Some pairs, specifically great big, big fat, and whacking great, may also be acquiring greater unity and grammaticalizing in Present-Day English. On the larger level, the cluster construction may have become entrenched as an abstract pattern, which allows further productivity. A qualitative comparison of tautological clusters to coordinated binomials in Middle English and Early Modern English shows large-scale stylistic overlap; the rise of the clusters is explained by providing a stylistically unmarked option as opposed to the binomials’ association with emphatic, rhetorical, and formal styles.

Word-formation intensification patterns in Old English, more precisely those formations involving spatial elements (e.g., þurhbitter ‘very bitter,’ foreaþe ‘very easily’), are investigated by Belén Méndez-Naya. Such synthetic intensifying forms are much more common in Old English than in Present-Day English but are nevertheless already surpassed by analytic intensifying adverbs even then. Following up on the question of the relation of harmony between intensifying element and base, she analyzes both boundedness and semantic prosody. While none of the three common elements, for-, ofer-, and þurh-, show any clear tendencies with regard to semantic prosody, boundedness emerges as a clearer pattern. On the basis of their cognitive image schemas, all three items should prefer bounded bases, which is completely confirmed for ofer- and for-, but only partly for þurh-, which shows a bounded-unbounded parity for types, but more tokens with bounded bases. The latter may be explained by change, namely the tendency for bounded configurations to become unbounded with increasing use. These three are the only truly productive types in the spatial area, which has been established by their type-token ratio, the number/range of word classes and semantic categories among their bases, and the number of hapaxes coined by them. Among them, for- is identified as the most grammaticalized item. The other types looked at, forþ-, in-, of-, or-, and uþ-, are found to be very restricted and unproductive. Their restriction to few or even one base only (e.g., ormæte) is compared to the similar collocational restriction of relic adverbial intensifiers in Present-Day English, such as precious few; we may see in these Old English cases the last stage of obsolescence.

The final paper, by Turo Vartiainen, takes a broad theoretical perspective on the question of whether degree modifiers are evenly distributed across predicative and attributive adjective uses. Apart from the fact that a high-frequency construction might have greater importance for category (here, word class) formation and assignment, the distribution of constructions may also speak for different overall models of categorization, such as abstract prototype-based models (e.g., Aarts 2007) or more usage-based construction-specific models (e.g., Croft 2016). Case studies based on the spoken BNC2014 and on COHA are carried out to shed light on such questions. In two case studies, typical amplifiers and downtoners like those featured in issue 1 contributions as well as intensifying this/that are investigated, all of them showing a clear preference for degree modification in predicative use as opposed to attributive uses. The decline of the “Big Mess” construction, the only possible attributive context for this/that, has further strengthened the association between degree modification and predicative use. If one looks at the degree question from the other angle, namely by studying twenty high-frequency adjectives, the same general preference for the degree-predication link is found. A final case study finds convincing evidence for the proposal that the shift from verbal (very much) to adjectival (very) degree modification with -ed participles is due to the latter’s frequent use in predicative contexts. The paper makes a clear case not only for the intimate connection between predicative contexts and degree modification but also for including such frequency information on constructions in word class theory. It further indicates that there are indeed good reasons here for rethinking the traditional word class “adjective” along Croft’s (2016) smaller and overlapping categories.

5. Concluding Remarks

Although this double issue does not—and could not possibly—treat all the degree phenomena listed in section 1, it nevertheless amply illustrates the great breadth and the potential of research in this area. As each degree item or expression has its own history, often with interesting twists, it is worthwhile focusing on items like well, nigh, little, or so, and on other linguistic realizations beyond intensifying uses of adjectives and adverbs. Equally, linking degree research with related and larger questions, as we have seen for exclamatives or word classes, opens up intriguing research avenues. Finally, abundant opportunity for work remains in the historical sphere by extending research to wider and more interactional contexts, such as drama, letters, or pamphlets, and to considering the role played by genre variation overall in change.