Changing intensifiers in Late Modern English, 1700-1900: a historical socio-pragmatic analysis
The project will investigate the (changing) use of two groups of intensifiers, namely amplifiers scaling upwards (e.g. 'terribly', 'most') and downtoners (e.g. 'slightly', 'a bit') in British courtroom speech in 1700-1900. We will look into empirical evidence from that period and focus on factors underlying variation and change. A novelty in our approach is that we will pay special attention to the socio-pragmatic bases of change. We will draw on a new and unique electronic resource, the 14-million-word Old Bailey Corpus. This corpus provides access to language taken down in authentic, interactional speech situations from the mouths of speakers representing varying social strata and acting in different institutional roles in the court-room.
Main areas of investigation:
(1) Language change: how did intensifiers develop as a class and individual types? Our hypothesis is that intensifiers (different types and frequencies) will increase over the period and start mirroring PDE usage.
(2) Which syntactic, lexico-semantic and pragmatic factors related to intensification favour the emergence of intensifiers?
(3) Sociolinguistics: what role does a speaker's gender, age, rank, occupation and courtroom role play?
(4) Discourse-pragmatics: who uses intensifiers in which contexts and for what effect? Our hypothesis is that speakers have varying strategic needs of intensifiers, leading to a cline in frequencies in descending order from defendants, witnesses, via lawyers to judges.
Main areas of investigation:
(1) Language change: how did intensifiers develop as a class and individual types? Our hypothesis is that intensifiers (different types and frequencies) will increase over the period and start mirroring PDE usage.
(2) Which syntactic, lexico-semantic and pragmatic factors related to intensification favour the emergence of intensifiers?
(3) Sociolinguistics: what role does a speaker's gender, age, rank, occupation and courtroom role play?
(4) Discourse-pragmatics: who uses intensifiers in which contexts and for what effect? Our hypothesis is that speakers have varying strategic needs of intensifiers, leading to a cline in frequencies in descending order from defendants, witnesses, via lawyers to judges.
Final report
Changing intensifiers in Late Modern English, 1700-1900: a historical socio-pragmatic analysis
The project was aimed at investigating the variation and change of intensifiers, i.e., items such as ‘perfectly’, ‘very’, ‘hardly’, ‘a little’, in Late Modern English (1700-1900) and account for the factors promoting change towards the Present-day English system. While most previous research had tended to focus on single items in texts containing written language, the present project focussed on a larger group of intensifiers within the pragmatically circumscribed context of the courtroom, drawing on evidence from the Old Bailey Corpus (OBC), a large-scale historical corpus of speech-based court records.
Since the proposal stage, the project underwent changes, among the most notable the version of the OBC, categorization of intensifiers, and the statistical analyses envisaged. When the data collection started, the version available to us was the so-called OBCExt (Extended) version, which comprised 17 million words. Once the further expanded and corrected OBC 2.0 version comprising 24 million words became available, the remaining data collection was based on that version and the earlier data sets supplemented and updated accordingly. As for the categorization of the items, the intensifier groups originally envisaged comprised maximizers (e.g. ‘fully’, ‘extremely’), boosters (e.g. ‘greatly’, ‘highly’) and downtoners, the last-mentioned group divided into diminishers (e.g. ‘slightly’, ‘faintly’) and minimizers (e.g. ‘hardly’, ‘scarcely’). For a more nuanced picture, a further category of downtoners was distinguished, i.e. moderators (e.g. ‘somewhat’, ‘fairly’). Regarding statistical analyses, it became necessary to complement descriptive frequency statistics with multivariate analyses, notably multiple regression analysis, which enabled us to investigate the influence of a predictor (e.g. the speaker’s gender or social class) on change, setting aside the possible influence of the other factors.
The Historical Thesaurus of the Oxford English Dictionary was used to compile an inventory of potential intensifiers for computerized searches. Data collection was carried out using the search engine accompanying the OBC. The raw data files were screened for relevant instances so that each final output file was subjected to an independent analysis or checking round by each project member. Excel was used for the frequency counts and distribution figures, and regression analyses were carried out with R. To observe the change, the 200-year period was divided into five 40-year subperiods.
The courtroom speakers in the Old Bailey made use of a wide range of intensifiers. Altogether 62,852 relevant instances comprising 9,488 maximizers (23 items), 47,613 boosters (44 items) and 5,751 downtoners (19 items) were included in the study. The main results of the project include the following:
(1) The maximizers, boosters and downtoners differ in their developments across the 200-year time span covered by the study: while the overall use of the maximizers increased across the period, the use of boosters declined and that of downtoners largely remained stable; three individual items, i.e. the boosters ‘very’ and ‘so’ and the downtoner ‘(a) little’ experienced a major decline, influencing the overall results;
(2) Regarding the two main linguistic features studied, i.e. the word class properties and the semantics of the targets of modification (ToMs), the three groups displayed mainly similarities but also some differences; while adjectives and adverbs proved the best represented classes for maximizers and boosters, the downtoners somewhat unexpectedly mostly modified verbs; of the semantic qualities distinguished for ToM adjectives, ‘human propensity’ played the leading role, followed by ‘value’ in the maximizer and booster groups but by ‘physical property’ in the downtoner group;
(3) Regarding the influence of the extralinguistic predictors investigated, for maximizers and boosters time proved the most important factor in change, but for downtoners, a speaker’s role in the courtroom instead. The other predictors investigated, i.e. gender and social class, also revealed similarities and differences in the use of the items opted for across the speaker groups; for instance, male speakers used more maximizers, a finding bucking the so far attested general trend placing female language users in the lead in the use of intensifiers overall. We have aimed at accounting for this, and similar results yielded by the study, with reference to various situational and pragmatic factors at play in the courtroom.
We originally hypothesized in our project proposal that intensifier uses (types and tokens) will increase over the period and that the inventory will increasingly mirror the Present-day English status quo. This increase proved to hold for maximizers only. As for Present-day English, this group also anticipated Present-day English usage on several accounts, apart from a number of notable exceptions (e.g. the preference for ‘perfectly’ in the late modern courtroom). We also hypothesized that our study would allow us to observe individual intensifiers falling in obsolescence; this was also shown to be the case by our data (e.g. the very rare use of the form ‘full’ as in ‘full weak’). Further, our results proved correct our hypothesis that the ToMs and their semantic analysis would provide an important gateway to identify the formulaic and/or flexible properties of the individual intensifier types regarding their collocational preferences. As for the sociolinguistics of intensifier usage in the late modern courtroom, we were able to identify the innovators and conservative users of the different intensifier types, with the speech of laypersons (witnesses, defendants) in special focus. Finally, regarding the discourse-pragmatics of intensifiers, we hypothesized that both witnesses/defendants and lawyers would have more need of strategic intensifier usage and thus use them more frequently than judges. In overall terms, this hypothesis could be verified for witnesses and defendants, while both lawyers and judges generally remained less frequent users of intensifiers, along with victims.
Among the new research questions regarding our courtroom materials are the potential influence of the speaker’s age, the type of speech context in question (face-to-face question-answer pairs, narration in first or third person, etc.), and variation and change in the use of many other forms of intensification such as ‘this’/‘that’, ‘good’, and phrases/constructions beyond single words, e.g. exclamative ‘what’ constructions. Nor is it only discourse contexts, or only specific words that attract intensifiers, but also certain clearly circumscribed topical/semantic fields, e.g., intoxication and types of crimes. More work is due in this respect.
The project has generated further interest in research on intensifiers and their history. Contacts with the compilers of the OBC including their technical expert at the University of Giessen (Germany) have been very valuable; in the context of our statistical analyses, we also collaborated with an expert at the University of Bamberg, specializing in regression analysis and the use of R. In August 2018, we organized a workshop dedicated to the wider context of intensification at the 20th International Conference on English Historical Linguistics (ICEHL) in Edinburgh; the workshop led to our editorial project comprising the publication of a selection of the papers presented at the workshop and further papers commissioned; this enabled us to share views on methodology (materials, data collection) and analytical frameworks.
We have given presentations at five international conferences, i.e., the ICAME conferences in Hong Kong (2016), Prague (2017), and Tampere (2018), the ICEHL conference in Duisburg-Essen (2016), and at the international Late Modern English conference in Uppsala (2017). A book chapter on the use of intensifiers in a specific context is out; an article on maximizers is available online (the hard copies are in production); an article on the frequent item ?little’ is to be submitted to print by September 2020, and a comprehensive monograph about all the intensifiers included in the project analyses is forthcoming.
The general public received project information via an article in the Augsburg University research supplement to the newspaper Augsburger Allgemeine, reaching more than 200,000 readers.
PUBLICATIONS
**Monograph**
Claridge, Claudia, Ewa Jonsson and Merja Kytö. Forthcoming (contracted). Intensifiers in English: A Socio-pragmatic Analysis, 1700-1900. Cambridge University Press.
**Articles and book chapters**
Claridge, Claudia, Ewa Jonsson and Merja Kytö. 2019 (online, OA; printed version in production).
Entirely innocent: A historical sociopragmatic analysis of maximizers in the Old Bailey Corpus, DOI: https://doi.org/10.1017/S1360674319000388
Published online by Cambridge University Press: 23 December, 2019. OA.
Claridge, Claudia. 2019. Drinking and crime: Negotiating intoxication in courtroom discourse, 1720-1913. In Teresa Fanego & Paula Rodríguez-Puente (eds.), Corpus-based Research on Variation in English Legal Discourse. Amsterdam: Benjamins, 261-285.
Claridge, Claudia, Ewa Jonsson and Merja Kytö. Forhcoming (accepted for publication). A little something goes a long way: little in the Old Bailey Corpus. Special issue of the Journal of English Linguistics, eds. Claudia Claridge and Merja Kytö. OA underway.
**Special issue**
Claridge, Claudia and Merja Kytö (eds.) Degree and Related Phenomena in the History of English. Special issue (double number) of the Journal of English Linguistics. Forthcoming (accepted for publication).
**Popular science publication**
„Dead drunk“ oder „perfectly innocent“. Research Supplement, Augsburger Allgemeine, Summer 2018, p. 3. https://assets.uni-augsburg.de/media/filer_public/80/10/8010a37c-1e5a-429e-b921-b6b0f5ff01dc/wissenschaftundforschungina_beilage-sose2018.pdf.
The project was aimed at investigating the variation and change of intensifiers, i.e., items such as ‘perfectly’, ‘very’, ‘hardly’, ‘a little’, in Late Modern English (1700-1900) and account for the factors promoting change towards the Present-day English system. While most previous research had tended to focus on single items in texts containing written language, the present project focussed on a larger group of intensifiers within the pragmatically circumscribed context of the courtroom, drawing on evidence from the Old Bailey Corpus (OBC), a large-scale historical corpus of speech-based court records.
Since the proposal stage, the project underwent changes, among the most notable the version of the OBC, categorization of intensifiers, and the statistical analyses envisaged. When the data collection started, the version available to us was the so-called OBCExt (Extended) version, which comprised 17 million words. Once the further expanded and corrected OBC 2.0 version comprising 24 million words became available, the remaining data collection was based on that version and the earlier data sets supplemented and updated accordingly. As for the categorization of the items, the intensifier groups originally envisaged comprised maximizers (e.g. ‘fully’, ‘extremely’), boosters (e.g. ‘greatly’, ‘highly’) and downtoners, the last-mentioned group divided into diminishers (e.g. ‘slightly’, ‘faintly’) and minimizers (e.g. ‘hardly’, ‘scarcely’). For a more nuanced picture, a further category of downtoners was distinguished, i.e. moderators (e.g. ‘somewhat’, ‘fairly’). Regarding statistical analyses, it became necessary to complement descriptive frequency statistics with multivariate analyses, notably multiple regression analysis, which enabled us to investigate the influence of a predictor (e.g. the speaker’s gender or social class) on change, setting aside the possible influence of the other factors.
The Historical Thesaurus of the Oxford English Dictionary was used to compile an inventory of potential intensifiers for computerized searches. Data collection was carried out using the search engine accompanying the OBC. The raw data files were screened for relevant instances so that each final output file was subjected to an independent analysis or checking round by each project member. Excel was used for the frequency counts and distribution figures, and regression analyses were carried out with R. To observe the change, the 200-year period was divided into five 40-year subperiods.
The courtroom speakers in the Old Bailey made use of a wide range of intensifiers. Altogether 62,852 relevant instances comprising 9,488 maximizers (23 items), 47,613 boosters (44 items) and 5,751 downtoners (19 items) were included in the study. The main results of the project include the following:
(1) The maximizers, boosters and downtoners differ in their developments across the 200-year time span covered by the study: while the overall use of the maximizers increased across the period, the use of boosters declined and that of downtoners largely remained stable; three individual items, i.e. the boosters ‘very’ and ‘so’ and the downtoner ‘(a) little’ experienced a major decline, influencing the overall results;
(2) Regarding the two main linguistic features studied, i.e. the word class properties and the semantics of the targets of modification (ToMs), the three groups displayed mainly similarities but also some differences; while adjectives and adverbs proved the best represented classes for maximizers and boosters, the downtoners somewhat unexpectedly mostly modified verbs; of the semantic qualities distinguished for ToM adjectives, ‘human propensity’ played the leading role, followed by ‘value’ in the maximizer and booster groups but by ‘physical property’ in the downtoner group;
(3) Regarding the influence of the extralinguistic predictors investigated, for maximizers and boosters time proved the most important factor in change, but for downtoners, a speaker’s role in the courtroom instead. The other predictors investigated, i.e. gender and social class, also revealed similarities and differences in the use of the items opted for across the speaker groups; for instance, male speakers used more maximizers, a finding bucking the so far attested general trend placing female language users in the lead in the use of intensifiers overall. We have aimed at accounting for this, and similar results yielded by the study, with reference to various situational and pragmatic factors at play in the courtroom.
We originally hypothesized in our project proposal that intensifier uses (types and tokens) will increase over the period and that the inventory will increasingly mirror the Present-day English status quo. This increase proved to hold for maximizers only. As for Present-day English, this group also anticipated Present-day English usage on several accounts, apart from a number of notable exceptions (e.g. the preference for ‘perfectly’ in the late modern courtroom). We also hypothesized that our study would allow us to observe individual intensifiers falling in obsolescence; this was also shown to be the case by our data (e.g. the very rare use of the form ‘full’ as in ‘full weak’). Further, our results proved correct our hypothesis that the ToMs and their semantic analysis would provide an important gateway to identify the formulaic and/or flexible properties of the individual intensifier types regarding their collocational preferences. As for the sociolinguistics of intensifier usage in the late modern courtroom, we were able to identify the innovators and conservative users of the different intensifier types, with the speech of laypersons (witnesses, defendants) in special focus. Finally, regarding the discourse-pragmatics of intensifiers, we hypothesized that both witnesses/defendants and lawyers would have more need of strategic intensifier usage and thus use them more frequently than judges. In overall terms, this hypothesis could be verified for witnesses and defendants, while both lawyers and judges generally remained less frequent users of intensifiers, along with victims.
Among the new research questions regarding our courtroom materials are the potential influence of the speaker’s age, the type of speech context in question (face-to-face question-answer pairs, narration in first or third person, etc.), and variation and change in the use of many other forms of intensification such as ‘this’/‘that’, ‘good’, and phrases/constructions beyond single words, e.g. exclamative ‘what’ constructions. Nor is it only discourse contexts, or only specific words that attract intensifiers, but also certain clearly circumscribed topical/semantic fields, e.g., intoxication and types of crimes. More work is due in this respect.
The project has generated further interest in research on intensifiers and their history. Contacts with the compilers of the OBC including their technical expert at the University of Giessen (Germany) have been very valuable; in the context of our statistical analyses, we also collaborated with an expert at the University of Bamberg, specializing in regression analysis and the use of R. In August 2018, we organized a workshop dedicated to the wider context of intensification at the 20th International Conference on English Historical Linguistics (ICEHL) in Edinburgh; the workshop led to our editorial project comprising the publication of a selection of the papers presented at the workshop and further papers commissioned; this enabled us to share views on methodology (materials, data collection) and analytical frameworks.
We have given presentations at five international conferences, i.e., the ICAME conferences in Hong Kong (2016), Prague (2017), and Tampere (2018), the ICEHL conference in Duisburg-Essen (2016), and at the international Late Modern English conference in Uppsala (2017). A book chapter on the use of intensifiers in a specific context is out; an article on maximizers is available online (the hard copies are in production); an article on the frequent item ?little’ is to be submitted to print by September 2020, and a comprehensive monograph about all the intensifiers included in the project analyses is forthcoming.
The general public received project information via an article in the Augsburg University research supplement to the newspaper Augsburger Allgemeine, reaching more than 200,000 readers.
PUBLICATIONS
**Monograph**
Claridge, Claudia, Ewa Jonsson and Merja Kytö. Forthcoming (contracted). Intensifiers in English: A Socio-pragmatic Analysis, 1700-1900. Cambridge University Press.
**Articles and book chapters**
Claridge, Claudia, Ewa Jonsson and Merja Kytö. 2019 (online, OA; printed version in production).
Entirely innocent: A historical sociopragmatic analysis of maximizers in the Old Bailey Corpus, DOI: https://doi.org/10.1017/S1360674319000388
Published online by Cambridge University Press: 23 December, 2019. OA.
Claridge, Claudia. 2019. Drinking and crime: Negotiating intoxication in courtroom discourse, 1720-1913. In Teresa Fanego & Paula Rodríguez-Puente (eds.), Corpus-based Research on Variation in English Legal Discourse. Amsterdam: Benjamins, 261-285.
Claridge, Claudia, Ewa Jonsson and Merja Kytö. Forhcoming (accepted for publication). A little something goes a long way: little in the Old Bailey Corpus. Special issue of the Journal of English Linguistics, eds. Claudia Claridge and Merja Kytö. OA underway.
**Special issue**
Claridge, Claudia and Merja Kytö (eds.) Degree and Related Phenomena in the History of English. Special issue (double number) of the Journal of English Linguistics. Forthcoming (accepted for publication).
**Popular science publication**
„Dead drunk“ oder „perfectly innocent“. Research Supplement, Augsburger Allgemeine, Summer 2018, p. 3. https://assets.uni-augsburg.de/media/filer_public/80/10/8010a37c-1e5a-429e-b921-b6b0f5ff01dc/wissenschaftundforschungina_beilage-sose2018.pdf.