The treatment of canonical forms of idioms:
The present state and recommendations for improvement
The findings of the study discussed in this contribution are based on a database containing 141 idioms originating from literary works, ancient legends, fables and the Bible. Since the aim of the study is to focus on the inclusion of canonical forms in online dictionaries, all these idioms were sought in online dictionaries. The problems faced by dictionary users, especially learners of English, are identified and discussed. Special attention is paid mostly to the following: the use of the article preceding a noun that is the first constituent element in the idiom; the way of including information on possessives in idioms; the use of the infinitive marker to, which is obligatory in some idioms beginning with a verb; the way of indicating variations in idioms; differences in the use of the apostrophe; the inclusion of similes with the comparison marker as; and lower- or upper-case initial letter. Various inconsistencies in the inclusion of the canonical forms of idioms can be found if comparing different dictionaries or even one and the same dictionary. Doubtlessly, some idioms are more fixed than others, and some show a certain degree of variation. Therefore, it is essential to indicate any variation as unambiguously and clearly as possible, so as to make dictionary users aware that some idioms allow some flexibility in their form. Suggestions are made and guidelines are proposed for improving the level of consistency, thus resulting in a more consistent and more user-friendly inclusion of idioms.
Biography: Dr Alenka Vrbinc is Associate Professor at the Faculty of Economics, University of Ljubljana, Slovenia. She has been active in the field of lexicography, primarily learner lexicography and bilingual lexicography, but also in the field of phraseology. She has edited a German-Slovene dictionary and co-authored an English-Slovene dictionary.
Labelling strategies in a specialized dictionary: The case of OIDLE2
In a dictionary, the connotative value of a lexical item is described by labels. This contribution is aimed at studying multiple labelling in the Oxford Idioms Dictionary for Learners of English (OIDLE2). We sought to establish whether labels belonging to one and the same category combine with one another or whether multiple labelling consists of labels from different categories of labels. The database used for the analysis was compiled by searching manually through the dictionary and keying all the idioms with multiple labelling into our database. Altogether, 392 idioms or their senses with two or more labels were found in OIDLE2. The results show that labels expressing different types of diasystematic information are included. As for the frequency of labels, the three most frequent labels appearing in combination with other labels are informal, humorous and old-fashioned. The combination of four labels is used only once, ten idioms were identified with three labels, while the majority of label combinations consist of two labels. Additionally, issues related to multiple labels, combinations of labels expressing different types of diasystematic information and other issues related to labelling are discussed. The inclusion of diasystematic information largely depends on the type of dictionary and on its intended users. This is especially true of dictionaries intended for non-native speakers of a language, where one of the main functions is to promote the active use of a foreign language, and where every single piece of information included in the dictionary counts.
Biography: Marjeta Vrbinc is Associate Professor at the Faculty of Arts, University of Ljubljana, Slovenia. Her main research interests are contrastive lexicology and phraseology, practical lexicography, with the emphasis on English monolingual learner’s dictionaries, and learners' dictionary skills. She has edited a German-Slovene dictionary and co-authored an English-Slovene dictionary.
Opening Up the Digital OED
This paper will discuss recent work carried out in the context of an ongoing project: ‘The Life of Words: Poetry and the OED.’ It will discuss new digital enhancements and analyses of 2nd edition of the Oxford English Dictionary. The project works directly with the background data of OED2, enriching it with metadata, and, alongside other massive datasets, analyzing and comparing data with custom programs written in Python. Metadata now being added back to the dictionary concerns the genre and register of evidence quotations, as well as a number of other features, including the gender of the author, the type of publication (e.g. periodical, codex, etc.), and so on. So, for instance, it is now only a matter of a few lines of code to produce a quantitative assessment of the question, ‘how many women authors were quoted in OED1, and what kinds of works are they quoted for?’ Or: ‘What proportion of non-Shakespearean early-modern verse drama contributes the earliest citation of a sense extension, and how does this compare to A) Shakespeare, and B) works in other genres?’ In the first main part of the paper, I give an overall quantitative assessment of the generic make-up of OED quotations, comparing OED1 and OED2. I then offer a re-assessment of claims (and counter-claims) surrounding the linguistic inventiveness of canonical authors such as Shakespeare and Milton, based on benchmarks for the period and genre of their various works. Finally I sketch out a number of other potential avenues of investigation that an enhanced OED can open up.
Biography: David Williams teaches English literature at St Jerome’s University in the University of Waterloo. He is the author of a monograph on ethics and contemporary poetry (Oxford University Press), and a number of articles on poetry,
interactions between poets and dictionaries, and the Digital Humanities. He is currently at work on a monograph on etymology and modern poetry, and another on poets and the OED.
Dictionaries and linguistic secession: the case of Occitan
Modern Occitan is commonly divided into dialects such as Auvergnat, Gascon, Languedocien, Limousin, Provençal, and Vivaro-Alpine. This is not an exhaustive list, and these dialects are not clearly delineated. In fact, their status as dialects of Occitan is not universally agreed upon. Attempts to standardize Occitan have therefore not come without controversy. Speakers of some “dialects” have instead proposed competing norms, such as Pierre Bonnaud in Auvergne, Pierre Blanchet in Provence, and Jean Laffite in Gascony (Balaguer 2010). Given the divided nature of Occitan, lexicographers may intend their work to represent a particular dialect (whether they consider it part of Occitan or not) or Occitan in general. In my previous work (Williams 2016), I showed that out of 62 dictionaries of Occitan and its dialects from 1966-2014, only 32 included the word Occitan in their titles. Furthermore, only 7 did not include a region or dialect in their titles. The present work builds upon this discovery, looking beyond the dictionaries’ titles to the works themselves. After studying the descriptions of dictionaries of Occitan and its dialects from 2000-2015, these dictionaries fall into three categories: dictionaries of a language variety not meant to be considered part of Occitan, dictionaries of dialects of Occitan, and dictionaries meant to represent Occitan in general. This presentation will discuss the goals of these three types of dictionaries, and their role in the standardization of Occitan or of their respective language varieties (cf. Zgusta 1989).
Biography: Krista Williams is Assistant Professor of French at the University of Evansville. Her research areas include the definitions and translations of color terms, profanity in pidgins and creoles, and Occitan lexicography.
“Folklore in a Caribbean English Creole dictionary: Inclusion, Definition and Extraction”
Dealing with “folklore” entries in a dictionary involves questions, beginning with what should be included, and how such entries should be defined in terms of evaluative or judgemental descriptors. Focusing on the Dictionary of the English/Creole of Trinidad & Tobago (Winer, 2009), this paper discusses the processes of inclusion, definition, and possible extraction of lexicon in this domain. By broad definition (Brunvand 1968), anything that is not “standard” can be considered “folklore.” Because of the relative scarcity and inaccessibility of sources of cultural and linguistic information on Trinidad & Tobago, the dictionary tends towards an encyclopedia rather than a glossary. A deliberate effort was made to include as much cultural content and folk knowledge as possible, areas which do not easily limit themselves to single lexical items. Inclusion is discussed in terms of a) lemmatization – especially phrases / proverbs; b) definition and commentary; and c) citation. Types and limitations of data and sources included are also noted. The significance of words such as “superstition” and “witchcraft” and defining words such as “supposed,” – which imply that the editor is not committed to believing the definition – are discussed in reference to Pickett (2007). The second process addressed is extraction. What retrieval keywords and strategies can folklore-seeking readers use to locate relevant entries? Must this wait until digitalization of the work? Are there keywords in the defining process that could help rationalize and streamline this process in future?
Biography: Lise Winer is editor of the Dictionary of the English/Creole of Trinidad & Tobago (2009), author of Badjohns, Bhaaji & Banknote Blue: Essays on the Social History of Language in Trinidad and Tobago (2007), and co-author of “Grasping the nettle: Handling flora entries in dictionaries” (Dictionaries 2016).