Course: Exploring Online Language Resources

Originally posted at blogs.it.ox.ac.uk on January 21, 2016 by Martin Wynne. IT Services at the University of Oxford has decided to delete a large number of historical blogs, and this is one of a number of posts related to the Oxford Text Archive which are being re-published here, after being laboriously retrieved from the archive provided by the Wayback Machine.

A course in the IT Learning Programme at the University of Oxford in Hilary Term 2016 will explore how we can use online language datasets to explore language, history and culture. This course is the latest stage in the evolution of the ‘Corpus Linguistics’ course which has run for the past few years.

We now have at our fingertips huge amounts of language data in digital form, representing unprecedented opportunities for exploring and analysing language and discourse. How we can use the evidence of language usage in digital resources to draw conclusions about language, culture and society? Drawing on techniques and methods from corpus linguistics, this course will offer guidance on finding and evaluating digital sources, hands-on exercises to explore and analyse data, and some suggestions on how to assess, use and interpret evidence from digital sources.

Each session will focus on the exploration and analysis of a different corpus or dataset, with practical hands-on exercises for how to use the resource to find evidence to explore linguistic, socio-cultural and historical research questions. Participants are free to attend the whole course or individual sessions, although there will be benefits, to beginners in particular, to attend the whole course and build week by week on the techniques and insights offered by each session. Prior registration with the online services described below will save time on the day.

1. BNCWeb – exploring a corpus of late twentieth century English

12:30-13:30 Thursday 28th January (HT week 2) at IT Services, Banbury Road. Sign for free to attend the course here.

The British National Corpus is a very widely used and cited dataset, which was designed and built in the 1990s to provide a representative and balanced sample of modern British English, in speech and writing, across a number of varieties in a wide range of contexts. This session will introduce and explore basic concepts of corpus design and construction, and introduce techniques, functions and methods for corpus analysis. BNCWeb is a customized application of CQPweb, with the facilities to exploit and analyse the linguistic annotation of the texts in the BNC, and to make use of the detailed descriptions of the sources. Participants can access BNCWeb with Oxford single sign-on via a link at https://ota.oerc.ox.ac.uk/ (and can also register to use the BNCWeb service hosted at Lancaster University, which will be the back-up in case of problems with BNCWeb at Oxford).

2. CQPweb – exploring a range of corpora

12:30-13:30 Thursday 4th February (HT week 3) at IT Services, Banbury Road.  Sign for free to attend the course here.

The online application CQPweb offers an interface to a powerful corpus search and analysis engine which can be applied to any textual dataset. CQPweb is an open source software application, deployed at many institutions around the world to offer access to a wide range of corpora. This session will focus on mining a large corpus from Early English Books Online for historical information. Participants should register to use the service at http://cqpweb.lancs.ac.uk/.

3. corpus.byu.bnc – historical and cultural investigations

12:30-13:30 Thursday 11th February (HT week 4) at IT Services, Banbury Road.  Sign for free to attend the course here.

The set of large corpora hosted at Brigham Young University include contemporary and historical Corpora of British and American English, Spanish, Portuguese, and the Hansard Corpus of UK parliamentary proceedings. This session will further extend the exploration beyond linguistic research questions to explore historical and political texts. Participants should register to use the service at http://corpus.byu.edu/.

4. The Oxford English Corpus – lexicography and beyond

12:30-13:30 Thursday 18th February (HT week 5) at IT Services, Banbury Road.  Sign for free to attend the course here.

The Oxford English Corpus, and related datasets, offer the opportunity to  explore current and recent trends in the English language, via a very large and growing corpus which is regularly updated with new texts. This corpus is used by the lexicographers at Oxford University Press to create and update entries in the Oxford English Dictionary and other dictionaries, reference works and teaching materials, and can also be used to monitor and discover social trends via the discourses revealed in the data. The Oxford English Corpus uses the SketchEngine software to manage, filter and reveal patterns in these multi-billion word corpora. Log-in credentials for the Oxford English Corpus are kindly supplied by OUP and will be issued during the tutorial session.

5. Exploring modern European languages with CLARIN

12:30-13:30 Thursday 25th February (HT week 6) at IT Services, Banbury Road.  Sign for free to attend the course here.

A wealth of corpora and other language resources are becoming more easily available to researchers thanks to the CLARIN European Research Infrastructure Consortium. The UK has recently joined CLARIN as an Observer, allowing access to all UK researchers with institutional single sign-on via the UK Federation. We’ll take a whistle-stop tour of some of the available languages and corpora, with a focus on the facilty for Federated Content Search, finding hits for a search term across a wide number of resources held in different repositories. Access is available to all University of Oxford users, and via institutional single sign-on to users in higher education institutions from participating countries (see more about access at http://clarin.eu/content/easy-access-protected-resources).

The course will take place Thursday lunchtimes weeks 2-6, Hilary Term 2016, at IT Services, Banbury Road, and will be taught by Ylva Berglund Prytz and Martin Wynne of IT Services, University of Oxford. It is open to all members of the University of Oxford and there is no charge.

A Virtual Museum of Language for Oxford

Originally posted at blogs.it.ox.ac.uk on November 4, 2015 by Martin Wynne. IT Services at the University of Oxford has decided to delete a large number of historical blogs, and this is one of a number of posts related to the Oxford Text Archive which are being re-published here, after being laboriously retrieved from the archive provided by the Wayback Machine.

The University of Oxford is home to unrivalled collections of  materials and expertise relating to human language, dispersed across faculties, museums, libraries and other units. A pilot project is proposed to imagine a new virtual museum, which would bring together and feature online resources and information about physical collections at the University of Oxford, in the various museums, galleries and libraries. The visitor to the website would see images, video, sound and text, to entertain and to inform about some aspect of language and its use. The collections of the virtual museum would start by creating exhibits which draw together existing materials, and then be built up over time with a series of new online exhibits created by guest curators. Exhibits could feature digital images of artefacts in museums, opportunities for exploring language resources online, interactive language games, podcasts and blogs about research in the University. The intended audience would be the general public, including sixth form students.

The museum could work closely with, and not try to replace or compete with, existing outreach activities. It should be seen as a portal to help web visitors to find ways to navigate to the wealth of language-related online resources created by people in different parts of the University. Obvious partners, already involved in outreach, engagement, dissemination and knowledge exchange activities online would be TORCH, Oxford Sparks, Digital Humanities at Oxford, Digital.Bodleian and numerous websites provided by OUP for language learners and others interested in language. The museum could also provide a platform for new citizen science and crowd-sourcing projects.

Language is a feature of all disciplines in the university, sometimes directly as the sole object of study, but more often as one component in more complex objects and processes, and almost ubiquitously as the medium of communication. Linguists study language, but all human and social scientists study social and cultural phenomena which are infused with language. There are scientists who study physical, mental and medical aspects of language, but all conduct the large part of their communications via language.

The distributed collections would include obvious candidates for material from the following faculties: Modern Languages; Linguistics, Philology and Phonetics; Oriental Studies; Classics; English Language and Literature, but all would be welcome to contribute. One of the key uses of the museum would be to offer a route for dissemination and outreach for research projects by presenting their research and their outputs to a general audience, and the musuem could be a central pathway to impact for these and other disciplines.

In Social Sciences, initial exhibits could be sought from Anthropology (relating to endangered languages), Education (particularly relating to language teaching and learning), Oxford Internet Institute (language on the web). The Migration Observatory (http://migrationobservatory.ox.ac.uk/) is involved in large-scale linguistic analysis of public discourse about migration, as a sociological research topic and to inform social policy and to inform the general public, and has existing materials relevant for a museum exhibit on this topic (see, for example http://www.compas.ox.ac.uk/2013/pr-2013-migration_media/).

In Mathematical, Physical and Life Sciences Division (MPLS), the e-Research Centre and the Computational Linguistics Group in Computer Science are already home to research projects focussed on language, and Medical Sciences is involved in research, treatment and therapy relating to numerous aspects of human language. Exhibits could build on existing collaborations on themes relating to medical humanities, the history of science, and humanities and science (especially TORCH themes and networks), and the degree course in  Psychology, Philosophy and Linguistics (http://www.ox.ac.uk/admissions/undergraduate/courses-listing/psychology-philosophy-and-linguistics).

Oxford University Press could potentially provide exhibits relating to:

  • The Press Archive (e.g. old printing presses, books, artefacts relating to the history of the OED);
  • Oxford English Dictionary (OED);
  • Dictionaries and Scholarship online;
  • Relevant entries from the DNB (e.g famous linguists and linguistic innovators);
  • Monographs (some full-text online) on linguistics and history of language monographs, and also examples of different language varieties exemplified in publications from different periods;
  • Corpora of contemporary language  (e.g Oxford Corpus of English, Oxford Twitter Corpus).

Relevant exhibits from the Bodleian Libraries are practically limitless, and could certainly include materials from:

  • Centre for Digital Scholarship;
  • Centre for the Study of the Book;
  • Special collections;
  • Digital.Bodleian;

and all of the libraries in the University would of course be encouraged to create exhibits. The Language Centre and the Department for Continuing Education are also likely to be potential collaborators.

The representation of materials and artefacts in the Museums would be a key pillar of the virtual museum. All University Museums would be encouraged to contribute, with some obvious candidates being early language inscriptions and tablets in the Ashmolean Museum, audio recordings by anthropologists in the Pitt Rivers. The University Museum of National History could bring in an evolutionary perspective, and perhaps something on animal communication from. Collaborations with the the galleries and visual art should also be explored.

Featured themes could bring together cross-disciplinary perspectives, and could include, for example, real-time monitoring of public and social media discourse on selected themes (e.g. Europe, environment, migration, etc.),  500 years of the history of printing, language of the reformation (2017 will be the five-hundredth anniversary of Luther’s 95 Theses), the Language of the first world war, etc..

Some exhibits in the virtual museum would draw on existing online materials, e.g.

Simple interfaces would also be offered to search language resources online, such as the British National Corpus (BNC) simple search, and the similar services for many other languages, e.g. http://clarin.eu/content/services. Further possible services from the virtual museum could be ‘Ask an expert’, where language-related queries could be answered and discussed.Other potential participants outside the university could be invited to contribute, although the intention would be to maintain a strong University of Oxford identity and branding for the museum. Potential external partners could include the Oxford Brookes publishing course, Digital Oxford, schools, other publishers in addition to OUP, The Story Museum, and other museums and libraries. The CLARIN-UK network could also be invited to contribute exhibits. CLARIN-UK is a consoritum of linguistics experts in the UK who have come together to promote the use of online and digital resources in research in the humanitiies and social sciences and beyond. Potential exhibits here could be based on the Metaphor Map, SCOTS corpus, the ESRC Centre for Corpus Approaches in Social Sciences, text mining from the GATE team at the University of Sheffield, etc. Some of the activites and resources involved can be seen at the CLARIN-UK website.

As well as being inspired by existing collections and research in Oxford, the developing ideas for this museum have drawn on a number of sources, including discussions with the Ludwig Eichinger, Director of a new Museum of the German Language which is under development in Mannheim, and the proposals for an English Language Museum in Winchester (http://www.englishproject.org/resources/english-language-museum-winchester), and a proposal made for by David Crystal some years ago for a London Language Museum (http://www.davidcrystal.community.librios.com/?fileid=-4845 [PDF file]). These are all proposed as physical museums, but they usefully draw attention to the gap in the market, and the potential breadth of relevant exhibits. The Language at Leeds initiative also shows how language can be at the centre of a truly multi-disciplinary activity in a University.

More suggestions and volunteers to contribute are welcome!

Analyzing the Language of Social Media

Originally posted at blogs.it.ox.ac.uk on March 25, 2015 by Martin Wynne. IT Services at the University of Oxford has decided to delete a large number of historical blogs, and this is one of a number of posts related to the Oxford Text Archive which are being re-published here, after being laboriously retrieved from the archive provided by the Wayback Machine.

The eighth one-day event Corpus Linguistics in the South was held at the University of Reading on Saturday 15th November 2014, and focussed on research analysing the language of social media.

Dawn Knight (Newcastle University) spoke about the ‘spoken-ness of e-language’, exploring the positioning of online discourse in relation to the norms of spoken and written language. In short, is language online more like speech or writing?

There are several interesting aspects to Dawn’s project. Explicit permission to re-use and share the data was obtained from all contributors, who were active and popular participants in online discourses. Anonymization of private personal data was carried out, but the data set does not seem to be available for other scholars to use. Funding for the project came from Cambridge University Press, who, it appears, are not willing to share the data.

One starting hypothesis is that there is a continuum of formality for interactive language, with writing at one end and speech on the other (see, for example, David Crystal, English as Global Language, 2003). Can we map different forms (blogs, tweets, email, discussion forums, SMS) onto this continuum?

Pronouns and deictic markers are interesting. In spoken interaction there are typically references to people, actions and things in the shared immediate context, and (probably as a result of this) pronouns and deictic markers are typically more common in speech. Corpora also show that personal pronouns, adverbs and interjections are more common in speech. In this sense, the e-language corpus looks more like speech than writing, despite asynchronicity of the discourse and the lack of shared space. Dawn suggests that there might be an over-compensation, since in online forums we are more reliant, almost exclusively reliant, on language for interactional aspects of communication.

Some results went against expectations – ‘shall’ and ‘must’ are thought to be generally in decline, particularly in informal registers, but proved to be more frequent than expected in SMS language. Discussion forums proved to be most like speech in many ways, despite their low interactivity and asynchronicity.

Dawn’s approach looks promising, and the initial results are suggestive. Further research could involve visualization of the multidimensional comparisons between corpora, for example, to explore more fine-grained identifications of similarities between different e-language and language types in a reference corpus.

The next two papers in the morning focussed on the analysis of particular online forums. Daniel Hunt (QMU) explored the language of an online forum for sufferers of anorexia. An approach based on keywords showed ways in which participants in the forum present their illness as an entity external to themselves, thus presenting themselves as passive and unaccountable. Amanda Potts (Lancaster University) presented an exploration of ‘queer’ sexual innuendo in an area of discourse and human activity that was new to me – commentaries accompanying highly popular videos of Minecraft games. It wasn’t clear to me that either project was able to draw conclusions that they couldn’t have drawn from simply reading the texts in their relatively small datasets, nor that there was anything particularly interesting or significant about the particular issues, themes and areas of social media which they had chosen.

Amy Aisha Brown (Open University) set out to examine English in Japan via Twitter, but the analysis seemed to be of tweets in Japanese which included references to the English language, so I’m afraid that I was a bit lost. She seemed to draw the conclusion that fluency in English is generally considered in a positive way in the Japanese twittersphere. Amy used a Windows desktop application Tweet Archivist Desktop and a programme called KH Coder for cleaning and analysis. Again, the data is not being shared and no indication whether this might ever be possible.

Alison Sealey and Chris Pak (Lancaster University) reported on a small-scale analysis of references to animals on Twitter, part of a larger project examining discourse about animals. The project used an online service called Topsy to find tweets, but it wasn’t clear to me how the analysis and the results are arrived at, are or what the research questions are.

Rachelle Vessey (Newcastle University) took a more theoretical tack, characterising mainstream corpus linguistics as being mainly concerned with, and focussed on, notions of stability and normativity, and on standard languages. The idea of ‘superdiversity’, was presented as a cultural successor to multiculturalism, with an assumption of more diverse and fast-changing cultural formations. She has pursued this issue in the context of Canadian language politics, examining tweets relating to a recent controversy known as ‘pastagate’. The data in this project, like others here today, was somewhat complicated by the large number of retweets. The somewhat underwhelming conclusion was that the largely separate English and French language communities operate separately in Twitter.

Yin Yin Lu (Oxford Internet Institute, University of Oxford) is investigating the linguistics of the Twitter hashtag. She has used the Streaming API, which offers access to a restricted number of tweets according to filters. She used a keyword filter with the hundred most frequently used words in English, sampled in one-hour slots over a two-week period. (Interestingly, she didn’t have access to a server to do this on in Oxford, and used a server at a different University thanks to a family connection.) Her analysis focussed on a few examples of how hashtags were used in activist campaigns such as #bringbackourgirls.

The final talk from Diana Maynard (University of Sheffield) introduced a research project, ‘Decarbonet‘, which aims not only to investigate what people think about climate change, but to ‘raise awareness’ and foster ‘behavioural change’. Diana’s part in this is to analyse discourse about climate change in social media.

Overall, it was clear that there are still some serious hurdles to accessing social media data, processing it for use with standard text analysis tools, and sharing the results and datasets. I was hoping to find examples of workflows which could access big datasets and analyze them in close to real time, but I haven’t found that yet. I’ll keep looking!

Corpus Linguistics, Context and Culture

Originally posted at blogs.it.ox.ac.uk on May 22, 2014 by Martin Wynne. IT Services at the University of Oxford has decided to delete a large number of historical blogs, and this is one of a number of posts related to the Oxford Text Archive which are being re-published here, after being laboriously retrieved from the archive provided by the Wayback Machine.

The following is an edited transcript of a very short introduction to a panel discussion in which I participated with Bas Aarts, Stefan Gries, Andrew Hardie, Christian Mair, Peter Stockwell on Corpus Linguistics, Context and Culture on 2nd May 2014.

We are now tantalisingly close to being able to process and analyse very large-scale textual resources with relative ease; these resources represent significant sections of the human cultural record; the opportunities for digital transformations of research in many disciplines are enormous.

Researchers are starting to use these resources to find and ask new research questions, as well as to address some old questions with more and new data, on a bigger scale, more authoritatively, more systematically; this is starting to happen and will happen with or without corpus linguists.To engage more effectively in these new forms of interdisciplinary research, we should focus more of our resources and attention on overcoming some important technical and methodological barriers; the main technical barriers are a lack of professional, reliable, persistent and sustainable services open to all – this is what CLARIN is trying to achieve; in terms of methodology, humanities scholars need to take a step towards working on some connected and common research programmes, addressing questions susceptible to big data approaches – this is down to us, the research community.

John Sinclair 1933-2007

Originally published 28th June, 2007 at http://wynnetours.blogspot.com/2007/06/john-sinclair-1933-2007.html.

I first met John under the best of circumstances. It was in the idyllic surroundings of his Tuscan Word Centre, and I had four days of exposure to his ideas, his barbed criticisms of my work, and his unrivalled hospitality. I could also add that it was my first visit to Italy, and that I’d also just fallen in love, and my partner, a close colleague of John’s at the time, had ensured a place for me on the course, all of which added to the glamour and excitement of the trip.

But I have to say that on first acquaintance I found him somewhat arrogant, as well as deliberately and pointlessly controversial, and I didn’t agree with him on some key points about the design and analysis of corpora. Not to mention the fact that he appeared to fall asleep every time I starting speaking. Having said that, I had a wonderful and stimulating time in Tuscany, and met several people whom I still value as colleagues and friends.

In subsequent years, I was invited back several times to teach on TWC courses, meeting many more wonderful people, hearing the same talks from John and Elena, and also reflecting on his ideas in my work, and slowly realising their value. I will treasure for ever the opportunity that he gave to me to take part in his hugely influential work at the TWC in training a new generation of scholars in his ideas. Furthermore, I had the opportunity to work with him for a short time on the TELRI project, through which I saw the enormous impact he had on corpus and computational linguistics in Central and Eastern Europe, and the huge esteem in which he was held in many countries.

Which is not to say that I came to agree with John about everything. I will now forever regret not having completed in John’s lifetime and received his response to an unfinished paper under the working title of ‘There is no degree zero of text encoding’ in response to his appeals to eradicate markup and maintain the integrity of electronic text.

In 2001 I applied for a job at the Oxford Text Archive, but was unable to attend the interview because I was teaching at the TWC. Oxford kindly allowed me to have a telephone interview, which I conducted from John’s office. I was fortunate enough to get the job.

When I came to work in Oxford, I soon received a query from Birmingham about whether I could track down in the archive a copy of John’s 1963 spoken corpus which had been collected and analysed for the OSTI report. Birmingham University Press wanted the corpus, or a sample of it, to print in the republication of the OSTI report. Unfortunately, the copy of the corpus which we held in the OTA appeared to be incomplete. As far as we could work out, what is likely to have happened is that a deal would have been made with John in which he would deposit a copy of this corpus in the archive, and in return the OTA would scan some documents for him for his next corpus-building project. OTA texts were normally collected in this way by some sort of barter process in those days. But John liked to drive a hard bargain. It appears that he must have only given us half the corpus, without anyone at the OTA realising at the time. Unfortunately, this subterfuge rebounded on him 20 years later when he found that he no longer had a copy and wanted it back. If nothing else, this now stands as an excellent cautionary tale in the field of digital preservation. (The half-corpus is freely available now from the Oxford Text Archive, catalogued as the Lexis corpus, catalogue number 0173, http://hdl.handle.net/20.500.12024/0173).

I was thrilled to have the opportunity to invite John to talk at the first event I organised in Oxford, a seminar on corpus building, and to have him contribute a key chapter to a book which I edited, Developing Linguistic Corpora: a Guide to Good Practice. John not only completed his chapter ahead of the deadline, he contributed exactly the type of chapter I wanted, and more, and was also on hand to chivvy me periodically about the delays in publication. This chapter will stand, I hope, as a summary and testament to his views on corpus design (available free online at http://hdl.handle.net/20.500.12024/2951).

Whenever my email client told me I had mail from John, I opened it with trepidation, wondering what inactivity or backsliding I was to be taken to task for this time. He was a spiky character, although I found his views always stimulating and his criticism constructive. His uncompromising attitudes and strong work ethic were a constant inspiration.

After John attended a PALA conference in Istanbul in 2003 as a keynote speaker, he confided in me that he was horrified that, to his mind, no-one was following an authentically empirical and evidence-based approach to stylistics, and typically he had no hesitation in pointing out that I was as guilty as anyone. But his criticism led to the proposal of a practical and constructive organisational solution, namely a project to invite PALA members to write a short analysis of a poem, in which every interpretative assertion is backed up by clear linguistic evidence – a project which, to my shame, I was not sufficiently well-organised to have put into operation in time for him to be able take part.

After struggling for some years to find a text which adequately presented his theories and methods, I was extremely pleased to see the appearance of Trust the Text in 2004. John’s death is a great loss to scholarship, but this work, along with the republication of English Collocation Studies, stand as great testaments to his work. We can also look forward to the posthumous release of several papers currently in press from the highly productive final period of his life.

Yet another recent successful initiative of John’s was to get the agreement of the Scottish ministry of education to make corpus resources available to every child in Scottish schools. A corpus and an analysis tool have been developed for this project. When he recently unveiled the tool, I was astonished to see that rather than simply presenting the data to the student for them to conduct their own data-driven analysis and learning, the tool will preprocess the concordance lines, find typical examples and present only these to the user. After years of training from John and Elena to read concordances to find patterns in the co-text, I was horrified. The user won’t have the opportunity to see for themselves repeated patterns of usage in the corpus. I think this is a big mistake. But I have learned to expect that, once more, I will be wrong and he will be proved right, and that this project will start another revolution the teaching of English.