(not) writing in public

It’s been very quiet on the blog and there’s a reason for that. I thought I’d write about the reasons behind it.

At the moment I’m juggling several things – an early career researcher’s portfolio, if you will. There are numerous things that I’m involved with, some short term and one longer term, all focused on my area and which will hopefully open doors in the future. However, none of them are things I’m happy to talk about yet – or indeed discuss in detail with anyone but a few close friends and my immediate family. In some cases this is due to the sensitive nature of the project, in others it’s due to the wishes of other people on the project, in others still it’s because things aren’t fully confirmed and I’m loath to count my chickens before they hatch, in yet another yet it’s because it’s still a tentative thing and I’m not sure how closely I want to connect it with my academic identity.

If I were a cleverer writer I’d be able to write carefully, giving you enough to make me seem busy and exciting while withholding juicy details of the stuff I want to keep close to my chest. But I am not that writer, and my current situation is such that none of the things I’m involved in suitable for public consumption (or at least, not yet).

It’s all making me wonder about the nature of blogging and writing in public, and what this means for the early career researcher. Is blogging about our work always an unqualified good thing? What are the disadvantages? What does it mean to get a reputation as someone who thinks and writes in public? Can such a reputation have a detrimental effect – can it mean that you’re less likely to be trusted with classified data and with sensitive research?

It also makes me wonder about the nature of power expressed in these concerns: basically, who gets to research and/or write in public without repercussions? If I were a more senior researcher – had more clout, had the security of a permanent job – how would that change what I felt able to write about here?

Student mental health part 2

Following on from my last post about student mental health, here’s a post from the other side of teaching about making space for “quiet students”. There are some really interesting ideas there and it’s made me reflect on my own teaching practice.

When I was an undergraduate, one of the people I was taught by seemed to have an air of desperation and mute appeal whenever we scrutinised the floor rather than meet his eyes. I found it unbearable; it would make the seminars drag on (dull – I wanted to get to the interesting stuff!) and honestly, I felt kind of sorry for him. So I talked a lot (which I disliked, and disliked myself for not shutting up) but furthermore, I felt I was being forced into the position of the talkative student who risked looking daft just so that the tutor at least had something to build on. I wasn’t a quiet student, but I resented not being given the chance to be one.

When I started teaching, I was determined not to reproduce that dynamic but at the same time, didn’t want to pick on people. I think there’s room for silence as a pedagogical tool but it wasn’t something that I wanted to use on a regular basis. In addition, I knew what access requirements my students had requested; some of these requirements concerned seminar room dynamics.

Instead I got my students to talk in pairs, trios or small groups while I visited each group in turn, listening to what they had to say, encouraging them and making sure they would have something to offer. I then brought the whole seminar group together and elicited something out of each group. I don’t know how obvious it was to my students but everyone got the chance to speak in the seminar and usually did so.

Obviously this tactic wouldn’t work everywhere and in every type of seminar which is why I liked Sarah’s post so much. I also think that strategies for helping quiet students create a better environment for everyone – everyone gets a chance to participate, no one feels vaguely resentful like I did as an undergraduate, and these strategies help make a seminar a supportive environment where students can try out ideas.

Something I didn’t discuss was mental health as specific to PhD students; this isn’t due to me not caring but, rather, it being a bit too personal. Jessica has recently been writing about this – there’s more in the series, but I particularly liked PhD blues: mental health and the PhD student and Having “the chat” with your supervisor: what I talk about when I talk about depression.

I also like this post about the experience of doing a PhD while disabled or chronically ill and the sheer stubbornness it takes: disabled PhD students of the world unite, unite and take over

And yet our inability to show up has no significant bearing on our ability to contribute beautiful original things to the world. We have the experience of working successfully according to our own strategies: we must do, for how else could we be here, now? We have strategies to get around these walls in our world. We need only your support, your belief, and your acknowledgement that the stories here speak to a state of affairs whose days should be numbered.

In other words: we know how to do this. All we need is the right support, the right conditions. In this respect we are no different from any other PhD student, or any student, or any individual embarking on a project of any kind.

Every single PhD student has worked hard to be where they are. Every single disabled PhD student has had to do this work within a context where things may be harder than they are for your average bear. They are not the only ones. Nonetheless, their experiences represent a distinct category of experiences among many. As with so many things it is only by bringing these experiences before the eyes of the world that we can hope that things will ever improve.

Student mental health

I recently read a Time To Change blog post on starting university with a mental health problem and it made me wonder what advice I’d give to a student in that position.

I was an LGBT welfare officer at Nottingham and I considered it part of my role to know as much as possible about structures for student welfare and advice – everything from housing issues to sexual health – all of which stood me in good stead when I began teaching. If a student came to me with a problem, chances were that I’d know where to find information to help them – or at least know where to start looking.

Student mental health is one of the things I care a lot about and it both frustrates and terrifies me that information about student mental health can be so difficult to find on university websites. So with that in mind, here are three things that I’d especially like students to know.

Identify formal sources of support

You’ll probably be assigned a personal tutor who’ll be your first point of contact if you have any problems or issues. If you don’t click with them you can usually swap to someone with whom you’ll get on better. Some departments may have mentorship schemes where you can ask to be matched with a postgraduate researcher and have regular meetings with them.

In my experience, there is support for mental health issues in universities but these aren’t necessarily well signposted. For example, in my university, people who can help with mental health issues include department disability liaison officers (DLOs), the counselling service, the disability advisor based in the student union, the postgraduate disability advisor, the Disabled Students Network, a mental wealth group, HealthyU and a mental health advisor in Occupational Health. Because it’s a university, no one talks to each other and it seems to come as a genuine surprise to some of these that others exist.

I know, it sucks that you have to negotiate this and learn more about the arcane systems of a university than any of your peers, but on the other hand you will gain an unparalleled education in “organisational structures” and will be able to negotiate the shit out of any workplace or organisation you may end up dealing with in the future. Sorry, that’s not really much consolation.

I’d recommend talking to different people, groups and services and working out which of them are useful to you and which are not. If you don’t find a service useful or find that it actively upsets you, then don’t feel compelled to stay with it. Some people find peer support useful; others find counselling useful; others just want to check in with the DLO every so often. Whatever works for you.

Work out what you need to do to be formally recognised as disabled

Unfortunately, this is one of the things that differs between universities. You’ll probably have to contact your university’s disability office or student services to find out about this – in my experience, different departments range from incredibly clued up, helpful and supportive (like my current one) to them going “errr, you what now?” if you try asking (some others I could mention).

In the University of Nottingham, you generally start by disclosing a disability or Specific Learning Difficulty/dyslexia or a long-term medical condition. You’ll probably meet someone to discuss what reasonable adjustments you require to support your studies. Reasonable adjustments include things like getting work to you (for example, providing handouts and slides in advance of the lecture/seminar or in a different format), recording lectures, arrangements about group work, scheduling seminars, and arrangements for assessments and exams. You’ll then end up with a Disability Referral Form, which briefly outlines the nature of the condition and what sort of support you need – these then go to your department and, in turn, your tutors.

It’s a good idea to get one of these even if your mental health is well managed – the last thing you want in the middle of a crisis is someone turning around and saying “sorry, we can’t give you an extension because you’re not on record as having a disability”. Even if you don’t end up requesting different arrangements, they’re useful because they let your department know you exist – and if you ever do need additional support, it’s like a fasttrack ticket to help.

You won’t be alone in having a referral form either; both undergraduates and postgraduates have them, including postgraduates who teach. Your seminar tutor or lecturer could be among them.

Talk to us

Having taught students with mental health issues, nothing worries me like one of them seemingly disappearing off the face of the earth. I’d much rather they let me know they were alright but too anxious to attend seminars, or they’d switched medication and it was making them sleep through their alarms, or they were too depressed to leave their room. These things happen, and if the student lets the department and me know about it, we can do something about it – like move the student to an afternoon seminar group, meet them individually to help them catch up on work, reassure them about coming to a seminar or direct them to better sources of support.

It’s also better to do this sooner rather than later – let us know when there’s a problem developing rather than when you’ve got loads of work to catch up on and are feeling a horrible mixture of guilt and anxiety for missing so much work, making it impossible to approach your tutor. There are things we can do to help.

Ultimately, we want our students to get as much as possible out of university and develop intellectually, creatively and personally. Mental health problems mean that it might not be easy or straightforward, but it’s still possible to do very well – indeed excel – at degree level.

Thanks to Heather and Alex Brett for their much appreciated comments on this post

Conference bingo

Another conference season draws to a close. Heather Froehlich and I have been discussing a conference I recently attended and this is the result:

Conference bingo card

Play along!

Let us know how you get on! Feel free to leave suggestions in the comments.

Addendum: We forgot to mention food! I once spent three days miserably eating dried fruit and nuts at a conference because I’d brought that along as a snack, little realising that there would be no vegetarian option. Consider yourself winning if this happens to you, if such a thing can ever be described as “winning”.

Also, if your conference experience involves a dying rabbit and having to wash said rabbit’s blood off your hands, at that point you can reasonably be allowed to give up and just drink everything in sight.

Corpus Linguistics 2013

Ah, the biannual Corpus Linguistics conference…the biggest date in the Corpus Linguistics calendar, the event that Tony McEnery described as more of a bacchanal than a conference.

This is one of my favourite academic conferences – it was the first one I went to as a wee MA student and I’ve been to every one since. Luckily it’s biannual or my wallet would be a thinner, sadder thing. I’ve done a lot of growing up at this series of conferences – first conference I attended, first conference I was a student helper for, first conference I got funding for, first conference I’m presenting at post-viva – and it feels a bit like home turf for me.

Photo by Michael Pace-Sigge

Photo by Michael Pace-Sigge

It was wonderful to catch up with friends and to meet new people. One of the conference helpers, Robbie Love, wrote a great post on his experiences at CL2013 and I’m delighted that he found it such a friendly place.

I was presenting on my PhD research, this time trying to give an overview of how the three approaches I used in my thesis fit together: corpus linguistics, three approaches from critical discourse analysis and, underpinning all of this, an awareness of the social, political and cultural context in which the suffrage movement operated. I only had a limited amount of time to present so decided to show elements of the corpus linguistic analysis and some of the work I did on Emily Wilding Davison using Theo van Leeuwen’s taxonomy of social actors. However, I argued that without an awareness of the historical context, any analysis of these texts would be lacking.

I’ve uploaded the slides so have a look if you’re interested.

One of the cool things about big conferences is that you assemble your own experience. It reminds me a bit of big music festivals with headliners/plenaries most people make an effort to see, then dozens of interesting people and projects. That up-and-coming person you’ve been advised to see, that fascinating project you’ve seen an announcement about, that person that you’ll see because you know they’re a fantastic presenter, that thing you’ll go to because it sounds controversial and you suspect the response will be good, the friend you’ll go and watch for moral support, the person you used to work with back in the day and want to find out what they’re doing now, the one that you’ve never heard of but looks intriguing… Talking to someone else, I realised that it was almost as if we were at two different conferences! It was, once again, fascinating to see how many things corpus linguistics can be used to investigate.

Some of my highlights were:
Michael Hoey giving a characteristically energetic opening plenary
An intense discussion of p-values following Jonathan Culpeper and Jane Demmen‘s talk
Paul Baker‘s beautifully neat demonstration of triangulated methods
Claire Hardaker on trolling (she refused to give the site name of one place, referring to it only as the “internet hate machine” – naturally I knew exactly where she meant)
Alison Sealey on the People, Products, Pets, Pests project
An incredibly lively morning presentation from Laurence Anthony on AntConc
A whole team from Lancaster represented by Paul Rayson on the Metaphor in End of Life Care methodology
Sylvia Jaworska on representing the Other in tourism

Over the course of the conference we also drank a bar dry and drank 180 bottles of wine at the gala dinner, so good work everyone.

And, because Michael P-S claims it’s not a conference without me lurking on my laptop somewhere…

Photo by Michael Pace-Sigge

Photo by Michael Pace-Sigge

Photos by Michael Pace-Sigge can be found here.

My worst student

Today I’ve been grumpily following the Same-Sex Marriage Bill debate, contemplating my thesis corrections and pouring acetic acid into my sore ear so I have to admit, I’m not in the best of moods. And then I saw that the Times Higher Education decided to encourage academics to share their stories of their worst students on twitter. Aside from the obvious problems about professionalism and ethics, I don’t like the sneering.

You see, I was someone’s worst student.

Not in university – I’d mostly sorted myself out by then – but in sixth-form. I was doing an A-level in something I’d previously been good at and for which I was in the top set at GCSE, but at A-level my grades plummeted from As to Es. I couldn’t understand the material – I tried so hard and it constantly defeated me. I tried reading around the subject; I tried talking myself through it; I tried just knuckling down and memorising it. It slipped away from me, no matter what I tried or how hard I tried. It’s a horrible feeling to be so utterly powerless – to feel like your intellect has abandoned you, that whatever you try you’re going to fail, that you are stupid and worthless and wasting everyone’s time. It felt like being dropped into a world where the rules were opaque and all-powerful and I was constantly one crucial step behind. Every lesson was an ordeal, something to just survive for the next two hours and, eventually, to resent.

Naturally, the teacher and I loathed each other. He’d taught me when I was actually good at this stuff and in retrospect, probably couldn’t understand why I was suddenly so appallingly bad at it. If I was in his position and a student had suddenly gone from being one of the best in the class to the very worst, I would have sought help for this student. He didn’t. Instead he alternately ignored me – it was a very results-focused school and I was clearly not going contribute to his clutch of As – and bullied me. Because I was 17 and a bit of a twat, I made it quietly clear that I resented him every bit as he resented me. One day I snapped and told him that I wished I could drop this subject (I was very polite in my twattishness!). The next day, my Head of Year pulled me into her office and berated me for hurting his feelings.

I threw up every morning before school from the sheer anxiety of once more stepping into that classroom and once more, being utterly, helplessly adrift.

I have never, ever forgotten that feeling of being so totally lost. Not when I found a subject I loved, not when I got a First for my BA, not when I graduated from my masters, not when my examiners shook my hand after my viva, and especially not when teaching.

I’ve taught students who were uninterested, resentful and hungover – it’s one of the problems of teaching a compulsory Language module when most of the students would rather be doing Literature. I’ve occasionally got frustrated when marking. I know I don’t have much experience, but I hope I never get worn down by it. I hope I create an atmosphere in my seminars where students can make mistakes, test ideas that might not work or admit that they don’t understand something. I hope I can be sensitive to the students who struggle, and I hope I know when I’m out of my depth. I hope I never belittle or sneer at students – not when they frustrate me, not when they apparently don’t try, not when they appear to be hopelessly bad at something. I hope I am respectful and compassionate to the ones who resent me. I hope I am patient when it matters.

As an academic, we tend to be working in an area we love and that we’re good at. We’ve probably never been crushingly bad at something we now teach. Given the kind of grades we’re expected to get to enter a degree programme, then a Masters, then a PhD, we may never have been crushingly bad at any academic subject. I, with my E in that A-level, somehow sneaked in. I’m not proud of that grade, but the harsh lessons I learnt in that classroom have shaped my teaching forever.

Other posts:
Caroline Magennis: On Teaching
Kirsty Rolfe: Talking teaching on Twitter (and talking nicely to students)

The white male professor is in

Perhaps presciently, in my penultimate post I noted that “[a]s a student, I have never been taught by someone with a non-European non-white background – and when I teach, I am incredibly aware that this may have been the case for my students”. This week, UCU released a report focusing on women and Black and Minority Ethnic (BME) professors:

UCU summary
Report: The position of women and BME staff in professorial roles in UK HEIs (pdf)
Guardian: The university professor is always white

Some of the figures they highlight include:

  • Just one in five professors are women, despite making up almost half the non-professorial academic workforce
  • Just one in 14 professors (7.3%) are from a black and minority ethnic (BME) background
  • White applicants are three times more likely to get a professorial post than BME applicants

Naturally, I find these figures troubling – and, if I’m honest, not a little dispiriting.

The UCU report notes that BME UK nationals are particularly underrepresented – unfortunately, their data isn’t presented in a particularly helpful way to interpret this. Universities are brilliant places for worldwide collaboration and I’ve been lucky enough to work with people from all over the world. However, I do think it’s worth focusing on UK nationals because it highlights failings in our own education system and university recruitment processes. Appointing more international BME academics would be great for diversity but I’m also concerned that it would lead to universities failing to take a very careful look at recruitment of UK BME academics and the barriers that stand in their way.

As a recent example, this article on discrepancies in attainment between BME and white students was published two months ago. I find that really troubling – to me, if this is happening across an entire cohort of students, it suggests that something is going wrong at an institutional level. And crucially, if BME students are leaving university without the Firsts and 2.1s necessary for postgraduate study, that suggests problems for a future generation of UK BME academics – namely, that they won’t be there.

There are a few points I’d like to make about the report.

  • Firstly, while the UCU breaks down figures into “Black”, “Asian”, “Chinese”, “Other Asian” and “Other” it’s not clear how these groups are defined – who, exactly, is included in the “Other” groups? The terminology itself is…less than sensitive (we’ve heard of post-colonialism and Othering yeah?). For that matter, it’s not clear who’s included in the “Asian” category – I’m assuming people of Indian origin, but what about people of Pakistani, Bangladeshi, Sri Lankan origin? To me, there’s a lack of clarity about who was included in the report.
  • Secondly, this is the sort of area where an intersectional analysis would be really helpful; basically, what happens to applicants who are BME and women? Is their experience of institutional discrimination on two fronts reflected in their employment rates, or is something else happening?
  • Thirdly, a breakdown by subject area and discipline would be beneficial. My gut instinct is that STEM subjects might be a bit better than art and humanities at employing Asian and Chinese professors, but without data I’m wary of generalising in such a way.

I think this is a useful starting point but there are so many questions this report doesn’t answer. It’s clear that there is massive underrepresentation of women and BME academics at the highest level of academia – are universities going to do anything about that?

P.S. Thank you Nina Simone

Doing interdisciplinarity

Second in what seems to be an occasional series about interdisciplinarity. All posts can be found under the interdisciplinarity tag

One of the most daunting things about my thesis was that I essentially had to learn a new discipline. I could have treated my corpus of hundred year old newspaper articles like a contemporary corpus – the corpora I’d used until then (the British National Corpus, the Guardian corpus and two corpora I’d assembled myself, one of music reviews and the other of children’s stories) were of my time and cultural context and, while I obviously researched the area, I didn’t have to learn about a different time period.

However, I knew that if I was to do any kind of (critical) discourse analysis with these historical newspaper texts, I had to learn about the historical context they operated in. If I didn’t, I would be incapable of recognising discourses – they simply would not register as significant to my contemporary eyes. I would not understand their significance, or the impact they had. The nuances would be lost on me and my thesis would make for possibly an interesting corpus analysis, but useless for anyone interested in the historical or social aspects.

The trouble was that I hadn’t done any history since GCSE. At that age I was pretty sure that my interests lay in warfare, sickness, medicine, death and social history, and not in kings, queens and the nobility. The prospect of being taught by my history teacher for another two years was not a prospect I was willing to entertain. Despite loving history, I reluctantly chose not to study it at A level. As such, I had a lot to learn.

Be aware of the field(s)

My first task was simply to understand the historiography – who was writing about the suffrage movement and how they positioned themselves within the field. I tried to understand the debates within the field, and how people’s research responded to others’ research. I learnt something of the waves of research about the suffrage movement – the first waves comprising of personal memoirs and scholarship focusing on the WSPU, London-based organisations and leading figures of the movement, and a second wave starting in the 1970s and focusing more on hitherto marginalised figures, experiences, ideologies and organisations. There were also historians seeking to synthesise these various perspectives.

I also sought to understand the wider context of the time period – what was going on in terms of politics, family life, Empire, work, health? What were the assumptions about gender roles and how women were supposed to behave? These were the things that were going to come up in the newspapers and which I had to be acquainted with.

One of the ways I did this was to read – a lot. The other was to go along to undergraduate history lectures. These were useful in supplying context and when it came to the lecture on the suffrage movement, I was delighted that it was all stuff I was newly familiar with. This suggested that I was on the right track.

I am also immensely grateful to Dr Chris Godden and Dr Lesley Hall who were generous with their time and advice and who didn’t laugh at me when I described what I was researching, but were kind enough to give the impression I had something useful to offer.

Allow it to shape your research on a fundamental level

In my case, I’d played it safe and requested Times Digital Archive data from between 1903 and 1920. Choosing dates was always going to be arbitrary; 1903-1920 encompasses the period of time from the formation of the Women’s Social and Political Union to two years after the Representation of the People Act 1918 which gave women the vote. If I wanted a lot more data, I could have asked for 1860 – the decade in which the campaign for women’s suffrage emerged out of other social protest movements – to 1928, the year in which women were granted the franchise on the same terms as men. That was too much to take on for a PhD project, and I had to narrow my focus. I could have simply chosen a five or ten year period – say 1910 to 1915 – but this wouldn’t have made sense in the social context of the time.

My research in history quickly revealed that the outbreak of World War One led to a complete change in the suffrage movement; it forced them to engage with nationalism as revealed and understood through warfare. It was also clear from looking at the initial data that the newspaper’s shift in focus and the amount of news they actually printed changed after 1914. Including the years between 1914 and 1918 would therefore change the scope of the project and that was too much to take on for a PhD project. My research into history also gave me a starting year – 1908, on the cusp of suffrage direct action.

The historical background, therefore, played a crucial role in shaping the project and delineating its boundaries.

It’s not just the icing on the cake

Related to the previous point: I didn’t want the historical analysis to be something I only brought into play in conclusions.

This ties in with a corpus linguistics issue of having to know your data in order to do in-depth work with it. If you handed me a corpus from a culture that I know very little about – let’s say, for sake of argument, Arabic literature – I would probably be able to make some pretty graphs and identify some words which behave in interesting ways and which are perhaps worth further investigation, but because I don’t know the history and the context in which these texts operate, I would be unable to connect these interesting words to things that were happening that would affect the discourses. I wouldn’t know about events, wars, peace treaties, rulers and governments, innovations, philosophies, schools of thought, cultural shifts.

While my work is in (corpus) linguistics – my background is in linguistics, I’m (currently!) based in an English department and my thesis will be examined by linguists – I don’t think I could write a good linguistics thesis using this particular corpus without also writing about history. It’s something I’ve always had in mind when interpreting data – and not just in the sense of “writing about what I’ve observed”. In corpus linguistics, categorising and grouping your data can in itself be a form of interpretation; I found it was vital to understand the historical context in order to come up with meaningful categories. At the moment I’m looking at the news narratives of Emily Wilding Davison’s actions, hospitalisation, death, inquest and funeral procession; these were all reported, sometimes extensively, in the Times. However, a critical discourse analysis approach requires the analyst to be sensitive to what is missing as well as to what is present, and so it is vital for the analyst to be aware of the wider context of the text. This was where an understanding of the historical context is so important. For example, the Times notes that Yates represented Davison’s family at the inquest into her death. A passing comment, perhaps. However, with my knowledge of the suffrage movement, I was able to recognise this man as Thomas Lamartine Yates, husband of one of Davison’s close friends and unofficial WSPU legal advisor. In some ways, this raises more questions than it answers – was he providing his services because of his wife’s association with Davison or does it say something about Davison’s posthumous changing relationship with the WSPU? – but it makes for a richer analysis.

To go with the cake metaphor, I don’t think of my research as a linguistics cake with a tasty history icing. Nor do I think of it as a linguistics and history battenburg cake, nor a linguistics and history marble cake. instead, the two are inseparably mixed.

Be humble

I am not a historian by training. I have undoubtedly made mistakes and been blind to things – interpretations, undercurrents – that a person with 5+ years of intensive study of history behind them would notice. There’s a lot that I don’t know, and if someone asked me to discuss in detail the politics of Empire during this time period or the campaign for equal franchise in the 1920s, I’d struggle. I hope that I’ve been a respectful and sensitive outsider and my work can offer some useful insights to “proper” historians, but I also hope I can accept their inevitable corrections with grace and humility.

Humanities, sciences and interdisciplinarity

First in what seems to be an occasional series about interdisciplinarity. All posts can be found under the interdisciplinarity tag

suffrag* and words statistically associated with it, calculated through Mutual Information (MI)


A couple of weeks ago I read this article about treating humanities like a science and was a bit annoyed about it. In my experience, the big sweeping claims as illustrated in that article tend to be made by a) arts & humanities scholars who’ve suddenly discovered quantitative/computational methods and are terribly excited about it or b) science-y scholars who’ve suddenly discovered arts & humanities and are terribly excited about it. I’ve heard a fair number of papers where the response has been “yes, and how is this relevant?” because while it’s been extremely clever and done something dizzyingly complex with data, it’s either telling arts & humanities people stuff they already know or stuff that they’re not interested in. In my particular discipline people are very aware of the limits of quantitative work and we acknowledge the interpretive work done by the researcher. I do think quantitative methods have a place in arts and humanities, and in this post I’ll discuss some of the strengths of quantitative work.

Firstly, I should say something about my background and where I’m coming from. I’d describe myself as an empirical linguist – I look at language as it’s used rather than try to gain insights through intuition. My background is in corpus linguistics which basically means I use computer programs to look at patterns in large collections of texts. If this sounds suspiciously quantitative then yes…it is. Sometimes I look at which words are statistically likely to occur with other words, or statistically more likely to occur in one (type of) text than another, or trace the frequency of words across different time periods. My thesis chapters tend to have tables and graphs in them. I sometimes talk about p-values and significance.

However, these patterns must be interpreted. Computers can locate these patterns but to interpret them – to understand what they mean for language users – needs a human. As a discourse analyst, I’m interested in the effect different lexical choices have on the people who encounter them. I’m interested in power, in social relationships and in the ways in which identities and groups are constructed through language. A computer would find it difficult to analyse that.

So what can be gained from using corpus linguistics rather than purely qualitative approaches? Paul Baker outlines four ways in which corpus linguistics can be useful: reducing researcher bias, examining the incremental effect of discourse, exploring resistant and changing discourses, and triangulation

reducing researcher bias

Language can be surprising. We have expectations of how language is used that isn’t always borne out by the data. My MA dissertation looked at how male and female children were represented in stories written for children, focusing on how their bodies were used to express things about them. So, for example, I looked at his eyes and her eyes and what words were found around them. What I was expecting was that boys would be presented as active, tough and independent and girls would be presented as more emotional and gentler. What I found was that a) his eyes was much more frequent in the data than her eyes and b) that male characters expressed much more emotions than female characters. Part of this was because there was so much more opportunity to do so because of the higher frequency of his eyes, but the range of emotions – sorrow, joy, compassion – was really interesting and not what I was expecting from the research literature I’d read.

We also have cognitive biases about how we process information and what we notice in a text. We seek evidence that confirms our hypotheses and disregard evidence that doesn’t. We tend to notice things that are extraordinary, original and/or startling rather than things that are common or expected. If we select a number of texts for close, detailed analysis, we might be tempted to choose texts because they support our hypothesis. A corpus helps get around these problems by raising issues of representation and balance of its contents.

examining the incremental effect of discourse

Michael Stubbs, in one of my favourite linguistic metaphors, compares each example of language use to the day’s weather. On its own, whether it rains or shines on any particular day isn’t that significant. However, when we look at lots of days – at months, years, decades or centuries worth of data – we start finding patterns and trends. We stop talking about weather and instead start thinking in terms of climate.

Language is a bit like this. On its own, a particular word use or way of phrasing something may seem insignificant. However, language has a cumulative force. If a particular linguistic construction is used lots of times, it begins to “provide familiar and conventional representations of people and events, by filtering and crystallizing ideas, and by providing pre-fabricated means by which ideas can be easily conveyed and grasped” – through this repetition and reproduction, a discourse can become dominant and “particular definitions and classifications acquire, by repetition, an aura of common sense, and come to seem natural and comprehensive rather than partial and selective” (Stubbs 1996). A corpus can both reveal wider discourses and show unusual or infrequent discourses – both of which may not be identified if a limited number of texts are analysed.

exploring resistant and changing discourses

Discourses are not fixed; they can be challenged and changed. Again, corpora can help locate places where this is happening. A study using a corpus may reveal evidence of the frequency of a feature or provide more information of its pattern of use – for example, linking it to a particular genre, social group, age range, national or ethnic group, political stance or a small and restricted social network. A changing discourse can be examined by using a diachronic corpus or corpora containing texts from different time periods and comparing frequencies or contexts; for example, where a particular pattern is first found then where and how it spreads, if a word has changed semantically, has become more widespread, is used by different groups or has acquired a metaphorical usage.

triangulation

Finally, triangulation. Alan Bryman has a good introduction to this (.pdf) but it basically means using two or more approaches to investigate a research question, then seeing how closely your finds using each approach support each other. I tend to use methodological triangulation and use both quantitative and qualitative approaches. As well as supporting each other, using more than one method allows for greater flexibility in research. I like being able to get a sense of how widespread a pattern is across lots of texts but I also like being able to focus very closely on a handful of texts and analyse them in detail. It’s a bit like using the zoom lens on a camera – different things come into view or focus, but they’re part of the same landscape.

I find quantitative methods fascinating for the different perspective they offer. My background in corpus linguistics has also trained me to think about issues like data sampling, choosing texts to analyse and cherry-picking evidence. It’s taught me to think critically about what and how and why people search for in a text, and it’s made me methodologically rigorous. At the same time, dealing with so much data has made me very sensitive to language and how it’s used in different contexts. I think the author of that article might find some of the work in corpus stylistics fascinating – this is what my supervisor is working on, and having worked a bit with her corpus it’s easy to see how much qualitative literary analysis goes into it.

Returning to the article, I think this raises wider questions of how we approach interdisciplinarity, how we locate and approach research questions in fields not our own, and how we relate to colleagues in these other fields who are experts. If we are to engage in interdisciplinary research, then we are bound to be working in unfamiliar areas. We are going to encounter research methods and ways of thinking that are unfamiliar to us. The ways we approach things will have to be explained – why should a humanities scholar care about “a bunch of trends and statistics and frequencies”? How do we make these relevant to their interests and show them that these can both answer interesting questions and open up new avenues of research? Simultaneously, how do we gently make someone aware that they’ve just dipped a toe in our field and that there’s still much to learn?

This is something that I’ve had to learn. I’m not a historian by background or training, but my area of research deals with historical issues. I’ve had to more or less teach myself early 20th century British history; I did this through extensive reading, gatecrashing undergraduate lectures and talking to historians. In a future blog post I’ll discuss this further so if you have any questions, let me know and I’ll do my best to answer.

References:
Baker, P. (2006). Using Corpora in Discourse Analysis. London: Continuum.
Stubbs, M. (1996). Text and Corpus Analysis. Oxford: Blackwell

Open Access

Today the government announced that publicly funded scientific research should be publicly available for free. In principle, I think moving to an open access system is a good thing. However, like many others, I have reservations about the type of OA the Finch report recommends. Mark Carrigan has an extensive round-up of coverage and reaction and has a set of slides giving a good overview of the situation.

As Steven Harnad of LSE writes,

There are two ways for authors to make their research OA. One way is to publish it in an OA journal, which makes it free online. This is called “Gold OA.” There are currently about 25,000 peer-reviewed journals, across all disciplines, worldwide. Most of them (about 90 per cent) are not Gold. Some Gold OA journals (mostly overseas national journals) cover their publication costs from subscriptions or subsidies, but the international Gold OA journals charge the author an often sizeable fee (£1000 or more).

The other way for authors to make their research OA is to publish it in the suitable journal of their choice, but to self-archive their peer-reviewed final draft in their institutional OA repository to make it free online for those who lack subscription access to the publisher’s version of record. This is called “Green OA.”

As Beverly Gibbs writes, the high cost of Article Processing Fees places a structural barrier for early career researchers trying to get their work published. Martin Eve asks whether “publisher boycotts [will] offer up at least one generation of early-career researchers to the sacrificial slaughter so that the cycle can either be broken or, more likely, continue once more”. And Mark Carrigan wonders about the researchers who can’t pay.

As someone working in Arts and Humanities, I’m worried that the way we do research and publish means I’ll find it harder to get Article Processing Fees funded. Unlike other disciplines, we’re less likely to work in groups with a big name attracting the funding and I’m curious as to how this will affect early stage researchers trying to publish single authored work. If there is only limited money in the pot, then as a lowly PhD researcher I’m not sure how I’m meant to compete with far more experienced and higher ranking academics.

However, I am glad that academia is finally having this discussion. I’ve been interested in creative commons, copyleft and open access for about a decade now, and the current model of academic publishing has been a constant source of frustration. One of the things that shocked me when entering university as an undergraduate was how locked academic publishing is – it seemed years behind the open licensing I’d encountered online and it was an unpleasant surprise to suddenly come up against such restrictions.

The practice of accessing journals on behalf of friends in other institutions who don’t subscribe to that journal seems pretty pervasive, and I wonder just how widespread this is. It suggests that while publishers try their hardest to restrict access, academics are willing to find ways around these restrictions. In light of this, I wonder how much loyalty academics feel towards the current publishing model, particularly PhD researchers and early career researchers who have grown up with filesharing, peer-to-peer networks and alternative licensing systems.

While Gold OA seems problematic, I don’t think the wider issue of open access to research will – or, indeed, can – go away.