Last month, Steven Pinker’s The Sense of Style was published in Britain and the US. As a specialist on books of usage advice, and as someone who needs to write for a living, I find The Sense of Style interesting because it is partly a book about writing and partly a book with straight-forward usage advice. I also think it’s very important that linguists get more involved in this business of usage advice, and reclaim some authority as it were, from prescriptivists who are underinformed about linguistics.
I have to admit that I wasn’t overwhelmed by Pinker’s writing advice, which despite much solid advice, based in linguistics and cognitive science, isn’t always as concrete as I would like it to be. But perhaps this is because as a linguist, some of his insights on how sentences and texts are constructed did not strike me as particularly new or controversial. It probably will for someone without a background in linguistics.
“I love style manuals. Ever since I was assigned Strunk and White’s The Elements of Style in an introductory psychology course, the writing genre has been among my favourite literary genres.”
However, this is not the part of the book that I want to address here. Right now I’m mostly interested in the last – and longest – chapter of the book. “Telling right from wrong” takes up almost a third of the book, and is more or less a usage guide on its own. What I like about this chapter and – the book in general – is the linguistic arguments Pinker uses and the language myths he debunks with them.
“The idea that there are exactly two approaches to usage—all the traditional rules must be followed, or else anything goes—is the sticklers’ founding myth.”
Pinker tackles old chestnuts such as the use of flat adverbs (such as in “drive fast”), and those about the distinction between two and more than two. This category is represented by usage problems such as the use of alternative for more than two choices, the choice between between and among, and each other and one another, and whether or not it is okay to use a superlative in an “X of two”-type phrase (“may the best of the two win”).
Pinker says he loves style manuals. I have a different relationship with them: I think they’re interesting, but I don’t necessarily feel the need to take them home with me. Nevertheless, The Sense of Style will end up on my bookshelf.
The past weeks we have been looking into the ‘hugeness’ of the HUGE database. We want to show how many usage guides were in the database by year, what kind of usage problems there are and how many of each category.
Currently the HUGE database covers 240 years of usage advise with 78 usage guides from 1770 until 2010. To give an overview of the distribution of usage guides in the database I made a table of the first editions by year.
As can be seen in the chart it looks as if the 1980s and 1990s are over-represented with fourteen and fifteen first editions of usage guides respectively. However, the number of publications do increase in these two decennia. Of course usage guides were published in the years 1790-1820 and 1880-1900 but these were newer editions of previously published guides.
Most usage guides published in the beginning of the 1980s deal with British English but in the late 1980s more American usage guides were published. The boom in usage guides could be a counter movement against the closure of state funded grammar schools during the 1960s and ‘70s. It seems that this trend in commercially published guides started in the UK and later thrived in the US as well. In the 1990s a more even distribution in guides can be seen in both the UK and the US with eight and seven first published guides.
The usage problems in the database are tagged with various linguistic levels: syntax, morphology, etymology, spelling, lexis, semantics and punctuation. The complete usage problem list can be found in the manual. The usage problem list consists mainly of grammatical problems in the English language so it is not surprising to see that most usage problems in the HUGE database are concerned with problems relating to syntax and morphology. The second most prominent linguistic level is semantics. This relates often to problems between a choice of words. As the main focus is on the grammatical problems, the usage problems in the database are least concerned with spelling and punctuation.
Usage problems were furthermore tagged on more specific problem definitions. The word cloud below gives an overview of these definitions. The more prominent a certain problem tag, the bigger the word. According to the data the problem definitions number, concord, agreement, redundancy and comparative are in the top five tags.
If you are interested in using the database for your research you can request access here.
The next Bridging the Unbridgeable lunch lecture will take place on 21 October 2014, at 13:00 in Lipsius 308. During this session Amanda Delgado Galvan, a PhD candidate at LUCL, will introduce the language annotation tool ELAN and show how it can be used in language research.
ELAN is a tool that allows you to create, edit, visualise and search annotations for video and audio data. It was specifically designed for the analysis of languages and it can be used by anyone who works with video and audio data for the purposes of annotation, documentation and analysis.
All those interested in the topic are welcome! Bring your own lunch!
I always thought this is what the COD looked like:
Until yesterday, when I found a lovely, what might be described as a vintage copy of the book on the Free Books Table we have on the second floor of the English Department at Leiden:
What a lovely cover, I wonder who designed it. I also wonder who left it on the table, or who the previous owner, A.P. Parker Brady, was. The book, printed in 1928 (1st edition 1911), was bought at a local bookshop in Lochem, a small town in the east of The Netherlands. The title page is also worth showing here:
Like The King’s English (1906), the book was produced by the two Fowler brothers together, and at a time when the Oxford English Dictionary on which it is based hadn’t even been finished yet. You can read all about the writing of the COD in Jenny McMorris’s The Warden of English (2001), Henry Fowler’s biography.
So: thank you anonymous donor, for this lovely little book! And do let us have more such books if anyone has any to spare. They will be in good hands.
In the next couple of months, I will be conducting research on diachronic changes in English usage at the University of Freiburg by looking into the “Brown family” of corpora.
The Brown Corpus of Standard American English was the first of the computer readable, general, million-word corpora compiled by Francis and Kučera at Brown University in the 1960s. Its creation inspired a number of mirroring “family” of corpora, including its British English counterpart, the Lancaster-Oslo-Bergen (LOB) corpus. In today’s terms the one million-word corpora are considered to be rather small in size, however, back in the early days of corpus linguistics, the Brown corpus, which was stored on 100,000 punched cards, represented an emancipatory step in studying naturally-occurring language.
To my great and pleasant surprise, my new office is the home to the original texts which make up the two newer editions to the Brown family, the Freiburg-Brown (Frown) and the Freiburg-LOB (FLOB) corpora dating from 1991. I am looking forward to looking into them, through automatic searches instead of page turning, of course.
Alex Salmond at polling station
Yesterday Scotland has voted and decided to stay within the United Kingdom. Today newspapers are filled with punchy and informative headlines analysing the outcome of the Scottish referendum. When I was reading an article in The Independent, my eyes fell immediately on two little words: concede defeat.
You may ask yourself now what’s the big deal or what’s wrong with this expression. And you are rightly doing so. There is nothing ‘wrong’ with concede defeat. But a related expression, concede victory, is found troublesome by a few. In my research on the BBC I have come across the ‘usage problem’ of using concede victory instead of concede defeat. The BBC style guide declares the use of “concede defeat” as “wrong” and favours the use of concede victory. Intrigued by this issue, I have decided to see how concede is actually used in the reporting on the referendum vote in the online media.
Doing a simple google search for “Salmond concedes defeat” and “Salmond has conceded defeat” results in 4,030 and 412 hits respectively*, while the same search with “Salmond concedes victory” and “Salmond has conceded victory” show a completely different picture. The search including the present tense of concede remained fruitless, while the later search resulted in a meagre 8 hits. What was striking from these results is not only the low numbers for concede victory, but also the fact that concede victory does not seem to be a British problem. It was found only on the Irish RTE.com and wn.com homepages**. This is surprising as I was expecting to find some results from the BBC News homepage. What does the BBC use then? Have they changed their mind about the use of concede victory?
The OED states in his entry on concede that concede means to admit defeat, to acknowledge that an election is lost. If you take the search further and investigate the BNC, you will see the same pattern as in the online media search. Concede defeat occurs more frequently than concede victory.
click to enlarge
A simple glance at the BBC homepage instantly answers the question concerning the BBC’s stance on concede. Instead of using concede altogether, the journalist smartly used accept defeat. The BBC style guide also advises on this usage in order to avoid “the problem”. It seems that the BBC is aware of the controversy about the use of concede victory, which still raises the follwowing questions. Why does the BBC advises its journalists to use concede victory if concede defeat is more frequently used? Is the BBC fighting a long lost cause?
* Google searches were conducted in the morning of the 19/9/2014, so the numbers of hits are subject to change.
** wn.com: Shows results in google search, but cannot be retrieved from the news provider itself.
We are moving house this summer, and while packing up the books in my study I came across an article I wrote nearly thirty years ago but that I had completely forgotten about. It is about usage problems and their function in J.D. Salinger’s The Catcher in the Rye (1951). (Prescriptivism has clearly been one of my lifelong academic interests!)
I had written it as part of a Festschrift for Sarah Betsky-Zweig (1923-1999) when she retired from the chair of American literature at the Leiden English Department in 1986. The article deals with double negation, subject pronouns in object position, between you and I, and he/she/it don’t, and it argues that Salinger used features like these (there are many more in Holden’s language) to depict different types of speakers in a sociolinguistically highly convincing way. All these features, with the exception of he/she/it don’t, are dealt with in English usage guides, as anyone consulting our HUGE database will be able to see for themselves.
With Salinger (1919-2010) having been so much in the news recently (a biography by David Shields and Shane Salerno came out a year ago and five new novels are expected to be published posthumously) I thought I’d revive the article, to confirm Salinger’s brilliance as a writer, and I slightly updated it in the process. So here it is: “Between you and I: Non-standard grammar and The Catcher in the Rye“.
Salinger is no longer alive, nor is Sarah Betsky. But my copy of the book also contained her thank-you note, in which she made an interesting comment:
You made me remember the strange censorship of Catcher. In the 1950s we were forbidden (upon pain of being fired) to teach Catcher at the university in the U.S.A. Imagine! It would corrupt student language!
Hard to imagine these days that books would be forbidden just because they make a character confuse lie and lay, use like as if, singular they and that’s for whose!