Being a book editor in the digital age: an interview with Graeme Hirst

Sorry, no French stuff in this one–this is an interview with Graeme Hirst, a well-known researcher on language and computers and the editor of a popular series of books on that subject.  Come back in a day or two for a return to the subject of the implications of the statistical properties of language for second-language learners and how that plays out in one person’s life (mine).

Graeme Hirst is a professor of computer science at the University of Toronto. From a theoretical perspective, his research is focussed on lexical semantics, especially problems in the representation of linguistic and semantic knowledge; text classification, the study of how to categorize bits of language into things that can tell us something (do the contents of this note tell us that the writer is experiencing cognitive decline?); and the linguistic nature of argument. He is an internationally respected expert on quantifying the extent to which words are semantically similar to each other (his paper on this has been cited over 1400 times), how we use “chains” of related words to make discussions of a topic cohesive, and how we can build computational models of language that can help us understand how people make sense of ambiguities by using what they know about words. He has also put this theoretical work to very practical use in his work on methods for detecting Alzheimer’s disease, or cognitive decline, by looking at long-term changes over time in people’s writing. (Some of this description is lightly edited from his web page.)

Graeme is also the editor of a series of books called the Synthesis lectures on human language technologies. In that capacity, he has edited 35 books, written by authors ranging from the already famous to the subsequently well known. (Some books in the series that I’ve enjoyed include Linguistic Fundamentals for Natural Language Processing: 100 Essentials from Morphology and Syntax, by Emily Bender; Semantic Role Labeling, by Martha Palmer and others (full disclosure: one of the authors was my advisor–still anonymous!); Recognizing Textual Entailment: Models and Applications, by Ido Dagan and others; and Semantic Relations Between Nominals, by Vivi Nastase and others.)

Prior to this, Graeme was for more than 25 years, book review editor of the journal Computational Linguistics. So Graeme has been in a unique position to observe the dynamics of publishing books in a world where people don’t necessarily start their investigations of a topic with a trip to the library, as we did when I was a younger man, but by doing a Google search and looking for a free tutorial to read. Graeme was kind enough to let me interview him by email; here is what he had to say on subjects ranging from what the editor-author relationship is like in a technical environment, an editor’s perspective on writing, and business models for book publishing in a world where it’s not clear that technical people read books any more.

Me: How do you see the role of the editor of a non-fiction book?

Graeme: There’s no single answer to that, even assuming that we are talking specifically about academic books. Also, do you mean the editor of a single multi-author volume or (given your earlier allusion to book series) the editor of a series?

In this kind of editing, there is a spectrum between gatekeeping and (what I’ll call for lack of a better term) programming. The former is typical of academic journals: the editor’s job is merely to enforce quality control and all relevant papers that are scientifically acceptable (by some threshold that depends in part on how much space is available in the journal) are published, without regard to the specific topic of each paper. There is no attempt to “balance” the papers by topic within an issue or across issues or to ensure that any particular topic is covered. That’s in contrast to a popular magazine, say, where variety is important, and so is entertainment or interest value. In such a magazine, the editor will decide what topics to cover partly from his/her own knowledge of what’s going on and from listening to article pitches from writers. Writers will then be commissioned to write articles, with a view to creating (or “programming”) a magazine that pleases its readers with an interesting variety and which covers topics that “need” to be covered. Not all commissioned articles will necessarily be published.

An academic book series can be much like the academic journal, where scientific quality is the main consideration rather than variety or covering specific topics — although the assurance of a market for the book is usually taken into account too. It costs a lot more to publish a book than a journal article, and even non-profit publishers, such as university presses, need some assurance that costs will probably be covered by proceeds from sales. (That consideration obviously doesn’t apply to predatory publishers and vanity presses that for a fee will just toss any author’s camera-ready copy out into the world, or will charge the author for all production costs, without any editorial judgement at all.)

But my personal approach has been in the middle of the spectrum — not exactly a popular magazine but with some of the same aspects of “programming”. Since the series I edit is intended to be largely tutorial in nature, rather than research monographs, I actively identify important topics, especially emerging topics, for which an introductory book would be helpful to the research community and then try to find (sometimes successfully, sometimes not) appropriate authors to write it. And I also gratefully receive suggestions and proposals from potential authors. So I aim for both variety and coverage. I would not want to have two books on the same topic in the series, as that kind of redundancy and competition would not be fair to the authors who put a lot of time and work into their book.

Quality is also an issue. For my series, I am conscious of the series as a “brand”, and I want readers to have confidence that if a book is allowed into the series then it will be a good one and worth their while. Not all books can be equally totally wonderful, but bad books will tarnish the series and have a bad effect on the other authors. For scientific quality control, like any editor, I have to rely on the reviewers, as I can’t be sufficiently expert on most of the topics. I also care about quality in writing, especially as these are introductory books aimed at a relatively wide audience within the research field — graduate students reading for an advanced course, or a researcher who wants to come up to speed rapidly on a new topic — and I do my best to help authors improve their writing where I can. Production quality matters, too; typos and sloppy formatting reduce credibility. The series is professionally copy-edited and typeset, but it helps if authors deliver a Latex source file that is as error-free as possible.

Me: How do you see the future of book-writing and publishing in a world where it’s not clear that technical people read books any more?  I’m not sure that my students ever read books on topics in our field unless I make them do it–their natural impulse seems to be to go to Google and look for a web page that will explain it to them quickly, and without the hassle of a trip to the library or the cost of buying a book.

Graeme: It’s a cliché that no one has the time or attention span any more to cope with anything long, and books are no exception to that.  And as books become ever more expensive, we want a really good reason to lay out our money, and we’re loath to do that if we need only some parts of the book.  It’s no different to the rebellion against buying CDs or record albums; why pay a lot when there’s only one or two tracks I want, and I can download those cheaply (or free).  Meeting this demand was one of the appeals of the Synthesis series for me — books on very focused topics in (mostly) just 100–150 pages, that set the topic out for the reader far better than he or she could do just by putting together the results of a Google search.  The result, I hope, is books that work both for the reader who is motivated by fascination with the topic and for the reader who just has a pragmatic need to know the material without a deep level of interest.  But the future of book writing seems secure.  There is no shortage of people who have lots to say about topics that they are passionate about and who want to tell the world.

Me: Do you have thoughts about the distinction(s) between editing a non-fiction book and editing a work of fiction? I’m guessing that you don’t have the kind of relationships with your authors that I read about fiction writers having with their editors–you paying their rent, advising them on their love life, etc. On the other hand, I’m guessing that you do have some aspects of the novelist/editor relationship–say, advising people on how to get over “writer’s block,” procrastination, dealing with reviewer comments, and the like.

Graeme: I don’t have any experience in editing fiction. And I’m usually too remote from my authors’ lives to offer the kind of advice that more typically characterizes the relationship between an advisor and a graduate student. Also, it is explicitly my publisher’s job, not mine, to cajole, nudge, hector, and badger tardy authors.

As with most things we do, writers tend to underestimate how long the work will take. But there are some who seemingly just keep putting it off, or can’t organize themselves to put in the time when there is always a more-urgent deadline for a conference paper or similar that needs to be attended to first. What always astonishes me is the people who will put a lot of work into writing 50% or 80% or more of a book, and then simply can’t bring themselves to complete it, thereby throwing away an enormous amount of time and effort.

Me: You mentioned that you try to help authors improve the quality of their writing. How does that work? What are you looking for, where quality is concerned? Do you see more/less in the way of problems from academic writers? Computer scientists versus linguists? What’s your level of tolerance/support for the somewhat impenetrable writing that we academics often get rewarded for?

Graeme: Despite what I said, there’s actually not all that much that I can do as an editor to influence writing except to point out general principles, often rather superficial ones, such as “refer to people, not papers”. One occasional problem in multiple-authored books is that the authors use different terminology or different mathematical conventions and symbols in their respective parts, and I now instruct groups early on about how to maintain consistency and avoiding redundancy where each author re-introduces certain material at the start of their part.

I’m astonished how much I need to do at the formatting level. Many authors need to be told to proofread both their BiBTeX entries and the resulting bibliography. Or how to use math mode properly. Or design a table logically.

Me: Indeed, as a reader, I find the whole multiple-authors-using-different-terminology-and/or-symbols thing really frustrating, and occasionally a non-trivial barrier to comprehension.

So, lessons learnt, then: what general (or specific) advice might you offer to authors up front?

Graeme: It’s faster in the long run to do it properly the first time than to go back later to correct all the little inconsistencies and errors.

Me: What would you most want authors (and readers, I guess–or other editors) to get out of this conversation?

Graeme: Don’t say yes unless you mean it. Finish what you start. Aspire to publish a book in the Synthesis HLT series.

Me: What do you wish I’d asked you that I didn’t?

Graeme: “Welcome to Paris. Will you join me for dinner at my favorite Michelin-starred restaurant?”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Curative Power of Medical Data

JCDL 2020 Workshop on Biomedical Natural Language Processing


Criminal Curiosities


Biomedical natural language processing

Mostly Mammoths

but other things that fascinate me, too


Adventures in natural history collections

Our French Oasis


ACL 2017

PC Chairs Blog

Abby Mullen

A site about history and life

EFL Notes

Random commentary on teaching English as a foreign language

Natural Language Processing

Université Paris-Centrale, Spring 2017

Speak Out in Spanish!

living and loving language




Exploring and venting about quantitative issues

%d bloggers like this: