A guide for the perplexed author in semantics

Note: at the upcoming LSA annual meeting in Portland, I will be part of a session about the publication process. My role will be to talk about open access in linguistics. I just remembered that I had lying around a draft guide to publishing articles. So, here’s the draft. I’d be very grateful for feedback so I can improve this document but also figure out what makes sense to talk about in Portland.

You have written something that you think other people should read. You want the input of other experts. You want your ideas to make their way through the discipline. You want the excellence of your ideas to reflect positively on you. How best to achieve those goals? What follows are my recommendations. These are my opinions only, but I have reached them over my 25 years in the discipline. I hope that some might find this subjective guide useful.

My main prescription is: disseminate early, often, and relentlessly.

Make your work available

As soon as you have a handout or slides that you’ve used for an official talk, put that on your website. When you have a draft paper, put it on your website. In each case, also point people at it through whatever networks you are part of (your department, contacts at other institutions, your Facebook, Twitter, Google+ connections).

Post to repositories

By the time you have a revised draft that is ready for submission, it’s time to also share it via disciplinary repositories (LingBuzz, Semantics Archive, PhilPapers).

Excursus: Why not stop here? Why submit to peer-reviewed journals? At some point in the future, we might reach a system where you post something to the disciplinary repository, peer review occurs in the open, and you revise your paper accordingly. The paper gets rated and assessed by various metrics (likes, number of comments, number of downloads, number of citation). We are not there yet. And to be honest, I am a firm believer in high quality peer review of the old-fashioned kind. If and when we move to the kind of free-for-all I just sketched, I’m worried that quality work from young and hereto-forth unknown authors will not get the attention it deserves. Of course, whatever the superstars put in the repository will get all the attention it deserves (and more). But the expert feedback curation from good editors and peer reviewers of good journals makes it much more likely that the discipline will be exposed to the merits of new work even if it doesn’t come from the superstars.

So, yes, submit to peer-reviewed journals. I hope they’re here to stay.

Choosing a journal

When it comes time to submit your work to a journal, the first consideration in choosing a journal is whether it is an appropriate and adequately high profile venue for the work. Journals differ along various additional dimensions: quality of peer review and editorial feedback, speediness of the review and decision process, respect for author’s rights (including the right to make available preprints and postprints), quality (and existence) of copy-editing, quality (and existence) of professional typesetting (including whether they accept LaTeX source rather than insisting on a less sophisticated format), whether or not they publish an online-first/early version of a paper as soon as it is ready, speediness of publication of print version. If you are new-ish to the field, you should ask for advice from trusted mentors.

Don’t let them lock up your work

Many journals lock up your work behind a toll access barrier. You should attempt to mitigate that lock up, because it is in your best interest for your work to be as easily accessible by as many readers as possible. It is imperative that you carefully read the publishing agreement that the publisher will ask you to sign. Make sure to understand in detail which rights to your own work you are being asked to sign away. In fact, it makes a lot of sense to include this in your decision of where to submit. There is a handy service that lets you explore the policies journals have with respect to your rights as an author: Sherpa/RoMEO.

a. Unless forced otherwise by the publisher, keep the early versions of your work (sometimes called preprints) on your website and in disciplinary and/or institutional repositories, but add to the downloads all the bibliographic detail of the published version.

b. Unless the standard publisher’s agreement already gives you the right to provide open access to your final manuscript (sometimes called postprints; the version you prepared for final submission after receiving peer review and editorial feedback, but before copy-editing and publisher’s typesetting), try to insist on that right. You can try to make your signing of the agreement contingent on the publisher accepting an author’s amendment. See the SPARC Addendum.

c. An increasing number of publishers offer to make your paper open access if you pay them a fee. This is called an author-pays hybrid open access model. There is a suspicion that publishers charge excessive author fees (“double dipping” since they still rake in subscription fees for a journal that only contains sporadic open access articles). Some universities subsidize such fees. Fees can also sometimes be charged to grants that funded the research reported in the published article.

d. There are some journals that are entirely open access and charge author fees to fund their operations (this is often called gold open access). This is an unusual model in linguistics and at this point pretty much irrelevant to publication in semantics. Some universities subsidize such fees. Fees can also sometimes be charged to grants that funded the research reported in the published article.

e. There are some journals that are entirely open access and DO NOT charge author fees. These are typically funded through institutional support (this is sometimes called platinum open access). Semantics & Pragmatics is one such journal. Others in adjacent areas are The Philosophers’ Imprint and the Australasian Journal of Logic.

Don’t get involved with edited volumes

Sometimes you might be asked to contribute your work as a chapter in an edited volume or handbook. I have done this (sometimes after having a hard time getting a paper published in a peer-reviewed journal, sometimes because it felt like an easy way to get something out, sometimes because I felt an obligation to the editors). I recommend against it. Edited volumes are a poor vehicle for cutting edge work. They are not as rigorously reviewed as top journals. They are not recognized by promotion & tenure committees as particularly impressive. They often take an enormously long time to get published, often more than even the most egregiously slow journals. Don’t do it.

Any questions or comments? Comment below. Email me. Tweet in response to the announcement of this post on twitter. Leave a Facebook comment.

DSpace statistics

The MIT Libraries just made available a new website displaying various statistics about our Open Access DSpace repository. I have three papers in that archive. Here are the download counts for those papers:

And here is a graph with the cumulative download count over time:

I can’t compare this to the download numbers that these articles get at the official publisher sites because publishers don’t seem to share those numbers with their authors. But it makes clear that DSpace is a legitimate distribution venue and that I should probably make as much of my work available through it as is feasible. Stay tuned.

Language Science Press

It’s still early days but the open-access revolution in linguistics continues. Stefan Müller and Martin Haspelmath, with seed funding from the German Science Foundation (DFG), have started an open access monograph press in linguistics called Language Science Press. E-books will be free to authors and readers. There will be a print-on-demand option as well. There are several series planned, including one on “Topics at the Grammar-Discourse Interface”, edited by Philippa Cook, Anke Holler and Cathrine Fabricius-Hansen. I have enthusiastically agreed to be on the editorial board of this series. You can support the venture in many ways. Sign up at http://hpsg.fu-berlin.de/OALI/sign/.

Statement on Aaron Swartz

We are deeply saddened by Aaron Swartz’s death, and send our condolences to all who knew him. We are very mindful of his commitment to the open access movement. It inspires our own commitment to work for a situation where academic knowledge is freely available, so that others are not menaced by the kind of prosecution that he faced. We encourage everyone to visit www.rememberaaronsw.com, a memorial site created by Aaron’s family and friends.

Scott Aaronson
Sasha Costanza-Chock
Ellen Finnie Duranceau
Kai von Fintel
Richard Holton
George Stephanopoulos
Anne Whiston Spirn

Members of the MIT Open Access Working Group

[cross-posted from the OA Working Group wiki and Scott Aaronson’s blog]

S&P acquired by LSA

[Crossposted from S&P Editors Blog]

We are excited to share good news about the future of S&P. We have
been working with the LSA on moving S&P out of its current incubating
stage to the next level with fuller support. This morning, the LSA
Executive Committee unanimously approved an agreement to that effect.

As of today, S&P is a full-fledged LSA journal, alongside Language
but independent of it. The LSA will join MIT and the University of Texas
in providing financial support to the journal. In return, S&P is to
become a journal owned by the LSA and titled “Semantics and Pragmatics”
with the subtitle “A Journal of the Linguistic Society of America”.

The day-to-day operations of the journal will not change. The current
editorial team will stay in place. The policies and procedures,
including the open access nature of the journal, will remain as they
are. Big decisions will be made cooperatively by the LSA Executive
Committee, the editors, and the S&P advisory committee.

Both the LSA and the S&P team are excited about this partnership. Open
access is the future of scholarly communication and we intend to work
together to make S&P the best journal in its field and a model for our
discipline and others.

An S&P underground classic

[Crossposted from the S&P Editors Blog:]

Semantics & Pragmatics today published an underground classic, Craige Roberts’ famous paper "Information structure in discourse: Towards an integrated formal theory of pragmatics", which had previously been published in a volume of OSU Working Papers in Linguistics, and then circulated in a slighly edited manuscript form, but was never officially published. With the help of Anders Schoubye, Chris Brown, and Justin Cope, the old manuscript was transformed into LaTeX and formatted for the S&P stylesheet. Craige wrote a new afterword and prepared an annotated bibliography, which is linked from the afterword. We’re proud to be able to make this classic paper and the supplementary material available in an official publication.

Reissuing underground classics is a worthwhile undertaking, we believe. Some famous examples are David Kaplan’s "Demonstratives" published in Themes from Kaplan, Kripke on presupposition published in Linguistic Inquiry, and in a sense also Grice’s William James Lectures. There was also volume 7 of the series "Syntax and Semantics" entitled "Notes from the linguistic underground" (edited by Jim McCawley in 1976), featuring famous papers such as Karttunen’s "Discourse referents" and gems like "Why you can’t do so into the sink" by Lakoff & Ross. So, we are continuing a respectable tradition.

Question for our audience: which other underground classics in semantics and pragmatics should S&P consider publishing? You can email us, comment on our Facebook page or our Google+ page, tweet (cc’ing @semprag), or leave a comment on our Editors Blog.

News from the open access revolution

Three items of interest on the open access front:

I think it is becoming clearer all the time that academic publishing will turn inexorably and perhaps quite quickly towards full open access. The question is which publishers and which universities, scholarly societies, and funding agencies will be at the forefront and who will lag behind.

[I will only occasionally posts news items of this nature. If you want to follow the revolution more closely, I recommend Peter Suber’s Google+ updates and Stuart Shieber’s blog.]

Citation impact 2008-2010

[Reposted from S&P Editors Blog]

We were curious to see how S&P is doing as far as the impact of published articles on the field is concerned. Below we have compiled a list of all articles published in the four main semantics journals (Linguistics & Philosophy, Natural Language Semantics, Journal of Semantics, Semantics & Pragmatics) since 2008 (the year of S&P’s first published article) that have received 10 citations or more according to Google Scholar. There are not yet any articles published in 2011 on that list. So, let’s focus on the cohort of articles published between 2008 and 2010. The four journals combined published 141 main research articles in that time frame. 54 of those (= 38%) have received 10 or more citations. S&P published fewer articles than the other three journals (11 in fact: 1 in 2008, 3 in 2009, 7 in 2010), since we’re still ramping up the quantity of publications. But S&P already has an outsized share of the top impact articles: we have 5 articles in the Top 20, and an overall rate of 64% of our articles have already received 10 or more citations.

By all accounts then, S&P’s first three years were a resounding success quality-wise. Now, we’ll be working on increasing our quantitive share of the semantics market while not decreasing our quality share. You can help: submit your best work to S&P. You will receive top-notch and fast peer-reviewing and editorial feedback and fast time-to-print. Plus, your work will be openly accessible to anyone with access to an internet connection, rather than being locked behind prohibitive subscription barriers.

[There are other things to notice, such as the domination in the upper range of articles by NLS, distancing the two older journals JoS and, especially, L&P.]

  1. [54 citations] Kehler, Kertz, Rohde et al. – JoS 2008. Coherence and coreference revisited
  2. [53 citations] Hackl – NLS 2009. On the grammar and processing of proportional quantifiers: most versus more than half
  3. [48 citations] Chemla – NLS 2009. Presuppositions of quantified sentences: experimental data
  4. [42 citations] Schlenker – S&P 2009. Local contexts
  5. [37 citations] Rothstein – JoS 2010. Counting and the mass/count distinction
  6. [32 citations] Wilhelm – NLS 2008. Bare nouns and number in Dëne Sųłiné
  7. [31 citations] Barker, Shan – S&P 2008. Donkey anaphora is in-scope binding
  8. [28 citations] Geurts, Pouscoulous – S&P 2009. Embedded implicatures
  9. [28 citations] Breheny – JoS 2008. A new look at the semantics and pragmatics of numerically quantified noun phrases
  10. [27 citations] von Fintel, Gillies – NLS 2010. Must… stay… strong!
  11. [27 citations] Rullmann, Matthewson et al. – NLS 2008. Modals as distributive indefinites
  12. [27 citations] Magri – NLS 2009. A theory of individual-level predicates based on blind mandatory scalar implicatures
  13. [24 citations] Elbourne – L&P 2008. Demonstratives as individual concepts
  14. [22 citations] Chemla – S&P 2009. Universal implicatures and free choice effects: Experimental data
  15. [21 citations] Matushansky – L&P 2008. On the linguistic complexity of proper names
  16. [21 citations] Farkas, Bruce – JoS 2010. On reacting to assertions and polar questions
  17. [21 citations] Alonso-Ovalle, Menendez-Benito – NLS 2010. Modal indefinites
  18. [21 citations] Bale – L&P 2008. A universal scale of comparison
  19. [20 citations] Nouwen – S&P 2010. Two kinds of modified numerals
  20. [20 citations] Singh – L&P 2008. On the interpretation of disjunction: Asymmetric, incremental, and eager for inconsistency
  21. [19 citations] Kissine – NLS 2008. Why will is not a modal
  22. [18 citations] Gualmini, Hulsey, Hacquard et al. – NLS 2008. The Question–Answer Requirement for scope assignment
  23. [18 citations] Harris et al. – L&P 2009. Perspective-shifting with appositives and expressives
  24. [17 citations] Abusch – JoS 2010. Presupposition triggering from alternatives
  25. [17 citations] Ippolito – JoS 2008. On the meaning of only
  26. [17 citations] Hacquard – L&P 2009. On the interaction of aspect and modal auxiliaries
  27. [17 citations] Morzycki – NLS 2009. Degree modification of gradable nouns: size adjectives and adnominal degree morphemes
  28. [17 citations] Lascarides et al. – JoS 2009. Agreement, disputes and commitments in dialogue
  29. [16 citations] Bale et al. – JoS 2009. The interpretation of functional heads: Using comparatives to explore the mass/count distinction
  30. [16 citations] Gillies – S&P 2010. Iffiness
  31. [16 citations] Brasoveanu – L&P 2008. Donkey pluralities: plural information states versus non-atomic individuals
  32. [16 citations] Moltmann – L&P 2009. Degree structure as trope structure: a trope-based analysis of positive and comparative adjectives
  33. [15 citations] Villalta – L&P 2008. Mood and gradability: an investigation of the subjunctive mood in Spanish
  34. [15 citations] Syrett, Kennedy et al. – JoS 2010. Meaning and context in children’s understanding of gradable adjectives
  35. [15 citations] Lascarides et al. – JoS 2009. A formal semantic analysis of gesture
  36. [14 citations] Arregui – L&P 2009. On similarity in counterfactuals
  37. [14 citations] Chemla – JoS 2008. An epistemic step for anti-presuppositions
  38. [13 citations] Abbott – L&P 2008. Presuppositions and common ground
  39. [13 citations] Hacquard – NLS 2010. On the event relativity of modal auxiliaries
  40. [13 citations] Chaves – L&P 2008. Linearization-based word-part ellipsis
  41. [13 citations] Moltmann – NLS 2008. Intensional verbs and their intentional objects
  42. [12 citations] Koenig, Mauner, Bienvenue et al. – JoS 2008. What with? The anatomy of a (proto)-role
  43. [12 citations] Nicolas – L&P 2008. Mass nouns and plural logic
  44. [12 citations] Dekker – L&P 2008. A multi-dimensional treatment of quantification in extraordinary English
  45. [11 citations] Nouwen – NLS 2008. Upper-bounded no more: the exhaustive interpretation of non-strict comparison
  46. [11 citations] Sharvit – L&P 2008. The puzzle of free indirect discourse
  47. [11 citations] Gualmini et al. – JoS 2009. Solving learnability problems in the acquisition of semantics
  48. [11 citations] Brasoveanu – JoS 2010. Decomposing modal quantification
  49. [11 citations] Davis – JoS 2009. Decisions, dynamics and the Japanese particle yo
  50. [11 citations] Lin – NLS 2009. Chinese comparatives and their implicational parameters
  51. [10 citations] Martí – NLS 2008. The semantics of plural indefinite noun phrases in Spanish and Portuguese
  52. [10 citations] Beck – S&P 2010. Quantifiers in than-clauses
  53. [10 citations] Zweig – L&P 2009. Number-neutral bare plurals and the multiplicity implicature
  54. [10 citations] Francez – L&P 2009. Existentials, predication, and modification

[NB: data from February 5, 2012]

Total papers published by the four journals 2008–2010: 141
54 have received 10 citations or more (54/141 = 38%)

Share of the Top 54:

JoS: 15 papers (of 41 published 2008–2010) = 15/41 = 37%
L&P: 17 papers (of 55 published 2008–2010) = 17/55 = 31%
NLS: 15 papers (of 34 published 2008–2010) = 15/34 = 44%
S&P: 7 papers (of 11 published 2008–2010) = 7/11 = 64%

Attack from big money publishers

Speaking of open access, I hope that most of you have heard about the US Research Works Act, which is a bill before Congress that would roll back the open access policies of some federal grant agencies. I urge you all to do what you can to raise awareness of this. Here is some essential reading:

My position on this is exactly the one very forcefully put by Harvard Provost Alan M. Garber:

“We endorse the view that every federal agency funding non-classified research should require free online access to the full-text, peer-reviewed results of that research as soon as possible after its publication. There are three powerful reasons to take such a step. First, taxpayers deserve access to the results of taxpayer-funded research. It is their right. Second, public access maximizes the visibility and usefulness of this research, which in turn maximizes the return on the public’s enormous investment in that research. Third, public access accelerates research and all the benefits that depend on research, from public health to economic development, manufacturing, and jobs …”

[From: http://osc.hul.harvard.edu/stp-rfi-response-january-2012]

Update (2012-01-18): Good news. Two of the big guns, Nature and Science have come out in opposition to the Research Works Act and in support of the NIH Open Access Policy. It’s very clear that Elsevier and their cronies are isolated in the scientific community, but unfortunately they have the ears of many ill-informed congresspeople.

My open access policy

MIT, of course, has an Open Access Policy, which I am proud to have played a small role in establishing. Having that policy has guided my personal decisions about venues for publishing and reviewing, but I have noticed that I have not always been very principled and consistent in my decisions. So, it is time for my own personal policy. Here it is.

Kai’s Open Access Policy

  1. Journals: I will only publish in, review for, and serve on editorial boards for journals that allow authors to deposit at least the final manuscript version (“postprint”) in an open access repository (such as MIT’s Dspace or the Semantics Archive), without any embargo (such as having to wait for 24 months before making the OA version available).
  2. Book chapters: I will personally only contribute book chapters, if the publisher allows me to deposit at least the final manuscript version in an open access repository, without any embargo. (I will consider reviewing books or book chapters that are not OA-friendly, because books are a different business from research journals, although I wish that there was more movement towards OA books.)
  3. Books: I will only publish books myself that have a significant open access component, such as making at least the final manuscript freely available, or even the final published version while charging for print versions of course.

For current reference, here are the policies of leading publishers of relevance to our field(s), culled from the MIT Libraries list of publisher policies:

  • Elsevier: With 2011 revision of author agreement, requires authors to opt out.
  • MIT Press: In full cooperation. Allows posting of final published version.
  • Springer: MIT and Springer have established an agreement that extends flexible reuse rights to MIT authors of papers published in Springer journals. Among other rights, the final submitted manuscripts of MIT-authored Springer papers can be posted openly in MIT’s open access repository DSpace@MIT. Authors should sign the standard Springer agreement and do not need to submit an author’s addendum.
  • Wiley-Blackwell: Has indicated it will be requiring authors to opt out.

Obviously, this list doesn’t include some important publishers, such as Oxford University Press. So, before making any particular decisions, I will consult with whoever is asking me to publish or review for them.