Tom Pepinsky

Category: Research

  • Academic Publishers: Speech Acts, Coordination, and History

    From Inside Higher Ed: the editors and editorial board of a top-ranked linguistics journal, Lingua, are resigning en masse to protest their publisher’s exorbitant pricing.

    The editors and editorial board members quit, they say, after telling Elsevier of the frustrations of libraries reporting that they could not afford to subscribe to the journal and in some cases couldn’t even figure out what it would cost to subscribe. Prices quoted on the Elsevier website suggest that an academic library in the United States with a total student and faculty full-time equivalent number of around 10,000 would pay $2,211 for shared online access, and $1,966 for a print copy.

    This is big news, and it’s not the first time that Elsevier has come under criticism (see here). And the plan is a smart one: the editors are creating a new journal called Glossa in an effort to transfer the reputation and quality of Lingua to an open-access model.

    (One peculiar thing is that apparently there already is a linguistics journal called Glossa. I have no idea how that is related to this new Glossa journal.)

    Now, note the following statement in the Linguist List announcement, from Stefan Müller of the Free University of Berlin, which targets any linguists who might agree to step in as the new editors of Lingua.

    You may be flattered by the offer of Elsevier but think twice: the good reputation of the journal was built by researchers like us. This reputation is now transferred to the new journal.

    This statement is what linguists call a speech act. By stating that the reputation has moved to the new journal, Müller attempts to make it so. The perlocutionary intent is to convince other linguists to act in ways that are consistent with Glossa being the successor of Lingua—to submit their manuscripts there, to accord it respect and prestige, and so forth. There are some direct threats to any potential collabos too.

    Being on a newly established board of a journal that remains after the former board moved to Ling-OA will not look good on any CV. I would not hire anybody who did something like that and I would object in any search committee I am involved in.

    Obviously the act of creating a new journal itself will not change reputations, which is why persuasion is so important. The challenges, though, are many. The first is a coordination problem inherent in moving reputation from one journal to another. Let’s say that I would prefer to treat Glossa as the successor of Lingua. But if I am a self-interested scholar, should I submit my manuscripts to Glossa? Only if I am sure that other linguists will do the same, and that they will accord my publications there the same amount of respect that they have customarily accorded to Lingua.

    Is that possible? Some scholars will surely make this switch. But so long as rankings such as the infamous Australian Research Council’s ranking of ALL THE JOURNALS exist, in which Lingua is an A journal and the already-existing Glossa is a C journal, treating Lingua like Glossa will run directly against the professional and financial incentives of many linguists (in this example, the Australians, but surely others in citation-metric-obsessed national university systems too).

    We can conclude, then, that coordination problems will be substantial. But there is a second challenge here too. In the end, somebody is going to end up running Lingua, and the journal will therefore continue to publish, and it will continue to hold Lingua‘s backlist. Even if people stop submitting their best papers to Lingua, and stop reviewing for them, and transfer reputation to Glossa, it will still be the case that every time anyone cites an article published prior to 2015, it will go towards Lingua‘s impact factor and various other citation metrics. Again, so long as academics and administrators accord value to impact factor, we can expect history to make it difficult to strike at the heart of Elsevier’s business model.

    Overcoming the weight of history will be even harder than overcoming coordination dilemmas. If there are any linguists out there reading this, I do have an idea. The first issue of Glossa should be entirely comprised of articles that are citations of Lingua articles.

    Here is an example of what I have in mind. Take the recent article, “The prosodic expression of focus, contrast and givenness: A production study of Hungarian,” by Susanne Genzel, Shinichiro Ishihara, and Balázs Surányi. Glossa should invite Genzel, Ishihara, and Surányi to publish a one-line article of the following type.

    The prosodic expression of focus, contrast and givenness: A production study of Hungarian

    by Susanne Genzel, Shinichiro Ishihara, and Balázs Surányi

    This article is a citation of Genzel, Ishihara, and Surányi (2015).

    References
    Genzel, Susanne, Shinichiro Ishihara, and Balázs Surányi. (2015) “The prosodic expression of focus, contrast and givenness: A production study of Hungarian,” Lingua, Volume 165, Part B, 183-204.

    Now there is an article by Genzel et al. in Glossa. Authors interested in transferring reputation from Lingua to Glossa can cite the Glossa version. Lingua still exists, but if the editors of Glossa are successful at changing reputations, this is a way to strike at the citation metrics too.

    UPDATE

    Via twitter, David Mainwairing observes that Thomson-Reuters might not count these new suggestions as “citable items.” I think he’s got a good point, but the consequences aren’t necessarily that bad—except for in one specific (and unfortunately distributional) sense.

    The main goal is to remove the citations from Lingua. Even if Thomson-Reuters does not consider tiny Glossa articles as citable items, this mechanism does remove citations from Lingua. Glossa would only gain in the metrics through new full submissions—which, if reputation is transferred, will still happen, just more slowly than if Thomson-Reuters did recognize tiny Glossa articles as citable items.

    The problematic distributional consequences are for scholars whose work appears in old issues of Lingua. My proposed tiny Glossa articles mechanism will mean that their citation metrics will suffer so long as Thomson-Reuters does not recognize their new citations. The challenge for the ex-Lingua editors, and for linguists interested in protesting Elsevier through a new open-access, is to think through all of the consequences of moving to a new journal. In a world of citation metrics as administrative shortcuts for allocating funding and prestige, some distributional consequences are unavoidable.

  • Write Like You Imagine That You Talk

    Via Chris Blattman, I came across a post by Paul Graham called “write like you talk.” The advice here is sound in spirit, but is a bit misleading.

    I know this because in a previous life I took a course in sociolinguistics, taught by one of the world’s great demolishers of the idea that anyone ever utters anything that actually makes sense. One of the most memorable class activities was to record and then transcribe actual conversation. And then, upping the stakes, to record and then transcribe a formal lecture. I recorded and transcribed a conversation among my friends about a tamagotchi (all the rage at the time), and then a formal lecture by a professor widely regarded as being one of the best lecturers on campus.

    When you do this, this is what you learn. People do not talk in complete sentences. They certainly do not talk in paragraphs. People rarely have a point, and almost never build up to that point in any systematic way. A good lecturer (unlike a pleasant conversationalist) will have an overall structure, yes, but actual delivery comes with asides, anecdotes, incomplete sentences and stutters, and ums and buts, and my new enemy, “so, …”.

    Graham’s advice should not be to write like you talk, but rather to write like you imagine that you talk. Here is the distinction. You imagine that when you talk, there is structure to what you are saying that can be broken down into a series of component claims that flow naturally from one to the next. You imagine that you use proper grammar and clear language that is just precise enough but never obscurantist. You imagine that you have a point. Imagine if your writing did the same!