Garrett and Harris on assessing scholarship in documentary linguistics

New paper just out in Language:
I’ve just started to skim it and it looks like there are many points that this forum would be interested in! It might be a good candidate for a running review/discussion like was done with Adrienne Tsikewa’s paper earlier this year. Some nice points about documentary work in comparison to other disciplines (e.g. how a new edition of Chaucer “counts”, so editions of glossed and annotated text should count too).


I gave it a brief read earlier this week, and also thought that it would be a great candidate for a “running review/discussion” here.
I’m on-board!


It is great! (I read a draft a while ago) I think some running review/discussion would be very useful.

And if I can do a bit of shameless self-promotion, there’ll be an LSA webinar later this month about it. The webinar was only mentioned in passing in the last LSA update—official announcement still to come—but you can already sign in for it:

“The latter article by Drs. Garrett and Harris will also be the subject of a webinar on September 28 entitled, “Tenure, promotion, and academic review in documentary linguistics.” (More details on the webinar to follow)”


Maybe we can have a space here for follow up discussion after the webinar? (Unless every issue gets resolved by the end of the webinar :wink:)

1 Like

I for one would definitely love to see another discussion here!

I do wonder if people will be Zoomed-out immediately after the webinar though. What if we:

  1. Set up a topic for discussing the paper right away
  2. Set up a topic for discussion after the webinar
  3. Then, after people have time to think about it and chat in the threads, we could set up a discussion about both the paper and the webinar

I’m interested. I have two papers I’m presenting on this topic in Berlin in October. So, I’m interested to hear what is said here.

That reminds me, what happened with the JOHD special issue? Last time I checked there was only one paper there

1 Like

@cbowern There are two papers currently. The special issue is still open for submission. No one has left the project. One editor got really busy with work, another editor had an increase in family size. It seems that I have the most time at the moment (to give two papers, right!?!). But the two papers I am presenting include all the editors of the JOHD special issue as authors, and they have been really constructive and involved as co-authors. It is in the special issue’s editorial team’s plans to place a position or summary paper in the collection but we wanted to wait till there are a few more papers before deciding on a final direction for that. This is still a relatively young conversation.

My presentations will be videoed so maybe a link here might be of interest. You can find the abstracts linked from my list of upcoming talks: Recent & Upcoming Talks | Hugh's Curriculum Vitae

If someone wants to make a submission to the special collection, please do so! I’ll likely be the one to find peer-reviewers for the manuscript. There is still lots to say on the topic both from the linguistics side and the archivists side.

The LSA had their own committee (which we knew was happening), I suspect this paper is part of that general LSA discussion. Maybe this paper will sparke some response ideas in readers and we can publish those at JOHD.

1 Like

Just thought I’d chime in here with some of my own thoughts on the paper. In general, “Yes!”, everything discussed in the paper is good and I agree with it. I think that most of what is stated here is already somewhat well established within the LD community, so I get the feeling that one of the goals of this article might be simply to broaden the awareness of the idea of assessing LD collections and recognizing their value.

For me personally, some of the most interesting questions come from the details, such as the assessment process/criteria. Such a detailed discussion isn’t within the scope of this article, but I feel like it is the most exciting part of what they are describing, and it is one of the remaining pieces of the puzzle which have yet to be well defined. When we say “An archival collection can be assessed by other documentary linguists” (p. e12), what exactly does that look like? Is it for a “corpus journal” as Haspelmath and Michaelis proposed,? Or single-authored “book-review style” publications? Or something else?

I’d also be interested to see more discussion about how LD collection assessment criteria might be defined and balanced against different community/fieldwork/project contexts, and what other community feedback processes there could be (beyond the letters of support mentioned in the article).

It’s great that these ideas are being shared on such a prominent platform, and I look forward to seeing what ideas come out of the resulting conversation. If anyone is interested in more content on this topic, or even perhaps interested in writing an article on the topic, here is an old call for papers for the special issue that @hp3 and I are editors of (and still accepting submissions for!): JOHD Special Collection on Language Documentation - call for papers - Google Docs


Please feel free to start a topic on this!

(& hi Richard good to see you around these parts :grin:)

Hi everyone,
Thank you @hp3 for the update on the JOHD special issue. I actually wrote to Lauren yesterday to ask what was happening with that.

@rgriscom, in my slides for the panel, I’ll share one way in which I tried to get collections assessed for tenure. My own take is that it should not be through a published description of the collection (Haspelmath & Michaelis, Thieberger and even Garrett & Harris suggest something along these lines); in my view, that defeats the purpose of trying to get these “non-traditional” outputs assessed because then you’re being assessed on a (traditional) published output.

I think the discussion here has already been really good and I hope people can make it to the webinar on Wednesday (everyone is welcome; you don’t have to be an LSA member to register for the webinar).


@JROSESLA That is a great counterpoint to consider. We’re open to considering publishing counter points to :wink: Actually, the point here is that non-peer-reviewed works ought to be part of professional recognition. One of the things we have been encouraging potential authors to consider is how these professional situations are structured in other disciplines. One thing that I have always been intrigued by is how the field of archaeology works… That is, do archaeologists get “credit” for each dig they do, or procedure funding for? How do they get professional credit for fieldwork?

Here are two presentations on the topic:

1 Like

Jorge that’s a great point. I keep coming back to the parallel of the critical edition of a text. When someone in a language department publishes one, they don’t have to publish an article about the book in order to get it cited; the book stands on its own. Should be the same for a documentary collection


Garret & Harris discuss “source transparency” referencing the Tromsø recommendations. Those recommendations include a paragraph on what to cite which suggests an option of an individual citation in the reference list for each text used in a paper:

“Sometimes it is desirable to provide a reference to an entire resource that may be comprised of numerous components (e.g. files or folders), while at other times it is desirable to provide separate citations to the individual components that were used.”

Has anyone taken the “separate citations” approach? Would this approach serve the purpose of raising more awareness about who did the work to create corpora that were used in publications?


I find it really difficult to apply the traditional practice of “crediting by citation” to data. The “best” datasets from my point of view are datasets that come with a CC-BY license, because these allow derived versions. If you use the derived data, do you also cite the original data? And how many steps are you expected to go back in the provenance chain?

1 Like

In my 2019 IJAL paper in which I used audio data from 1 recording deposited in AILLA, I cited that specific resource (and not the entire collection). That also has the advantage of recognizing the speakers who helped create that particular record and not just the researcher (which is often how collections are cited).


I use individual citations (with collection identifier, file bundle/folder, file name, and when I can easily access it, timestamp) for all linguistic examples. Journals such as LD&C that require source transparency have been happy with this. Journals that usually don’t require that, such as Journal of Child Language, Language, etc. have also been fine with letting me do it. The reasoning here is that I welcome readers to look at the examples for themselves, and readers will need that level of specificity in order to find the example in the collection. So this is more in the spirit of open science/OA in general than specifically from a documentary motivation.

My articles usually also involve analyses that abstract over an entire collection, e.g. calculating the frequency of word X per 100 or 1,000 words. In those cases I follow conventions used in psycholinguistics/corpus ling papers, which is that the creation of the corpus/collection is described in a separate methods section and that section cites the entire corpus.