The OWL and the Rice Farmer

In a recent email I was asked by a European Linguistic Professional SIL Colleague about the ease of use of ELAN.

Is ELAN easy enough to use, so that an Indonesian rice farmer can use it?

I am offended with the ‘rice farmer’ terminology which is prevalent in SIL discourse. However, to understand this term, and its usage I think it is important to understand another term used prior to the rice farmerThe Ordinary Working Linguist. This older phrase has gained some traction as evidenced by its adoption in the linguistic literature1, 2, 3, 4, 5, 6, 7, 8.

The OWL or ‘ordinary working linguist’ is attributed by George Lakoff to Charles Fillmore as a self-describing term. Though it is impossible to say if Fillmore actually coined the term. If he did, then it is an interesting frame by the discoverer of cognitive framing. I have yet to meet an ordinary linguist. I have met smart linguists, extraordinary linguists, novel linguists, quirky linguists, geeky linguists, and ornery linguists — but never an ordinary one. The second issue I take with this contrived frame is that it suggests that somewhere there might be a non-working linguist. Now, either I have never met a lazy linguist, or I have been foiled by some collective endeavor. That is, the linguists I have met have somehow all decided that I am an extremely boring person — the kind to be avoided, and all have feigned busyness! Granted busyness is not the same as “working”, but I have never met a linguist who is not engaged in some sort of activity which has a mistakable similitude to work!

All the same, the notion of “OWL” seems to be very useful to a large group of linguists. For instance, Gary F. Simons suggests that Helen Aristar Dry proposed a book in the early 1990s with the title Computers and the Ordinary Working Linguist.9, While David Nathan, Peter Austin, Andrew Black, Piotr Bański, Mike Cahill, Jeff Good, Steven Moran, Scott Farrar, and René van den Berg all use the term in their works.

In honor of Fillmore, if we take a framing view of the phrase “ordinary working linguist” we can see that many of the authors are discussing the concept within the frame of linguistic computing — the computer tools that linguists use to do their work. However, it can be argued that there is an SIL specific shift in the framing of the phrase (usage by Cahill, Schroeder, Simons, Black, and van den Berg). Van den Berg’s use is most articulate:

the typical SIL fieldworker is what has been called an ‘owl’, that is, an ‘ordinary working linguist’ who has been given enough training to enable him or her to do the job; this is usually a minimum of two semesters of field linguistics at graduate level. A typical SIL field worker is therefore very much a generalist, without specialist knowledge in any branch of linguistics.

Van den Berg’s usage is both confirmed and then contrasted by SIL International on its website where it contrasts the term with Linguistics Coordinators, Linguistics Consultants, and International Linguistics Advisors. It is clear to see that within the shared frame by SIL staff, that Linguistic Consultants are those who have more training than typical SIL field workers. For several reasons, including reduced staffing, and paradigm shifts within project management related moral ownership within the tasks of language development, SIL has moved from ex-pat driven projects to nationally staffed and ethnolinguistic community driven projects. SIL’s minimally trained “OWLs” moved up the organizational chart to become project managers/facilitators, leaving open the “bottom” of the chart, ergo the new need for ‘rice farmer’ as a newly perceived class of computer user.

I find the term rice farmer 14 times in the internal SIL International wiki and I find “OWL” 83 times. These terms surface mostly among SIL computer app developers, but I also hear them used by SIL policy and procedures developers, SIL program planners, and SIL educators. To me this term comes off as a class distinction term (and maybe there is nothing wrong with class distinction; I mean the Jews had class distinctions at the time of Jesus, the Cretans had classes at the time of Titus, the British have it, and castes are alive and well in India, etc.10). However, as I find the term being used in SIL, a more transparent term might be ‘low socio-economic and low educational capacity individual’.

The problem is that in my work with Steve Marlett in Mexico, I worked with ethnolinguistic moronity language users who were not ‘low educational capacity individuals’ rather they were high educational capacity individuals. When I went to Malaysia and worked with Sean Conklin, the same thing, ethnolinguistic minority language users were not low educational capacity individuals. When I went to Nigeria and worked with people like Becky Paterson and David Rowbory, the people doing language work with us were not low educational capacity individuals. In fact the Nigerian rice farmer/pastor/student/translator/cultural guide/language activist who was on our language project team can and does use ELAN after some training (and uses Facebook without training to boot — FYI: I need training on ELAN too!). So, why then do SIL staff use terms like ‘rice farmers’ when they talk about designing software for users? Something seems to be wrong with this picture. My guess is that there is a real reason why SIL staff use the term, but it is not the reason that SIL staff attribute as the causation of the observation of “low educational capacity individuals” - as we observe dark skinned colleagues who try and use SIL software and don’t intuitively know how to use the tools. My guess is that the reason for the use of the term is because it is part of the SIL corporate ethos. SIL staff use (or at least have in the past) their educational courses at various SIL training events as filters to find the ‘best and brightest students’ and recommend those for various field service within SIL. Then in the advanced classes SIL staff obfuscate the materials just to make them harder, to see if those same students are able to still keep up. So, as I see it, there seems to be this undefined class difference in SIL’s corporate sociology preferring those who can distill obfuscated communications under stress and dis-preferring those who cannot distill obfuscated signals under stress.

So for the following four reasons I strongly dis-prefer the terms ‘rice farmer’ and ‘OWL’ as they often appear in SIL discourse:

  1. I don’t think the ‘rice farmer’ terminology is actually very useful for professional reasons. Software design is about communicating. It is about communicating relational concepts in a visual context. It is about connecting new analogies with salient concepts. User Interface (UI - what a software user sees and interacts with) is the gateway to User Experience (UX - the emotional, logical, task based context of activities that are attempting to be facilitated by the software designer and accomplished by the software user). Therefore what UX is really about is communicating ideas and creating positive and successful interactions. If the UI fails to create the designed UX then it doesn’t matter if the user is a rice farmer or a Ph.D. candidate from an Australian university. The end result is that the user is frustrated and will not use the software willingly. Therefore, singling out the ‘rice farmer’ from the white guy is not really a useful construct because all users, whether black, white, tan, farmer, or stockbroker, are subject to the same constraints and frustrations of software which is not intuitively implemented. It is my experience that programmers (not just SIL programmers) usually do not intuitively know how application users think or behave. UX design is an art and a science that takes experimentation and a solid grasp of information design principles.
  2. Let’s not forget that, at least in software design we are talking about the ‘rice farmer’ with a computer, an iPad, or an android device.
  3. We never use the American Corn Farmer, or the Norwegian Reindeer Rancher as analogies, and I am a bit tired of singling out people from tropical areas because their skin is darker, or their average height is shorter, or because they grow rice in their backyard. The focus on the ‘rice farmer’ does not help us deal with the white guy’s problem with the software and robs us of a balanced view of the real communication challenge.
  4. Using the term ‘rice farmer’ seems to put the ethnolinguistic minority language users and nationals in a sub-standard class relative to ‘real linguists’ or ‘real missionaries’. Yes, my African colleagues who do have their own farms and families, also use SIL software; I first drove a tractor at age 13 and grew up on a farm, so what is the difference? By making the ‘OWL’ or the ‘rice farmer’ as some uninitiated or limited capacity person, the “reason” for new design (i.e the reason that WeSay was created to work with FLEx) doesn’t solve the fact that even for highly-educated-white-professionals FLEx is still confusing. The issue is a visual communications issue not a smarts or intelligence issue. Linguistic concepts can be communicated about beautifully and communicated simply, both in the classroom and in software.

I hope this post motivates those in SIL to take a deeper look at User Experience issues in the software tools and websites we build.

  1. Bański, Piotr. 2010. Why TEI stand-off annotation doesn’t quite work: and why you might want to use it nevertheless. In Proceedings of Balisage: The Markup Conference 2010 (Balisage Series on Markup Technologies 5), n.p. Montréal, Canada: Balisage. ↩︎

  2. van den Berg, René. 2013. SIL and its contribution to Oceanic linguistics. Language and Linguistics in Oceania 5. 33–74. ↩︎

  3. Black, H Andrew & Gary F Simons. 2008. The SIL Fieldworks Language Explorer Approach to Morphological Parsing. In Nick Gaylord, Alexis Palmer & Elias Ponvert (eds.), Texas Linguistics Society 10: Computational Linguistics for Less-Studied Languages, 37–55. Stanford, California: CSLI Publications. ↩︎

  4. Cahill, Mike C. 2001. Farewell from the Coordinator. Notes on Linguistics 4(4). 203–204. ↩︎

  5. Farrar, Scott & Steven Moran. 2008. The e-Linguistics toolkit. In Workshop in the 4th IEEE International Conference on e-Science, Indianapolis, Indiana, 10 December 2008, n.p. IEEE Computer Society. ↩︎

  6. Good, Jeff. 2010. Valuing technology: Finding the linguist’s place in a new technological universe. In Lenore A. Grenoble & N. Louanna Furbee (eds.), Language Documentation: Practice and values, 111–132. Amsterdam: John Benjamins Publishing Company. ↩︎

  7. Nathan, David & Peter K Austin. 2004. Reconceiving metadata: language documentation through thick and thin. Language Documentation and Description 2. 179–188. ↩︎

  8. Schroeder, Leila. 2008. Bantu Orthography Manual (SIL E-Books 9). Revised 2010. Dallas, Texas: SIL International. ↩︎

  9. Which would be later published as: Aristar Dry, Helen & John Lawler (eds.). 1998. Using Computers in Linguistics: A Practical Guide. 1st edn. London: Routledge. ↩︎

  10. The United States doesn’t have classes, only more successful capitalists and less successful capitalists. — ⟨snark⟩ ↩︎

Hugh Paterson III
Hugh Paterson III
Collaborative Scholar

I specialize in bespoke research at the intersection of Linguistics, Law, Languages, and Technology; specifically utility and life-cycle management for information products in these spaces.