Day 2  *photo*  Gwama Rapid Word Collection workshop arrangement of roles

The number of participants in this workshop seems to have settled at 17, not counting the six of us who are non-Gwama (Anne-Christie—who is the Coordinator and helps answer everyone’s questions and is primarily responsible for quality control; two typists—David, and the wife of an SIL employee; two government employees—who will be responsible for continuing the Gwama word-collection after these two weeks; and me—functioning as record-keeper and as consultant when questions arise). So there are 23 of us working in the very large meeting room I described previously. I am so thankful that we have such a large space to work in.

We are seated around six tables arranged around the room. 4-5 participants are seated around each of three tables oriented parallel to each other near the platform at the one end of the room; they form the word-collection groups. Three other participants are seated at a table toward the back of the room (near the entrance); they are adding Amharic glosses (short definitions) to the Gwama words collected by those toward the front. Two other tables are situated between the word-collection groups and the platform, but perpendicular to them; I sit at one of these, doing the record-keeping; one of the typists sits at the other, doing data entry. David does his share of the data entry in a separate room because it’s less draining for him if he can work in solitude, rather than in the presence of other people.

You may have observed that we added a typist today who was not present the first day. Since David was able to keep up by himself on Monday with the output from the two glossers, adding yet another typist meant we needed to increase the number of glossers, as well. Thus, Anne-Christie asked one of the members of the word-collection teams to change roles and help with the glossing. For the better part of the day, the glossers were not quite able to keep the typists steadily supplied with work, so there were periods of time where the typists were waiting for the next folder to be completed by the glossers and had time on their hands. By the end of the day, however, that seemed to be changing, and I’m expecting that tomorrow we’ll see that there will normally be at least one folder waiting for a typist to claim it.

Yesterday, we began to see folders accumulating in the glossers’ “inbox,” as two people glossing for three word-collection groups was an insufficient number. Today, with three people working on glossing, the glossers’ output seemed to pretty well match the rate at which new material was given to them. Thus, I have the impression that we struck a good balance today, with three word-collection groups, three glossers, and two typists (one of whom is a fluent touch-typist).

As Anne-Christie warned me ahead of time would be the case, the ‘weak link’ in this workshop is the level of fluency in Amharic of the Gwama participants. In this part of the country, Oromo is used much more as a trade language than is Amharic, so while those taking part in this workshop do speak Amharic, they do not have the depth of vocabulary and understanding in that language that they do in either Gwama or Oromo. In practical terms, this means that they are often struggling to understand what the questionnaire (written in Amharic) says, so they have difficulty understanding the topic (semantic domain) for which they are being asked to collect words. Therefore we have added another step in our assembly-line process: Anne-Christie looks over the words collected before the folders are given to the glossers, screening the content to be sure that the words are correctly categorized, and crossing out any that are not. This saves the glossers valuable time that they would otherwise spend translating words that don’t need to be translated. (These words will inevitably be collected when the correct semantic domain is treated, and they’ll be glossed there, so they will not be “lost.”) The most salient example of this was when the names of various carnivores were written on the response sheet for the semantic domain intended to collect names used for various kinds of meat that people eat (e.g., beef, mutton, veal). The intent of the semantic domain was simply misunderstood by the group that dealt with that topic. They understood that it had to do with eating meat, but they misunderstood who was doing the eating.

In news peripheral to the workshop itself, I think we have found a ‘favorite restaurant’ here in Asosa. I had a very nice omelet there the other day, and today Anne-Christie had vegetable soup (with vegetables in it that we haven’t seen in the local market, so we’re wondering where they got them), David had rice and sauce (the first time we have been able to order that in any of the restaurants where we’ve tried, and we have tried several times), and I had ‘tɨps’ (roasted beef in broth) with bread. (See photo.) David spoke for all of us when he said, “We’ll definitely come back here again!”

Photo of Johnny Walker, from previous workshop in Amaaro, eating roasted beef in broth with bread.

written by Kevin Warfel