Sunday, November 29, 2015

Corrected Bambara poster; thinking about best practices

 Bambara poster on hand-washing from International SOS
In an earlier posting, I highlighted a poster in Bambara on hand-washing, which had evidently been composed in a non-standard 8-bit font making it all but unreadable (see image on top right). I have written International SOS, on whose site the poster is available, and Translators Without Borders (TWB) which is listed on the poster, to signal the problem.

Since the issue has already been a subject on this blog, I would like to follow through with presentation of a corrected version (already shared with TWB), in the spirit of 2Ds+4Rs (reasons to repost African language materials). I'd also like to use this to raise the larger issue of the best licensing for such public education materials in languages that are less-resourced and/or less-widely spoken.
Text from poster with tracked changes

This material is © 2015 AEA International Holdings Pte. Ltd. I am sharing it, as I have other materials in African languages from other sources, with attribution, in my understanding of fair use. However, this is not ideal. It would be much more helpful for the original purposes of creating these materials to use a Creative Commons license to encourage the kind of thing I'm demonstrating here, as well as more substantive re-use of such material. Would it be possible for organizations involved in authoring and/or translating extension and public education materials in African languages - not only for health but also other domains of social and economic importance - to agree on a set of "best practices" for facilitating their review, dissemination, revision, and re-use?

For this particular document, what I did was to copy the text into a Word document and first run 3 search and replaces, since the problem with the document stemmed from it being in an old font that had certain letters substituted with "special characters" needed for Bambara. These were q ɛ, x → ɲ, \ → ɔ. Then less than a dozen minor copy-edits, mainly changing capital "I" to lower case "i" - the software on which the original was produced was evidently expecting English text, and auto-corrected free-standing i's (i = second person singular pronoun in Bambara, not capitalized ) to capital I's. It was very quick set of corrections, but did not include any review of the translation itself - that would be easier now with the text that follows.

K’i tɛgɛ ko ni safinɛ ye, o b’a to ko banakisɛ tɛ jɛnsɛn

Safinɛ ta: Ji gansan tɛ bɔgɔ ni tulu bɔ yen, olu minnu bɛ se ka banakisɛw dogo.
I tɛgɛ ko murumuru 15-20 nin cogo la.
A baara mumɛ ye murumuru 40-60.

1 I bolola nɛgɛw bɛɛ bɔ, o kɔ, i tɛgɛ ɲigin ni ji ye
2 Safinɛ k’i tɛgɛ la
3 I bolow jɔsi ɲɔgɔnna
4 Se i bolo ni tɛgɛ yɔrɔ bɛɛ ma
5 I bolo ni bolokɔni n’u tugunda bɛɛ saniya
6 Furancɛ min bɛ i bolokɔnikunba ni dɔ in cɛ saniya
7 I sɔninkɔrɔrɔla saniya n’a jɔsili ye I tɛgɛkɔnɔna na
8 A sɛnɛnko ka ɲa jiɲuman na
9 Fini walima sɛriwɛti jɛlen kɛ ka orobinɛ datugun, o kɔ, i tɛgɛ laja ni kura ye

Nin ja in labɛnna kalan de kɔsɔn ani fana tiɲɛ don a bɔtuma na. Dɔkɔtɔrɔw ka
ladilikanw bilanɔna tɛ.. Nin ɲininkaliw waloima hamiw minnu b’aw la nin
kan, a’ y’aw ka dɔkɔtɔrɔ sɛgɛrɛ.

© 2015 AEA International Holdings Pte. Ltd. All rights reserved.
Unauthorized copy or distribution prohibited.
Handwashing Steps Poster– v2 BAMBARA

Saturday, November 28, 2015

Dioula, Jula, Dyula, or Diula?

Hint: It's all the same language.

The New York Times recently featured an article on speakers of West African languages in the Bronx (a borough of New York City) that mentioned "Diula." From experience I realized that it probably is a less-often used spelling of the variety of Manding spoken mainly in Burkina Faso and northern Côte d'Ivoire, usually spelled Dioula (the French spelling), Jula (the usual spelling in the language itself as well as in English), or Dyula (also used in English and French). There are apparently other variant spellings such as Djula, Dyoula, Joula, and Juula (per various sources).

The number of spellings for this language therefore rivals the number for Temne, which I discussed in an earlier posting. However in the case of Jula/Dioula the existence of multiple spellings is partly related to French phonetics (French being used officially in almost all the areas where Jula is spoken, and the language of the first European descriptions of Julas and  Jula language). In French, the hard "j" sound in Manding (or English) requires using a "d," and the "u" sound corresponds to the French "ou," not the French "u."

So now it looks like all the permutations between "Jula" and "Dioula" have been covered. The "dy" digraph appears to have been introduced in the 1960s and was indeed preferred over "j" for Manding in the famous 1966 expert meeting in Bamako on unification of African alphabets. By the time of the 1978 expert meeting in Niamey, however, that was changing: Côte d'Ivoire retained "dy" while Burkina Faso (then Upper Volta) was using "j." I'm not sure that "dy" is still used in Julakan (the Jula language) in Côte d'Ivoire. Nevertheless, it still figures in the repertoire of alternate spellings used for the language name in English and French.

Unfortunately the multiple spellings probably lead to confusion, such as appears to be the case in the NY Times, which has in the past used Dioula, Dyula, Jula, and Diula. Also the use of French spellings in English can lead to mispronunciations by Anglophones - "dee-oo-la"  instead of "joo-la" (I've noticed this kind of thing on multiple occasions over the years).

And then there's Diola, Jola, or Joola

In any event, one should be clear that there is no relation between Jula/Dioula and the dialect cluster known as Jola, Diola, or Joola (pronounced more or less like "joe-la"). That is spoken in the Casamance, Senegal and neighboring areas of Gambia and Guinea-Bissau.

Friday, November 27, 2015

Breaking the "dialect barrier"*

You hear a lot about "language barrier" - a term that I for one would just as soon replace with "language divide" or "language gap" - but what about navigating the less daunting but sometimes trickier differences between dialects or varieties within a language or dialect continuum? Yet this must be common to the experience of many people around the world, including most of Africa.

Personally my biggest experience with this was transitioning in 1985 from Djenné, Mali, where as a Peace Corps volunteer I gained a certain competence in the Maasinankoore variety of Fulfulde, to Pita, Guinea, in the heart of the Pular-speaking Futa Jalon region. Actually it was unlike anything I've done before or since, even though I've encountered dialect differences more briefly elsewhere.

Making the transition was kind of fun - and occasionally funny. The biggest initial difference I found between Fula varieties has always been the greetings, but they are relatively simple to learn. Then there was keeping track of different verb endings, not making the initial consonant shifts in verbs, and a few other grammatical differences. Most of this was on the fly, but as I mentioned in an earlier posting, it did help me to have Sonja Fagerberg-Diallo's learning materials for both varieties of the language.

Then there were some differences in vocabulary - including some well-known (such as the first two below) and lesser known pitfalls:
  • caggal = "back (anatomy)" or "after" in Maasinankoore, but cagga = "prostitute" in Pular
  • ɓaawo = "back (anatomy)" or "after" in Pular, but ɓaawo = "female genitalia (very impolite)" in Maasinankoore
  • pati = "don't" in Maasinankoore, but pati = "grandmother" in Pular
  • kaani (pl. perfective of verb haanude) = "should" in Maasinankoore, but kaani = "ugly" in Pular
You can imagine that such differences could suddenly deflect a conversation into incomprehension or laughter.

Fortunately nothing I encountered in Futa Jalon was too extreme. As alluded to above, I spoke and listened, which is more or less how people from the region who move to a new area would bride a language or dialect gap. Except that I had a few advantages: One was that being a foreigner with any competence in the language was a bit of a novelty in Guinea at that time, so I think expectations were not high (and maybe some missteps were attributed to accent). I was also able to ask and get pointers from my Guinean co-workers in the community forestry project since they all also spoke French. And I was fortunate to have help of a local teacher, Amadou "Sernas" Bah, in going through an extensive Pular word list. (I had done something similar but more extensive in Djenné with Maasinankoore that led to the later Fulfulde Lexicon.)

The risk of meaning one thing but saying the opposite

I recall a Malian co-worker in Djenné, Boubey Touré, who came from Rharous in the Songhai area in the north of that country, but who never tried speaking his language (what Ethnologue classifies as Songhay, Koyraboro Senni) with speakers of the Djenné dialect of Songhai (a distinct variety of  Songhay, Koyra Chiini), for fear that some word he would say would would mean the opposite in local speech (as he put it). So he ended up mostly relying on Bambara in forestry extension work.

When I went to Futa Jalon, I actually encountered the kind of situation Boubey referred to, but in Fula. The accomplished (roughly past tense) negative ending for active voice verbs in Maasinankoore is pronounced -aayi (even though generally written -aali), wheras the future ending for active voice verbs in Pular is pronounced -ayi. Essentially the same sound. So coming from one background and not knowing, one could actually say a lot of things that mean almost the opposite in the other location.

The bigger lesson though is that an African extension worker like Boubey who is fluent in one language variety never had any training or materials to allow him to negotiate the dialect / language (variety) differences in his own country, while someone like me gets to go there as a Peace Corps volunteer, get all kinds of training for a second language, and then support and materials to transition to another variety of that language.

Greetings differ most markedly between dialects

I have much less experience with Manding varieties, but recall my first trip to the Bougouni area of Mali where people greeted me with "I ni se," a Bambara greeting I hadn't encountered before. It wasn't a different pronunciation of the widely used greeting, "I ni ce," and though it sounded disconcertingly close to the female response to greetings "Nse," it definitely wasn't that either. Then I realized this was the verb "se," to arrive. I ni se is a greeting on arrival. Welcome! Apparently this is common only in that region.

It is always the greetings that are the most different between varieties. One more example from Fula: In Maasinankoore one could say on parting, Alla hollu en ontuma (God show us another time), while in Pular I only ever heard a short form of that, En ontuma. Most other greetings are just different.

Thank you for your understanding

On a break from a PanAfrican Localisation workshop in Centurion (Tshwane), South Africa, Adel el Zaim, Dwayne Bailey and I took a walk over to a shopping area, and on the way out of one shop I looked back and read a sign "Dankie vir u ondersteuning." (What?) I asked Dwayne and he explained that it actually meant "Thank you for your support." English and Afrikaans are of course two separate though related languages, with some cognates, and evidently some faux amis as well. But this sort of example, even allowing for different spellings, of a concept ("under"+"standing") shifting meanings is something found between dialects as well.

Another example is from my move from Mali to Guinea: In Maasinankoore, the verb doomude (or doomugol) means simply to wait, but in Pular it has the sense of waiting around at mealtime to be invited to eat.

* Barrier or divide or gap?

Quick note on terminology. There is an advantage I think to using a "divide" or "gap" rather than "barrier" to describe the situation where lack of common language prevents or hampers communication. In effect what we're actually talking about is a "communication gap" in need of a common language "bridge" or learning some key differences between dialects to exchange ideas and information and do so accurately.

The gap/divide and bridge metaphor is I think quite powerful in portraying the problem and solution in the case of people with very different languages. In the case of dialects, the divide is not as wide or deep, and the problem is not as much inability to communicate, as running into mini-gaps such as differences in vocabulary or terms that are used differently. The knowledge need is more nuanced - figuring out what you know and don't know. So "dialect barrier" is no more apt here (I used it humorously in the title), but not sure of a better analogy.

Thursday, November 19, 2015

Two ebola posters in Krio, and revenge of Bambara Arial

On the theme of reviewing health education materials on ebola (and generally) per 2Ds&4Rs, here are some examples of why it is important.

First, two different posters in Krio, and quick analysis following. The first one comes from the International SOS collection (.pdf originals were converted to .png to display here):
Krio ebola poster from International SOS
The other Krio poster comes from the US Centers for Disease Control and Prevention (CDC) collection.

Krio ebola poster from CDC
Although I do not speak Krio, the language in the first poster looks to be closer to that used in a Krio announcement by the US Embassy in Freetown last year (see previous post on this blog). Can anyone familiar with the written form of the language tell if the second one is Krio or English with some Krio words? (Of course, Krio is an English-based creole, but it isn't English.)

Second, is a Bambara translation of instructions on hand-washing, again from the International SOS site.
Bambara poster on hand-washing from International SOS

It seems to have been a good translation, but produced with an old 8-bit font (apparently Bambara Arial), which when printed or converted to .pdf on a computer without that font, produced the odd and unreadable results you see. In the age of Unicode, this shouldn't happen.

At some stage such efforts in African languages need to be reviewed, just as materials in English or French would be. Not to find fault, but to correct and also to identify the types of problems to watch out for in future such work. This can be done in the context of an archive of African language health education materials, but another issue is how to build in review of such materials before they get disseminated.

Wednesday, November 18, 2015

Ebola education materials in African languages need an archive

We are approaching the second anniversary of the genesis of the ebola epidemic in West Africa in Meliandou, a village in the "Forest region" of southeastern Guinea, and hopefully also its end in that country. The the epidemic of course also ravaged neighboring Liberia and Sierra Leone, which have recently been declared ebola-free, and threatened to expand across the region before concerted local and international efforts turned the tide.

During the worst of the crisis I tried to highlight some of the ebola education efforts in African languages, with attention to some aspects that might affect their effectiveness to communicate messages. In that process I suggested that it is helpful to make health education materials in these languages widely available not only for their intended purpose, but also to facilitate improving them for future use, as summarized in "2Ds+4Rs: Why repost ebola info in African languages?"

At this time, with the peak of the epidemic thankfully behind us, it is important to find a way to archive ebola education materials in African languages. Other materials and lessons learned in English and French are certainly also important, but certainly won't lack for attention. It seems that there are three challenges for materials in African languages:
  • The languages are many, but materials in each relatively few
  • Materials are dispersed in different virtual and probably physical locations
  • Although some materials are accessible on internet, other materials on local or national levels are not, and in fact may in time be lost
So for instance the Ebola Communication Network and International SOS sites each have some posters in African languages (along with other languages) and a few in the same languages. That's fine, but why not also bring all the African language materials from those and other sites, to a site only on African languages, and group the materials by language? Such a site could be maintained by one or more of language organizations like the African Academy of Languages (ACALAN), health organizations like the WHO Regional Office for Africa, or academic organizations such as the African Language Materials Archive (ALMA).

In addition to the digital materials shared on websites - text as well as audio and video - I imagine there must be more materials in diverse languages on national levels. How to gather those is another question.

Sunday, November 15, 2015

Early Writing in Indigenous Languages

On the theme of the importance of stories, I thought I'd pass on a call for chapters for an academic publication on writing by children in their first languages. The scope is global, including Africa. It is announced by Dr. Arieh Sherris, who is currently a Fulbright Visiting Scholar at the University of Education, Winneba, in Ghana.

From the call:
The lion’s share of the world’s living languages face a bleak future. A growing consensus of linguists predicts that by the close of the 21st century 50-90% will disappear. Efforts to reverse this trend are underway worldwide. The purpose of this edited volume is to provide case studies of revitalization efforts at schooling early writing among children between approx. 3 and 12 years in lesser-known languages from Africa, Asia, Australia, Europe, North America, and South America.
For those interested, the full call is available on the Society for the Study of the Indigenous Languages of the Americas (SSILA) website.

Thursday, November 12, 2015

More on "community languages" in Africa

Following up on one of the topics of the previous post on this blog, here is some more recent information on use of the term "community language" in African contexts. That post discussed what I saw as a contrast between on the one hand a common use of the term in the West to describe a language of a minority community as differentiated from the dominant language of the majority population, and on the other hand, an African usage emphasizing a language spoken in a wider area, even as a lingua franca - implying "community" in a larger sense.

Regarding the latter, I came across an updated version of the 1985 UNESCO list of African community languages in the form of a map (image of the original PDF is on the right) and accompanying document, "Language of Instruction Policy and Practice in Africa," published in 2004(?) by the distinguished Nigerian linguist, Prof. Ayo Bamgbose. There are a number of differences from the original list, beginning with inclusion of South Africa, and listing of 212 languages (the previous version had 159).

Prof. Bamgbose also offers a shorter, and as I read it slightly different, definition for "community language" than that in the 1985 UNESCO survey:  
languages that are used for inter-ethnic communication
This formulation would seem to leave out widely spoken maternal languages that aren't also lingua francas (although any African lingua franca is also a mother tongue for some, and any first language that is widely spoken would likely also be used as second language by some). Still, this is different than the definition one finds in most references.

Of course, "community language" at its most basic just means a language spoken by some community. A quick follow up search found other uses of the term in discussing language in Africa that shade somewhat between what I have been contrasting as a Western focus on communities apart, and an African one on communities that link.

For example, a 2010 policy advocacy brief authored by Adama Ouane and Christine Glanz - "Why and how Africa should invest in African languages and multilingual education" (UNESCO & ADEA) - refers to "community language" once, after bringing up the importance of "learning to read and write in the language/s used at community level." And two recent edited publications feature papers discussing "community language" in various ways (the linked titles bring up passages using the term):
In the previous post I had suggested that it might be interesting to bring "community language" back into discussions of language in Africa. Evidently some people are already there...

Wednesday, November 11, 2015

"Community" & "national" languages in African contexts

Working on preparing the PanAfrican Localisation (PanAfriL10n) wiki for relaunch, I recently returned to one particular reference - a 1985 UNESCO document and its use of two terms which have meanings in Africa that differ somewhat from mainstream (Western) definitions: "community language" and "national language."

The document in question is: UNESCO Regional Office for Education in Africa, 1985, African community languages and their use in literacy and education: A regional survey. A wikified version of its Appendix 1 presenting 159 African community languages is included in the wiki. The publication covers quite a bit more than that and despite its age it is well worth the attention of anyone interested in language planning in Africa.

Community language

It was the UNESCO document's definition of "community language" - for which there is also a page on the wiki - that had me opening it up again for the first time in years, and then looking up other definitions for comparison. First of all, the term as used by UNESCO evidently came out of a Subregional Seminar on National Languages and Teacher Training held in Dar es Salaam in 1979, itself also funded at least in part by UNESCO. Community language is defined as:
the dominant and general means of communication in a district or province or similar large area ... which serve[s] the purposes of general communication over fairly wide areas within countries....
In this sense, a community language is a widely spoken first language and may also serve as a local lingua franca - quite common in many African societies. Contrast that with the definition from

a language spoken by members of a minority group or community within a majority language context
This usage apparently emerged in the mid-1970s in Australia,1 as an alternative way to describe languages other than English and Aboriginal languages, in the place of problematic terms such as immigrant language or minority language. It is also used now in Britain for non-English, non-Celtic languages,2 and apparently is used in similar ways in other countries in the West.

No problem with either of these takes (although inevitably there are critiques of any term used to describe any population or its attributes), but it is important I think to note the different frames of reference involved, and value of understanding these in their own contexts.

As far as I am aware, "community language" has not been used much over the last 3 decades in discussing the sociolingistics of multilingual Africa (apart from a brief discussion in a recent book by Tove Skutnabb-Kangas3; I'd be grateful for any other examples). So it might be time to bring it back into discussion.

A subcategory of community languages in the survey is "shared community languages" - what are now more commonly called "cross-border languages" (a term used internationally4). Once you look at languages across borders, the type of use in each country becomes a factor in characterizing overall usage. The graphic at the right is an example of what that looks like in the case of Hausa.

National language

The UNESCO document also offers a very clear definition for "national language," a term with multiple uses and that always seems to have a hazy meaning. In general, "national language" tends to be used differently in Africa than in the West, to mean "language of the nation" (of which there may be more than one, but are not the Europhone official languages), rather than "nationwide language." I've discussed this previously in critiques of Ethnolgue's use of the term, in 2013 and 2014, but without reference to the definition from Dar es Salaam, 1979:
a national language is defined strictly as either (a) an African language that is also an official language, or (b) a language that has been decreed to be a national language of a country. It must be noted that according to this definition:
- all mother tongues are not necessarily national languages
- French, English, Portuguese, and Spanish are not national languages, even though they may be official languages
In many African countries, one or a number of African languages are given the legal status of "national language" and in a few, all languages spoken indigenously on their territory are included in this category. In short, "national language" in Africa most often can be plural. Contrast with this definition, also from

the language spoken and written by the majority of people in a country; also, the official language of a country, recognized and adopted by its government
There are two issues with the latter definition - first, it is incomplete, and second the conflation of "national" and "official." The latter confusion may be common ("official language" being another term with more than one definition and context of application), however in Africa there is most often an explicit distinction with regard to languages between statuses of "official" and "national" that the Dar es Salaam definition reflects.

African frame of reference for discussing languages

The differences in how these two terms are used in Africa on the one hand and in the West on the other merit a brief reflection. "Community" in one case seems to be more a matter of joining, where in the other, it is more a matter of setting or being set apart. "National" in one case is more plural, while in the other it is more exclusive. That is not to make a value judgement, but to the extent these impressions are valid, it is a difference with deeper origins and implications.

Also, where there is a difference, the definition used in Africa is less well-known. This issue was already familiar from discussions about meanings of "national language," so I found it of note that there seems to be a similar dynamic in the case of "community language."

Community & national languages as part of a system

Part of what is interesting about the effort begun in Dar es Salaam in 1979 and continued with the UNESCO study in 1985 was that defining terms like community language and national language was not just an exercise in categorizing languages and distinguishing among their use, but more an effort to conceptualize a system of language use in multilingual Africa. This system involved a "five-tier categorization of African languages in relation to development as follows:
  • Mother tongue languages
  • Community languages
  • National language or languages (i.e., decreed as such)
  • Languages of African intercommunication
  • International languages"
This system, as I read it, is one of overlapping layers across which members of a multilingual society move, each with greater or lesser ease depending on education and linguistic portfolio. Outlining it in such a way calls attention to different levels of languages for purposes of planning and development.
At the same time, one could argue that what I have been calling the Western definitions of community and national languages also can be seen as part of a system. But that is a different system, wherein the national language is dominant, and each community language persists among a demographic minority in specific limited areas.

(See also: More on "community languages" in Africa)

The PanAfriL10n Wiki

Back on the topic of the PanAfriL10n wiki - I have taken longer than planned to get it to the point where it can be announced formally. The goal is presentability, and to that end some glitches in the transfer had to be corrected, and some features added. In terms of content, some pages needed to be eliminated or rewritten. Full updating will require input from others (another level of challenge), but some updates and additions are more important to the overall product, so they are being done now.

The longer term future of this wiki will depend on a number of factors, but if I'm investing time in it now, it is because there seems to be demand and there is not another resource that does what it attempts to do. More on all that when its return is formally announced.

1.  Michael Clyne. 1991. Community Languages: The Australian Experience. Cambridge University Press.
2. Colin Baker and Sylvia Prys Jones, eds. 1998. Encyclopedia of Bilingualism and Bilingual Education. Multilingual Matters. (p. 99)
3. Tove Skutnabb-Kangas. 2013. Linguistic Genocide in Education - or Worldwide Diversity and Human Rights? Routledge.
4. The French language Wikipedia has a lengthy article on this subject - Langue transfrontalière - but at time of access it had very little on Africa.