One of the two foods highlighted in my chapter about soybeans in Africa in The World of Soy (published already 8 years ago now) was a pungent West African condiment, traditionally made with “African locust beans” but increasingly with soybeans, and known by various names such as: sumbala, sunbala, or sungala (in Manding); daddawa or dawadawa (in Hausa); iru (in Yoruba); netetu (in Wolof); and ojji (in Fula). The other food was tofu, which is also an interesting development, but here I’ll focus only on the condiment, and the ongoing natural resource issues behind a shift in its main ingredient.
Increasing use of soybeans is a good story, but the decrease in use of locust beans is not, reflecting declining numbers of African locust bean trees due to loss, low regeneration, and little planting, and decreasing production by existing trees where rainfall levels have gone down.
What is sumbala…?
Sumbala – I’ll use the term from Manding here, although I used the Hausa term in the book chapter – is widely used across West Africa to season sauces. The process of making sumbala centers on fermentation of seeds, most commonly those of the African locust bean tree, Parkia biglobosa (among the traditional alternatives is a close relative of the more humid tropics, P. bicolor). More about the tree and what’s happening with it in a moment, but about its seeds – evidently they are quite nutritious. Locust beans have high protein (like distantly related beans) and various other nutrients. The fermentation process, aside from changing taste, also affects the profile of available nutrients.
Production of sumbala – generally done by women as a home-based industry – is a multi-step process from preparing the seeds to boiling them to setting them aside in the proper conditions to ferment (without innoculent to get the fermentation started). The bacteria of the fermentation are in the Bacillus group, primarily Bacillis subtilis (a variety of the latter is also used to ferment soybeans in a somewhat different process to make a Japanese food called nattō 納豆).
The final sumbala product can be seen in local markets as a ball or patty. These may be traded at some distance within the region. Nowadays it is also packaged in a crumbled form for sale in urban markets and for export.
African locust bean tree, or nere
Parkia biglobosa, often called in English African locust bean tree, or more rarely African carob, is common in Africa, and especially West Africa, roughly between 5°N and 15° N latitude – from the edge of tropical forests to the edge of the desert (“mean annual rainfall” of 400-700mm). The tree of course has names in the various languages of the region such as: nɛrɛ or nɛtɛ (Manding); ɗorawa (Hausa); irugba (Yoruba); uul (Wolof); and nareewi or netehi (Fula). “Néré” in French and English comes from the Manding, as evidently do the Fula terms.
The tree is valuable for its seeds, of course, and also for other products, such as the nutritous yellow pulp of the pods. 
The Lost Crops of Africa (vol. 2) chapter on the African locust bean tree discuses some of those uses. In terms of traditional medicine, bark and leaves are apparently effective in treatment of some infectious diseases.
Due to its value, the locust bean tree is usually conserved in fields when most other woody plants would be cleared. However it has been noted for a while that natural regeneration is apparently not sufficient to replenish the population – which recent studies from Burkina Faso and Nigeria confirm. Reasons given for this include the harvest of the seeds for sumbala, and loss of seedlings to drought or browsing livestock.
Efforts to plant African locust bean trees are apparently few, even though it is simple to seed directly or produce in nurseries.* One example of planting in southern Burkina Faso is interesting but seems to be limited in scope. It is not clear whether planting this species is part of any large scale project or extension effort, although it was mentioned in a 2010 recommendation of species for the Great Green Wall project.
Another challenge is lower rainfall levels, which aside from affecting survival of seedlings also lower production of seed pods – as observed in another example from Burkina Faso.
Soybeans to the rescue?
It is in this environment, with generally declining availability of locust beans but increasing markets for sumbala, that soybeans emerged as an alternative. This substitution is described in more detail in the book chapter, but it appears to have been an innovation by Nigerian women in the 1980s. I am not aware of any studies of its dissemination – by local networks or projects, nor whether it may have been an innovation in several locations experiencing the same shortage of locust beans. I first learned of this substitution in 1999 while in Mali, and it was treated as a new development – indeed a project had introduced the use of soybeans for sumbala a couple of years earlier.
Soybeans have a couple of advantages for making sumbala, notably as a field crop they can be produced quickly enough to respond to demand, and having thinner seed coats, require less boiling time (hence less firewood in typical production) to prepare for fermentation. On the other hand, soybeans as a field crop require more labor to produce than locust beans, which are simply harvested from trees, and the soy sumbala deteriorates faster in storage.
In terms of taste, opinions I heard were that soy sumbala was comparable to that made with locust beans. However Margaret Shao, in a master’s thesis referencing her research in northern Ghana, found that locust bean sumbala was preferred. And further that because of this, soybeans were sometimes used as filler with locust beans in making sumbala to combine the advantages of both. Beyond that, I have not seen any studies of preferences.
Looking to the future
The Africa locust bean tree is valuable in many ways, of which the use of its seeds for making of sumbala is particularly important. Soybeans are established as a valuable crop and food in West Africa, and have added an important option to producers of sumbala in the face of increased demand for their product and decreased supply of locust beans. The two – soybeans and locust beans – are in a substitution relationship that is relatively novel. But the emergence of soybeans and decline of locust beans does not mean that the former should be expected to supplant the latter as the main ingredient of this West African condiment.
Use of soybeans to produce sumbala, by helping meet demand for the condiment, may actually help reduce pressure on supply of locust beans, thus perhaps indirectly favoring regeneration of the African locust bean tree. However, various factors make reliance on natural regeneration of the African locust bean tree unrealistic. The main hope for maintaining this tree as a component of African dry forests and sustainable source of seeds and other products will have to be its deliberate production and planting both as a tree crop in reforestation projects and as a “food tree” crop in community-level agroforestry.
In memoriam
I would like to add a brief remembrance of Prof. Sidney Mintz, one of the editors of The World of Soy, and of course a distinguished scholar whose work profoundly influenced his field of anthropology, as well as the lives of many. Prof. Mintz passed away last December after an unfortunate accident. The New York Times obituary called him the “father of food anthropology” and indeed his work on the project that resulted in The World of Soy was in this vein. According to his website, his next project was “looking at fermentation, a too little-noticed subject, when we consider that as much as one third of the food we eat is fermented.” He had a tireless intellect and genuinely inspiring approach to scholarship and to life.
My only connections with Prof. Mintz were as an undergraduate student years ago in an intro to anthropology course he taught at Johns Hopkins entitled “Human: Being and Becoming,” and of course in the soy project, including a panel presentation at the 8th Symposium on Chinese Dietary Culture in Chengdu. There are many more fitting tributes to his life and work. I can only say I was privileged to have learned from him.
* The thick seed coat may require some treatment – “scarification” – to facilitate penetration of water and germination. Manuals recommend several methods including boiling briefly, soaking in sulfuric acid, or physically nicking the seed cover (one study, for example, compares methods). For a moderate production, on the order of 100 as I recall, I used the following simple method in Djenné, Mali during the early 1980s: Soak seeds in water overnight, remove swollen seeds for planting (we produced them in plastic pots in the nursery for later outplanting), return non-swollen seeds to fresh water to soak, with option to nick the outside of the seeds with a pocket-knife to hasten the process.
