A publication of the Archaeological Institute of America

Special Introductory Offer!
Africa's Earliest Bananas Volume 59 Number 5, September/October 2006
by Peter Robertshaw

A new discovery peels back the history of our favorite fruit.

[image] Banana propagation requires intensive human intervention; the discovery of banana phytoliths in Uganda may push back the arrival of agriculture in equatorial Africa to at least 5,000 years ago. (Alexander Joe/AFP/Getty Images) [LARGER IMAGE]

Many years ago, my wife and I bought our daughter a stuffed toy gorilla that was clutching a banana in its hand. Now, almost anyone would assume that gorillas have been eating bananas for as long as there have been gorillas. Like most archaeologists who work in Africa, however, I doubted that bananas were an important part of a gorilla's diet because I was taught that bananas had been introduced to Africa from Southeast Asia only within the last 2,000 years. As a result of a recent project in Uganda, however, now I am not so sure. When did the banana cross the Indian Ocean to Africa? As you might have predicted, the answer is slippery.

I have spent many years investigating the rise and fall of Bunyoro, a precolonial kingdom that ruled in the wet, tropical region of western Uganda during the last millennium. While I was doing my work in one part of Uganda, David Taylor, now professor of geography at Trinity College, Dublin, was drilling cores in highland swamps in the Rukiga Highlands, a corner of Uganda famous, as it happens, for its gorillas. David was identifying pollen in his cores to reconstruct climate and vegetation history. Eventually we met and ended up writing several papers together--though not in the pub--exploring the relationships between climate change, political economic history, and vegetation shifts in Uganda. Then we persuaded the National Geographic Society to give us a grant to collect sediment cores from papyrus swamps adjacent to the archaeological sites I had been investigating. And so the summer of 2001 found David and me, as well as Julius Lejju, David's Ugandan doctoral student (now teaching at Mbarara University), standing at the edge of a small, papyrus-choked swamp at the site of Munsa, where deep ditches a couple of miles in length were originally dug about 500 years ago.

Direct evidence of ancient agriculture usually takes the form of charred seeds and other plant remains. But that is no use when it comes to bananas because nobody ever broke a tooth on a seed in a banana milkshake. This is where phytoliths enter the picture. Phytoliths (literally "plant-stones") are microscopic silica bodies found in the stems and leaves of plants. Whereas the plant dies and decays, phytoliths are virtually indestructible and so become incorporated into sediment cores. Happily, different families, and sometimes even species, of plants produce phytoliths of distinctly different shapes.

When Julius told David that he had found some banana phytoliths in our cores, we were pleased but not surprised since bananas are grown at Munsa today. The earliest documentary reference to bananas in Africa is from the sixth century A.D. Historians and archaeologists often thought that bananas were probably introduced to Africa via Madagascar, which was colonized by people from Southeast Asia in the first millennium A.D. This idea, however, was thrown into doubt some years ago on the basis of linguistic, botanical, and geographical evidence, all of which suggested that bananas might well have been introduced much earlier. There was no way to prove this, though, a few years ago a team of Belgian scientists had caused a commotion of their own by reporting that they had discovered banana phytoliths dated to 500 B.C. in Cameroon, thereby pushing back the date for the first appearance of bananas in Africa by about a thousand years.

Then Julius unleashed a bombshell: He had identified several banana phytoliths toward the bottom of one of the cores, in sediments that accumulated more than 5,000 years ago! Now here we were with bananas that were well over 2,500 years older still. Were they really banana phytoliths? And if they were that old, how could they have reached the middle of Africa so long ago? All of a sudden, we were not only challenging the assumption that bananas only reached Africa in the last 2,000 years, but that African connections to the Indian Ocean world may be more ancient than we previously supposed.

Peter Robertshaw is professor of anthropology at California State University, San Bernardino.

© 2006 by the Archaeological Institute of America