The line between the nature and learning is often blurred. Simply said, if someone's genetic code contains an instruction "when X, do Y", should we say that the behavior Y was caused by the code (is innate) or by the presence of X (is learned)?

The story implies than in conditions X1 baboons behave violently, and in conditions X2 baboons behave peacefully. The word "conditions" here includes both their environment and their history, and it seems that X1 = "lack of food OR recent history of violence" and X2 = "enough food AND recent history of peace". When there is enough food, both X1 and X2 seem self-perpetuating, though the experience with X2 is very short yet.

My prediction is that adding violent baboons to the group would create an X1 situation. (A less certain prediction is that even without external baboons, sooner or later the group will generate a violent individual. Problem is, due to the "sooner or later" part, the second prediction is unfalsifiable.)