For some things you have no options but to take what information is available. If a community doesn't have a literary tradition then oral histories is all that can be managed. The option is to simply let such things die - a sad fate.
But you need to be careful. First person interviews are very valuable, but are obviously colored by the life history of the one interviewed. This is generally recognized.
Two works that I've found valuable come to mind. The first is The Ten Grandmothers by Alice Marriott. A group of scholars interviewed members of the Kiowa people of Kansas. Some of those interviewed had been children before the destruction of the buffalo culture of the plains. There was little else in the way of information available.
The other is Nisa: the Life and Words of a !Kung Woman by Marjorie Schostak. The title character was a San woman from the Kalahari as their lifestyle started to be overcome by the modern world. Again, the San have no written tradition, so oral interviews is all that was open.
There are many other examples, of course, including the work of Margaret Mead.
But all such works need to be honest about sources. The data is influenced by the lives and views of those interviewed. But it is also necessary to assure that assumptions made by the interviewers don't color the story. Missionary observers of the Hopi People, for example, made many very serious mistakes of interpretation and caused a lot of damage.