Oral tradition can be both valid and valuable, but it requires a different sort of source criticism than is required for written sources. Historians are trained to evaluate and use written sources, so they tend, naturally, to trust conclusions based on them. And frankly, it is easier and more reliable to draw on written resources because the source criticism is more easily applied. After source criticism is applied to oral sources, one is left with a lot less certainty.

Anthropologists and folklorists are trained to apply source criticism to oral tradition, but the conclusions they reach about the past are likely to be more vague than the ones historians reach, simply because of the nature of the resource. That does not mean oral tradition is less valuable. It can provide amazing insight into the past, and it can be quit reliable. One simply needs to approach it differently.

A recent study of Australian indigenous oral tradition is provocative and suggests how impressively-accurate material from oral sources can be. The following is a synopsis I wrote up on this for a project I am working on. Perhaps it will help:

Nicholas Reid, Patrick Nunn, and M. Sharpe, “Indigenous Australian stories and sea-level change,” paper presented at the 18th Conference on the Foundation for Endangered Languages: Indigenous Languages: Value to the Community, Okinawa, Japan, 17-20 September 2014; and see the report, John Upton, “Tales of Ancient Sea Rise told for 10,000 years,” published January 25, 2015 in Climate Central. “That’s the conclusion of linguists and a geographer, who have together identified 18 Aboriginal stories — many of which were transcribed by early settlers before the tribes that told them succumbed to murderous and disease-spreading immigrants from afar — that they say accurately described geographical features that predated the last post-ice age rising of the seas.” The observation was that some of the stories recorded landforms that had been inundated up to 12,000 year before present, but most 10,000 BP or slightly more recently. The study suggests that stories retained accurate information about the Ice-age geography for 400 generations. They suggest that similar stories among the Klamath Native Americans in Oregon retain information on the eruption of Mount Mazama, 7,700 years ago.

This is a huge problem for all historians, and especially when trying to decipher human history from before the written word, ie pre-history. And that of course even makes the assumption that written history is by default more credible, which is far from true (and much written history is just adaptations of oral history).

The field of history is in itself always problematic for these reasons, we can never know a definitive truth which is why we have to use the hermeutic method of analysis, and in the hermeneutic method oral history is as valid as any - but it still has to go through the process.

This is a really interesting question and has the potential to delve deep into the darkest corners of human understanding of history. My understanding of the interaction of oral and written histories is mainly based on the ancient Greek tradition, and specifically Homer and the Greek/Athenian play writes. I'm going to own up now and state that this is not a yes/no question and there are a huge number of differing academic opinions on just this topic.

I think something that has been missed a tiny bit in the other comments is that oral and written histories are not mutually exclusive, neither are they necessarily completely equatable. Let me explain in the context of Homer's Illiad and Odyssey. A key figure in such an investigation is Milman Parry. I hope this example helps you in your question, but I know that it may not be fully comparable.

Ok, so to begin to understand the intersection of linguistics, oral and written history we need to peel back the layers one at a time.

Contextually: it is widely understood that the Illiad and the Odyessy were originally oral stories, more specifically poems that were spoken aloud to audiences and perhaps even accompanied with music (although this last assertion is very speculative). Homer speaks about heros and their deeds an addresses his audience as his contemporaries several times in formulaic language. In fact the name 'Homer' may not refer to one person but in fact be derived from 'Homeridae' - professional singers. This is very important for understanding oral history because it means that the epic stories of Homer were not the preserve of one person, but were remembered and constantly shared between more people with even greater audiences. With this understanding we can look at Homer that we have now, frozen in a final state in it's written form, with the understanding that it evolved from a linguistic, oral tradition.

An excellent paper on this by Barbara Graziosi and Johannes Haubold (Graziosi, Barbara & Haubold, Johannes (2005). Homer: The Resonance of Epic. London: Duckworth.) puts it like this:

"Repeated words and phrases such as 'swift-footed Achilles' trigger a chain of associations which, we suggest, work like acoustic resonance. They suggest associations in the mind of the audiences and readers that are crucial to the story, yet do not appear to be consciously invented on the spot. And even if they are, their main function is not to capture the moment that is being described in a unique way, but to tie that moment to the larger tradition and thus endow it with resonance."<

So, how does this help us? Well, it shows that oral traditions are tied into a wider scope of cultural understanding and a dialogue between the speaker and the audience. The problem we have here is that my example is about a story. While there may be some historical basis for the events in the Trojan War, historians would be VERY hesitant in treating the text as a historical document in the same way that we would for a source such as, for example, Herodotus (although he isn't without a whole host of other problems to do with historical validity and oral history, but that's for another day).

So, we need to know how valid an oral historical tradition is in direct comparison to a written one. Essentially is a written record more likely to contain accurate historical 'truth' compared to a written one, which is far less tangible and verifiable, for the simple fact that it is contained within a cultural tradition and shared memory, compared to a book.

This is where Milman Parry and his research comes in. Parry is really the foremost source on the oral tradition relating to Greek epics and is well worth researching if I haven't put you off the subject for life. To boil a life-time of work down to a few sentences; Parry worked on the theory that South Slavic epics and Greek ancient epics had a demonstrable similarity that means that it is possible for HUGE (in the epic sense) amounts of information, contained within a poetic, performance i.e. oral format could be memorised and recited accurately to an expert audience. The audience are as much a part of preserving the tradition as the speakers. The oral history belongs as part of a collective memory as either speaker or listener, part of the life of everyday people. This does not mean that the poems were not subject to change, just that, according to Parry's research, their were likely certain aspects that HAD to stay the same or the epics would not have been remembered as easily and the audience would have noticed the difference. By writing these stories down, two primary things occur. 1) You freeze the story into that moment in time, a written record provides tangible evidence that at the time of writing, that was how the information was understood. 2) You start negating the need for an oral tradition at all. If people can read it for themselves, or perform to an audience without having to go through the lengthy process of memorizing the whole thing. The story/history starts to take on a more solid form. This is not to say that written histories are infallible, just that they are more verifiable, accessible (and easier to translate!) than an oral equivalent.

So, ultimately, the difference between an oral and written tradition is far from clear. I think, based on the work of Parry, that for a historian to dismiss the evidence of an oral tradition out of hand, simple because of it's format would be a gross oversight. Depending on the era of history under investigation ANY information is important to gain a full understanding. What is also important to appreciate is that just because it is written or spoken does not make an history more right or more wrong than the other. Oral traditions are subject to change within the collective or individual memory, just as written texts can be edited, lost or copied out inaccurately. Assuming that both contained accurate information in the first place! From Parry's work and the evidence from Homer, it seems entirely plausible that, as you say, entire tribal genealogies were put to memory. Homer, for instance, names every single soldier who dies in the course of the Illiad. In a very famous passage in Book II, Homer recites the so-called 'Catalogue of Ships'. If Parry is right, this would all have been committed to memory, along with the rest of the epic, constantly being verified, at every performance, by an 'expert audience.' Now, I'm not entirely sure if that relates to the traditions that you are speaking of, but it certainly seems feasible that the oral and written traditions could be equally as useful when investigating the history.

In short, every source has a place in proper historical investigation. Oral traditions, although different, could have just as much significance as written equivalents or counterparts.

Milman Parry's collection at Harvard University can be researched online here - Centre for Hellenic Studies