Indeed a single piece of oral evidence may in a particular context be more reliable than a single piece of written evidence as just one instance. @Blindbohemian already provides such an example. The only thing I could add onto the discussion at this point is that many written records are based directly off oral records (i.e. historians roaming around asking people about a certain event)."Oral traditions can be reliable; written records can be unreliable." is saying oral can be more authentic than written
The most important thing to consider is not the mode in which the evidence was produced or collected, but instead to assess other properties of the evidence. Is the evidence corroborated elsewhere (bonus points if corroborated by other sources which are unlikely to have had contact with the original producer of evidence)? Is there any reason to suspect bias if the producer of the evidence is a person (do they have an agenda)? And many, many more questions we have to ask about every piece of evidence.
That is why I say we cannot just assume a recount of history is incorrect simply because that civilisation did not write down their history.
- 4