The first fundamental problem with the problem you are positing here is epistemological, meaning concerning the theory of knowledge. And it is: What is a historical fact?

The problem with facts is that facts do not assert themselves, they need to be asserted and that process often involves a process of interpretation.

Take f.ex. the end of the Western Roman Empire. Most people would assert that it is a fact that the Western Roman Empire ended in 476 C.E.. But what happened in 476 B.C.? On September 4, 476, Odoacer forced then 16-year-old Emperor Romulus Augustulus to abdicate and installs himself as ruler over Italy. Odoacer is seen as the "first Barbarian Emperor" and that is said to end the Ancient Roman Empire. But what about Romulus Augustulus? He was the son of a man who although had worked his way through the ranks of the Western Roman military but started his career as a diplomat for Attila the Hun. Similarly, the Eastern Roman Empire did continue to exist for a considerable amount of time. And yet, we use Odoacer – despite the fact that he too claimed the title of Emperor – as the end of the ancient era and Rome and the beginning of the Middle Ages. This, in effect, is something established by British historian Edward Gibbon in his The History of the Decline and Fall of the Roman Empire and has become accepted "fact" by now.

Another example of this comes in a way simpler form: The end of WWII in Europe is in Western Europe commonly dated with May 8, the date when Germany signed the unconditional surrender to the Allies. But that's not the case everywhere in Europe. The Soviet Union and its successor states celebrate the day of victory as the end of WWII in Europe as May 9 because the unconditional surrender went in effect at exactly midnight between May 8 and May 9, 1945. So, what is the factual date when Germany unconditionally surrendered to Allied Forces?

What I am trying to get at is that what we consider historical fact is based upon the interpretations. Interpretations both by the people who recorded an event as well as interpretation by historians and others who take these sources and put them into context. The writing of history is never based upon finding just one fact, it is the process of constructing meaningful narratives by putting past events into a sequence – with a beginning, a middle, and an end – and context and making sense of them. At its base, this can take the form of casual sense like saying because the Battle of Waterloo was lost, Napoleon could be exiled. But this reaches further in the sense of what the meaning of these events for later history up until this day has.

We can be certain that Napoleon lost the Battle of Waterloo because a myriad of sources written by contemporaries of the event told us so and we can be certain that napoleon was exiled because contemporary sources – including Napoleon himself – told us so. Because we have this evidence, we can treat this as an established fact, though what defeat and exile means in this context is – once again – something that is an interpretation of the events that occurred, both by us and the contemporaries writing about them.

Since we can't go back and observe the events ourselves we have to rely on certain established standards about historical evidence to establish the facts from which we construct our narrative. And even if we could back, who's to say that we come to the same conclusion than the contemporaries for we are of a different time and different times have different ways of giving meaning to events. Slavery is something that has a distinctly different meaning for us than for those people who practiced it.

What we will rely on is therefore the meaning and interpretations as recorded in the historical sources recorded by the contemporaries. Because those don't change. But what the mean for us, how they are related, causal and otherwise, will often change over time. Gibbon's version of the fall of the Roman Empire is now regarded as outdated by historians, simply because Gibbon did not have access to all sources we know of today and because Gibbon is interpreting them in a manner that is the result of a specific mindset of his specific time, which differs greatly from ours.

And so, what we can conclude is that the only thing that does not change are the sources themselves. But what they mean and often what is considered fact itself changes over time and in different historical socio-political settings. The fall of Rome in 476 B.C. Gibbon did consider fact and we do not anymore. That WWII ended on May 8 is a fact for half of Europe and for the other half that it ended on May 9 is fact. And both are, in a sense, correct. But this is the thing about history and really engaging with it: It's not a series of dates and events, it's a process of putting order to events as recorded by historical sources and interpreting them. And in some cases, multiple interpretations can be correct at the same time for what we consider meaningful and hos we construct meaning can change over time and in different contexts.