Most people in the West think that in the realm of history, there is no such thing as certainty, but only (varying degrees of) plausibility, and that it is, therefore, not satisfactory to use the category of “proof” for historical matters.
Whether this is right or not, is important in general terms in the search for historical truth; and it is important also in terms of the assessment of the foundations of faith, because Christian faith (or biblical faith) is, in important ways, founded upon historical events, especially (but of course not exclusively) the resurrection of Jesus (see, e.g., 1 Cor. 15:14-17).
In order to answer the question raised here, we need to consider the following eight points:
1. Historical science is different from other sciences, especially from most natural sciences, both with respect to its object as well as with respect to its methods and ways of proving, and its ways of collecting data and presenting results.
It is, therefore, inadequate to apply categories used in natural science for historical investigations. While something may be “proven” in a chemical laboratory by repeating an experiment and observing that the results are the same as in a previous round, this cannot be done in the realm of history. It is also inadequate to expect “proof” for events in ancient history of the sort that is available in the present through photography and other modern technological means. No one in his right mind will doubt the existence of Napoleon or the special event of his death on the isle of St. Helena just because there are no photographic pictures available documenting specific events of his life or death.
2. When we talk about events (or states), we need to distinguish between various aspects of them.
One of the most fundamental – and obvious – distinctions, as far as events are concerned, is between the crude “what?” of an event, as opposed to questions related to its “how?”, or questions related to its “why?"
3. The category of “fact” does exist in contemporary historical science
(see, e.g., Furay / Salevouris, The Methods and Skills of History, pp. 12, 16, 28, 146, 245, 248). Knowledge about “facts”, however, includes the notion of “certainty” as a necessary corollary. Nevertheless, it also needs to be acknowledged that the degrees of certainty with respect to certain events (or states) – or aspects of these events (or states) – vary. There is, thus, a continuum between “plausibility” (or “probability”) and “certainty”. In many cases, it will be possible to establish the crude “what?” of an event (or state) with certainty, while questions related to its “how?”, and even more to its “why?”, may be answered only with varying degrees of plausibility.
4. Questions about historical events (or states) need to be investigated and answered based on evidence.
Evidence is normally a “witness” of some kind or other, like an eye-witness report, a piece of ceramics, a coin, etc.
5. The degree of certainty will, among other things, depend on the number of available witnesses.
In historical science, a single witness will be taken as a decisive piece of evidence only with great caution. The more the (author of a) single source has proven to be credible, the more one will need to take it seriously. Even if there is only a single witness, it cannot be discarded outrightly; that is, there is no general rule that will allow us to assume that the event (or state) reported did not take place.
The more independent witnesses that are available, the higher the degree of certainty will normally be afforded to a specific event (or state).
6. Even if there are more than one witness, the problems connected with the search for historical truth are not solved automatically, particularly (but not only) when the witnesses contradict each other.
There is always the possibility that one or more witness may err, or even lie – or, in fact, all of them may err or lie. Therefore, one has to find out which of the witnesses (if any) has spoken correctly.
7. Because of the fact that single or multiple witnesses may err or lie, it is necessary to cross-examine witnesses, in analogy to how witnesses are cross-examines in a court-room setting.
The reliability of each witness has to be assessed (among other things) by taking into consideration possible defects, biases and tendencies of an author, or by investigating the value of other statements issued by the same author with respect to other events.
In principle, this procedure is also used in connection with biblical texts, to the degree that they are studied in the context of a strictly historical investigation.
8. If the net of independent evidence for an event (or state) and its context is tight enough, a point of absolute certainty will be reached, where the development of an alternative scenario becomes absolutely impossible.
For example: the murder of Caesar, or the murder of MLK. But even if the “what?” of such events will be absolutely certain, things still often look different with respect to (aspects of) the “how?” and, often even more, the “why?”. New evidence may appear that elevates previously uncertain elements of an event to higher levels of probability, perhaps even to absolute certainty in some cases. But in many cases this stage is not reached, because the evidence is limited.
The result: Is there “proof” in history?
According to what we have pondered so far, we may conclude: Something is in fact proven in history if there are several independent, reliable witnesses that concur in their witness, and if the event (or state) can be understood as a meaningful element in a comprehensive course of events (or sequence of states). More or better proof, or a different kind of proof, is not available for every event (or state), at least in all eras of history before the introduction of live-recording technologies.