Reading Snippet - In Defence of History (4)

Formulating historical theories from sources

When discussing the considerations regarding sources, it's not the sources' nature that was the focus; the process of utilising sources to create theories was also under scrutiny.

A key distinction the author was keen to make was sources and evidence. While evidence arise from sources, sources do not automatically become evidence. Instead, when a hypothesis is formulated, it needs to be supported by evidence, and sources would then be trawled through to identify the relevant ones that prove or disprove the hypothesis. This gives rise to another question to sources - when do sources become sources rather than staying as snippets of information?

There are historians who argue that all snippets are historical sources waiting to be mentioned. However, another more discerning view is that snippets are abundant, but they do not become sources until they have been identified, discussed and promoted by historians. So snippets would reside in newspapers and archives with no significance or contribution, but once a historian has seen it, decided to consider it and taken efforts in writing it out, then it has formally entered the historical studies arena for future candidacy as evidence or subject of debates.

In the process of using evidence to generate a new historical theory, how true is the post modernist argument of "there is no truth" and "there are multitude styles of creating varying theories from the same evidence, and they can all be valid"? If these arguments are true, then practically all theories are as good as historical literature that carry limited value, and efforts to extract "the right conclusion" are futile.

The authors' counter-argument is the "puzzle theory", in which when a variety sources are used to construct the web of evidence, a picture would emerge which has a limited set of interpretations. The most extreme example would be the Holocaust, which in no way could be argued to be an exaggeration or that it has never existed (as a prolonged massacre). It is possible to extract different theories regarding the rise, the evolution and the driving forces behind the Holocaust based on a similar set of evidence, but they would broadly lie in the same band.

This argument has forced the post-modernist to retreat and claim the "there is no truth" lies in the final interpretation by the readers of the theories presented. When sources are considered and web of evidence has been constructed, the historians are in the process of piecing together valid historical theories and it is OK to be bound by the web of evidence into a band of possible theories. However, as soon as the process of penning down the theories has completed, the process of creating historical studies outcome ceases and the resulting literature becomes open to readers' own interpretation, and in this sense the truth can never be perfectly & comprehensively presented to the readers, and is therefore a literature to the readers in this sense.

The combination of this revised post-modernist view of "historical truth and readers' interpretation" and web of evidence raises a critical question about any arising theory - how can we ensure that they are valid & supported by evidence and could be reasonably challenged? After all, the risk is that the web of evidence is poorly constructed and counter-evidence are omitted by mistake or intentionally, or that the band of possible theories is wrongly attributed and that inherently erroneous theories slipped through the web.

A conventional answer to this is that the academic circle develops and evolves a consensus and act as the gatekeeper against wrong theories or poorly argued theses. But these is not infallible, as demonstrated by the German academia during the Nazi period whose gatekeeping allowed pro-Nazi theories to pass. The scope for collective human error was great.

From the "Abraham case", the author evaluated such a risk from three angles. The first is the challenge from an opposite "meta camp", as a band of possible theories still has 2 or more extremes, and theories raised towards one end would attract criticisms and attempted discreditation from the other ends, but the to-and-fro could help flag up flaws in arguments or evidence gaps, and help to deepen & widen the inspection into the validity of theories. Gatekeeping does not come from consensual view, but opposing views putting a sceptical spin on each others' studies.

Another angle is the strength of the web of evidence - it is inevitable for evidence to carry mistakes & gaps arising from typos, translation differences or missed counter-evidence. It is possible for these mistakes & gaps not to affect the overall validity of the theory, but the more mistakes and gaps found through inspection, the more suspicious the theory becomes. Even if the theory is ultimately valid, the paper should be deemed weak and be rejected.

The third angle is identification of patterns of mistakes at a macro level - analysis logics, data collection methodologies and calculations & attribution. If a researcher deliberately missed out a vast repository of evidence known to be unfavourable to his/her theory, or assumptions behind calculations are too simplistic or plain wrong, or that exceptions are not considered at all, then the theories should again be deemed weak and be rejected.

The process of turning sources into evidence then to validated theories is complex, and further complicated by the risk of mis-interpretation by the reader. However, what becomes clear is the need for rigour at every stage of the process, and subjecting the process and outcomes to hostile inspection at micro and macro levels, instead of having a simple "consensus test". The arguments and challenges should not be feared - they help to force a deep re-think on the linkage between sources, evidence and the theory, and offered an opportunity for theories to be refined into a more robust state. In the same vein, the post-modernist interpretation of history was challenged, and refined into something far more defendable, acceptable and thought-provoking to the overall research community.