Evidence myths and the current state of knowledge

As someone who has spent hundreds of hours bending over data trying to find evidence that could possibly support a theoretical point, I think I have some authority to say that I do not believe in research that is not based on solid and sound evidence of some kind. However, it never ceases to astonish me how some people believe that dependable research has to provide the ultimate proof and absolutely incontestable evidence. Perhaps we should take a look a little bit more carefully at these assumptions.

First, there is the issue of proof. When I was at school, we were taught that our solar system contained nine planets and we should know them because astronomers had proof that that was so. Now every kid knows that actually astronomers have proof that our solar system is formed by eight planets since Pluto was downgraded to the category of a dwarf planet. This astrological episode illustrates quite well that evidence is not something that inherently spells ‘the truth’. Evidence, of any kind, has to be analyzed, categorized and interpreted and, therefore, its existence does not necessarily guarantee that the ultimate truth has been discovered. Analysis and interpretations will always be dependent on the current state of collective knowledge in any discipline and on the knowledge, insight and imagination of those individuals analyzing the data. Scientists up to the beginning of the 21st century did the best they could with the telescopes, computers and satellites they had at the time. It is not that their research lacked scientific rigour; it was simply limited by the means they had at their disposal and by their then current state of knowledge.

This reminds me of a short article by Lisa Jardine (2006, online) where she discusses people’s perceptions that scientific evidence is always unequivocal and that experiments ‘ought to prove’ scientific theory ‘once and for all’.

How odd it is that non-scientists think of science as being about certainties and absolute truth. Whereas scientists are actually quite tentative – they simply try to arrive at the best fit between the experimental findings so far and a general principle. Because most of us want more certainty, believing the experiments ought to prove the scientific theory once and for all. We cannot afford ourselves the luxury of waiting for evidence which clinches the theory. We are going to have to learn how to participate in debates which are not about certainties.

The reliability of evidence is also limited by the means we have to transmit and share it. In the case of textual data, of any kind, there is the mediation of language to be considered. The issues involving signifier and signified and the social construction of meaning do not need to be discussed here now, but if anyone believes that it is possible to assert that just because something is said or written this constitutes evidence of the ‘truth’ behind an event, I would seriously recommend some extra reading and thinking. When we consider some spoken or written text, the only ultimate proof that we can possibly have is that these things were said and written; all the rest has to be contextualized and interpreted.

Yet, how trustworthy our evidence is also depends on how much we are prepared to believe in the data our research participants provide. In social sciences, we basically depend on what people tell us. We cannot accept their testimony as incontestable truth but neither can we, in my opinion, start from the principle that they are all lying. Making truth still depends on what Ricoeur (2004, p.278) calls the ‘witness’ triple declaration’ (1) I was there; (2) believe me; (3) if you don’t believe me, ask someone else.’ Data analysis is intrinsically based on a conscious decision of accepting participants’ declarations, being aware that such declarations are coloured by their assumptions and beliefs, previous experiences and socio-historical backgrounds. If we are lucky, what we have at the end of the process is, at its best, strong evidence but this will still never be beyond questioning. Neither should it be.

Knowledge is advanced when we question things that are known and which have been proved. Science, and I mean hard sciences as well as social sciences, is the art of making well informed questions, admitting ignorance and systematically searching for answers. Knowledge is always a building up activity, for even when we have to wipe the slate clean and start anew, we are building up on other people’s mistakes. Knowledge is never definitive but a constant becoming as someone, somewhere, someday will always find evidence that may confirm or contest the evidence and the beliefs we now hold.


  • Jardine, L., 2006. Believe it or not: The battle over certainty. BBC. [online] 28 Apr. Available at: <http://news.bbc.co.uk/1/hi/magazine/4950876.stm&gt; [Accessed 28 Jun. 2013].
  • Ricoeur, P., 2004. The Conflict of Interpretations: Essays in Hermeneutics. London: Continuum.