Scientific findings are always changing

This is another archival repost originally written for the old blog in oct 2007.

In an episode of Friends, Ross and Phoebe have a somewhat implausible argument about Evolution. The argument ends with Ross conceding a point about probability and the inability to prove anything outside of mathematics (of course, the script writers avoided these technical details and make it look like he’s conceding defeat). To Ross’ despair, Phoebe thinks worse of Ross for making this concession, and having anything less than 100% solid certainty. As Thomas W. Martin writes in Seed:

In the present cultural climate, altering one’s beliefs in response to anything (facts included) is considered a sign of weakness.[1]

Some people are not only resistant to specific changes, but they are resistant to the entire concept of change. Raised in a system where all knowledge is certain and dictated in an old book of scrolls, one must have quite some difficulty with the idea that other people could consider knowledge to be provisional approximations, refined in response to experiment. When confronted with this aspect of the scientific method, creationists ask: “how can we believe any of it, when it could change tomorrow?”[2][3]

Science, of course, does change over time. New ideas develop, and old ideas fade away. But it’s not quite a simple as the caricature. We can categorise the sort of changes we see. There is the routine turnover of hypotheses and assumptions which turn out not to be supported by the evidence — for example, the aquatic-ape hypothesis, or the assumptions that humans must have hundreds of thousands of genes (we have 20-30,000) and humans must have diverged from the other apes tens of millions of years ago (we diverged from chimps 4-5mya). But evolution is a well supported theory, not a young hypothesis or assumption. Theories undergo a process of refinement, rather than a constant turnover, as a new piece of the puzzle fits in — from the small scale of discovering a new fossil or protein, to the large scale of the synthesis of evolution and genetics, or the oft-cited Einsteinean revolution of physics. At the level of the theory, these changes are either trivial, or the illumination of a non-fatal error or omission: our knowledge grows and changes, but is not overturned. Most, if not all of the historical examples of major scientific revolutions completely overturning previous ideas, have been the triumph of science over pre-scientific superstition — Copernicus and Darwin being the obvious examples. So, while we will discover many new things about evolution, and it may yet even go through a significant transformation, it will not be overturned in favour of creationism.


  1. Thomas W. Martin, 2007. Scientific literacy and the habit of discourse. Seed Magazine
  2. Egilson, G. H. 2003. ”For creation”. Post on
  3. Matthews, Michael. 2003. ”A century of fraud”.

Leave a comment

Your email address will not be published. Required fields are marked *