Estimates suggest that over five million people have now watched the debate that was streamed live last week between CEO of the Planetary Society, Bill Nye “The Science Guy,†and president of Answers in Genesis, Ken Ham.
The debate benefited from a fairly concrete question—“Is creation a viable model of origins in today’s modern, scientific era?â€â€”although the conversation suffered at points from a lack of clarification about what made a model of origins viable. (Both Nye and Ham alternated between taking a model’s viability as dependent on whether one could successfully develop technology or otherwise be scientifically innovative while accepting the model and taking a model’s viability as dependent on whether or not the model accords with available observational data.)
While usage of the term ‘viability’ was shifty, both parties were clear which view they were referring to when using the term ‘creation.’ They were referring to a version of young earth creationism based upon a literal interpretation of the book of Genesis — i.e. the view that God created the world in six literal days roughly 6,000 years ago and that there was a world-wide flood roughly 4,000 years ago. The question of whether this specific position is viable today as a scientific model of origins was the question at hand.
In addition to citing empirical data, both Nye and Ham made use of a number of philosophical assumptions in making their cases. The differing philosophical assumptions that are part and parcel of embracing either a naturalistic or religious worldview were to be expected. However, many of the most philosophical interesting (and often philosophically suspect on Ham’s part) assumptions were those relating to philosophy of science, confirmation theory and epistemology.
A prime example is the difference between Nye's and Ham’s understandings of prediction. Nye’s understanding of what counts as a successful scientific prediction is in keeping with mainstream science. A prediction is successful if at some actual point in time it is put forward and then at some actual later point in time what was predicted is found to be the case. Ham’s understanding of a successful scientific prediction appears to be broader. It would include Nye’s types of predictions, but Ham also seems to count a view as having predictive power if the observational data we actually have is what we would have predicted on the view even if we didn't have the observational data we do in fact have. It’s a sort of conditionalized or “as if†predictive power. Thus, on Ham’s account, to claim that the way the world is aligns with the type of world one would posit on the creation model is to claim that the creation model has predictive power. But, on Nye’s account, this fails to count as an example of the creation model being used to make a prediction — because predictions are things which are confirmed by observational data that is collected at a later time than when the prediction was made.
Having the kind of conditional predictive power that Ham is referring to is without a doubt a valuable thing. It is a necessary condition for a coherent worldview. Without such cohesion a theory is at best in tension with observational data and at worst contradictory to it. But for a model to be predictive in Nye’s sense provides additional grounds for taking that view to be viable—both on the grounds that predictive power in this sense is highly useful for innovation and on the ground that predictive power in this sense is often a good indicator that the model is on track with the truth. Thus, to show that a model is predictive in Ham’s sense is to provide less evidence in favor of the model’s viability than it is to show that the model is predictive in Nye’s more standard sense.
The difference between how Nye and Ham understand prediction is the result of a more fundamental theoretical difference — i.e. a difference in their views about what counts as an acceptable “starting point.†A recurring point for Ham was that both he and Nye were dealing with the same evidence — the same observational data — and that “it comes down to our interpretation of the data.†This may be true. But some interpretations are just better than others. In science, considerations like a model’s parsimony, predictive power or explanatory fit are criteria that are frequently used to determine how good an interpretation of the data a given model is. Another important factor in assessing the quality of an interpretation is the interpreter’s starting point. Just as some interpretations of the data are better than others, so certain starting points are better than others.
In mainstream science (or any discipline that embraces an evidentialist epistemology), a theorist’s starting point is with her evidence — and in the case of science this is her observational data. Explanatory models are posited which appropriately account for the data and predictions are made, in part, in order to confirm or disconfirm the models posited. When your starting point is your evidence, your goal becomes to construct the model that best fits your evidence. And you are able to use prediction as a way of learning whether or not you’re onto something. For Darwin and most of the evolutionary theorists that have followed him, this seems to have been the approach. After examining the observational data, the evolutionary model was posited because it was taken to be a good fit for the data. From there, scientists used the model to make predictions, and in many cases they discovered what they had predicted they would.
Ham freely admits that his starting point is not the observational data. His starting point is his model. While a good scientist’s job is to put forward the model that best explains the data, Ham’s strategy is to find a way to make the data fit the theory he already has. On such a strategy the minimum threshold for success is obtained so long as the data is not in outright contradiction with the model (and Nye would argue that Ham’s model doesn’t obtain even that). My point here is simply that if your starting point is not the evidence, then your starting point is inferior from a scientific or evidentialist perspective. Of course, one could reject such a perspective. But in the context of this debate — where the question is whether or not creation is a viable scientific model of origins today — that simply wasn't an option for Ham.
In addition to adopting a problematic theory of science and of confirmation, Ham also overlooked some important epistemological points. On multiple occasions, he made a point of arguing that you can’t “prove†anything about the past and criticized Nye for “assuming things about the past that aren't necessarily true.†Ham acted as if his and Nye’s mutual inability to deductively prove things about the past somehow put their positions about the past on equal scientific footing. If we grant Ham the point that we can’t deductively prove things about the past, there is still a big problem with his mindset. In short, Ham seems to assume that all fallible theories share a certain sort of equality.
But this is not the case. If you and I walk into your kitchen and see that a vase has been knocked over, it may be the case that neither of us can prove with certainty what caused the vase to fall. But if you hypothesize that your cat knocked the vase over and I hypothesize it was a herd of purple unicorns, clearly your hypothesis — although it lacks certain proof — is the better hypothesis. In short, Ham ignores the fact that epistemic justification comes in degrees. To show that an opponent’s view is fallible is not sufficient to show that an opponent’s view is not the superior position.
What’s my point?  It’s that the philosophical moves Ham made cannot be used to ground an acceptance of creation over evolution. This, of course, does not rule creationism out — either new evidence could be acquired making creationism more likely or one could choose to simply have faith in a worldview that runs counter to the evidence. But the latter clearly is no longer  science — and if creationism is being accepted on grounds of faith contrary to evidence, then the answer to the question of the debate is surely no.
-Mark Satta
Leave a Reply