Debunking Plantinga: We Can Evolve Ways To Discover Truth

In the video below, Paul Griffiths, a philosopher of science from the University of Sydney, explores how knowledge can evolve, against arguments from people like Plantinga who claim that natural selection would only select for useful beliefs but give no confidence that our beliefs attain to actual truth.

For much more detailed treatment specifically on the nature and structure of Plantinga’s arguments, check out Tyler Wunder’s discussion with Luke Muelhauser from a few years ago on The Pale Blue Dot. Wunder wrote his dissertation explicating and responding to Plantinga’s work.

Also “Plantinga Schmantinga” from Reasonable Doubts is good too.

"Getting out of religion and back into nature is incredibly difficult. Cult-ural Science, cult-ural religion, ..."

Have I Considered Catholicism Sufficiently?
"Following on from my last past here - As pervasive as the term original sin ..."

Have I Considered Catholicism Sufficiently?

Browse Our Archives

Follow Us!


What Are Your Thoughts?leave a comment
  • http://wateringgoodseeds.tumblr.com/ Shira Coffee

    OK, I finally have time to comment on this. I’ve listened to it a couple of times, and also found the paper he mentions (at http://paulgriffiths.representinggenes.org/wordpress/wp-content/uploads/publications/B1_33.pdf )

    First off, it’s kind of misleading to label this as a refutation of Platinga, since Griffiths doesn’t actually address any argument of Platinga’s. Instead, he addresses a related (?) argument put forth by Stephen Stich. This strikes me as a bit bait-and-switchy.

    In any case, I think Griffiths makes two excellent points in the video. First, it is incoherent to talk about selection “for reproductive success” or “for fitness”. Reproductive success / fitness are universal evolutionary goals, and cannot be used to explain any particular adaptation or group of adaptations. It’s like a monotheist explaining that something is so because God wills it. There is just no explanatory power there. Second, Griffiths is right to argue that an imperfect adaptation (like the heuristics and biases employed by the human brain) may nonetheless confer some important fitness advantage. The issue is to deduce the nature of that advantage.

    This is where I think Griffiths gets it wrong. Attractive as it would be to demonstrate that our brains and senses are adapted for “tracking truth”, this formulation is nearly as overgeneralized as the idea of fitness. If you ask why organisms “track truth”, you have to take a step backward and begin to look at ways that is beneficial to the organism. Moreover, you have to explain how “tracking truth” was beneficial for very simple organisms, or even for proto-organisms in RNA world, etc.

    Furthermore, there is good evidence that our brains are not simply imperfect trackers of truth, because they actually and actively fool us. The most basic example of this is the creation of the phenomenal-self illusion, what Metzinger calls the “ego tunnel” and Dennett calls the “Cartesian theater”. This is not a cheap illusion, either, since it requires rather a lot of calorically expensive brain power to maintain. It is therefore obviously an adaptation rather than an accident, and the question to answer is, what does it contribute to fitness?

    It seems to me that a far better high-level answer to the question of what brains and perceptions contribute to our fitness is this: they help perform the very difficult and complex work of choosing the best next behavior for an organism.

    This explanation is superior to “tracking truth” for several reasons. First, it is applicable to very simple organisms and within very simple and primitive ecosystems. (For instance, even in prebiotic “Chemistry World”, the order in which reactions occur would probably affect a given chemical system’s “longeivity”.) Second, such strategies can be implemented by various adaptations — from enzymes in Chemistry and RNA World to proto-sense-organs on cell walls to ways of laying down memories of what worked in the past for, say, flatworms, to complex computational strategies in humans.). Third, these adaptations readily scale and build on each other. We still have lock-and-key structures on our own cell walls, and the cones in our eyes make use of the same chemicals that allow single-celled photosynthesizers to rise and sink in the ocean to gain access to usable wavelenghts of light and minimize exposure to harmful wavelengths.We still make use of the same Skinnerian, ABC learning that flatworms use, etc. Fourth, it’s perfectly possible to both model and directly test the ways in which various kinds of organisms “choose” behaviors depending on circumstance. We can very likely find the exact mechanisms. It is, by contrast, rather ridiculous to try to test ways in which simple organisms “track truth”.

    Finally, from a philosophical point of view, the idea of “best-next-behavior strategies” would give a much firmer basis for understanding human issues such as intention. (Why yes, I am reading Anscombe… why do you ask, lol.)