The Naked Scientists

The Naked Scientists Forum

Author Topic: Can an experiment measure bias in creationists?  (Read 2256 times)

Offline Supercryptid

  • Hero Member
  • *****
  • Posts: 606
    • View Profile
    • http://www.angelfire.com/sc2/Trunko
Can an experiment measure bias in creationists?
« on: 01/09/2013 05:18:19 »
Take note that I am using the term "creationist" mostly in regards to old Earth and young Earth creationism, not evolutionary creationism.

I recently had a debate with an old Earth creationist and it got my gears turning. It seems to me that creationists have a couple of prominent views: any naturally-occuring trait in a living thing is "optimally-designed", whereas any trait that is the result of a mutation (regardless of any perceived beneficial effect to evolutionists) will inevitably represent a "degradation of some optimally-designed function". What makes this interesting is that it implies that, given sufficient information on the function and genetic origin of a trait, creationists should be capable of recognizing whether it arose in nature ("designed") or whether it arose from a mutation ("degradation of design"). I suggest that an experiment could be prepared which could test to see if this claim is true. Importantly, the experiment could be designed to see if bias has any effect on their observations.

Consider the following set up: A creationist is presented with a pamphlet containing extensive information on three different microbe types. The species involved are kept secret, and are simply labelled "Microbe A", "Microbe B" and "Microbe C". The information regards the details of the function, cellular mechanisms and genetics behind citrate metabolism present in each microbe type. The creationist is then asked to write an essay about each of these three microbes (as short or as long as he/she wishes) describing the perceived origins of these traits and justifying their position.

Unknown to the creationist, one of these microbes would be the citrate-utilizing E.coli mutant that developed during the Lenski experiment: http://en.wikipedia.org/wiki/E._coli_long-term_evolution_experiment. The other two microbes would be of two different types which have the ability to metabolize citrate in nature (that is, there are no laboratory-based experiments that witnessed any such citrate-metabolizing mutation arising firsthand in them). For the sake of the experiment, the mechanisms in these three different microbes need to be as similar as possible. The reason that their identities are kept secret is to eliminate bias. I strongly suspect that:

(1) If a trait is known to exist in nature, a creationist will automatically search for justification to call that trait "designed".
(2) If a trait is known to arise from mutation, a creationist will automatically search for justification in calling that trait "degradation".

By preventing the creationist from knowing the origin of these citrate-metabolizing traits, they are forced to come to their conclusions based on the facts alone. There are different results that would be possible and each has different implications:

(1) The creationist concludes that all three traits are designed, in which case it would be proof that a trait which is known to have arisen from a mutation can appear to be designed (even to a creationist).
(2) The creationist concludes that all three traits are degrading mutations, which would mean that two traits which they would have otherwise perceived as designed somehow look imperfect or damaging upon closer inspection. This would call into question what a designed trait is supposed to look like.
(3) The creationist concludes that the mutated trait looks designed, and one or both of the natural traits look degraded. This would call into question the ability of the creationist to recognize what a designed trait is in the first place.
(4) The creationist can tell that there is a difference between the one mutated trait and the two natural traits, which would support their own arguments as internally-consistent.

Here is where the bias really comes in. After the creationist finishes the essay, he/she is informed that one of the traits is mutation that happened in the lab and two are naturally-occuring. They are then asked if they would have written the essay differently if they had been aware of the origin of the three traits prior to beginning. If the creationist got it "wrong" (wrong from a creationist perspective, that is), then they would mostly likely say that they would have changed their conclusions (no doubt to support the standard creationist perspective). If the creationist does do this, then that would demonstrate that they do indeed have biased thinking (biased based on the perceived origin of a trait).

Now with good science, we can't go on one test to tell us much. We would need to test many creationists to see if there is an over-arcing pattern. Preferably, the creationists need to be well-educated and even professionals in some field of biology. We would also need to keep the creationists that are taking the test from knowing that it is bias that is being tested for (to further reduce the possibility of bias, ironically). The test also needs to be genuinely fair: all of the information provided to them of the traits must be true and not written in any deceitful or leading way. We actually have to give them a chance to demonstrate that they can tell the difference between design and degradation from the physical evidence alone.

I also suggest that every aspect of the experiment be open to public knowledge (via the Internet or other means). This is both so that no one can claim that the test was unfair (as they can view the pamphlet and essays themself) and so that flaws can be recognized and corrected so that an improved experiment may be performed in the future if need be.

One of the challenges of this would be in giving the creationists sufficient information about the traits to be fair without revealing that one of the microbes is the (now famous) Lenski E.coli strain. If they are able to deduce that it is the Lenski microbe, then the risk for bias is too high and the experiment becomes useless.

Any thoughts on how to improve this?
« Last Edit: 01/09/2013 05:24:22 by Supercryptid »


 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4729
  • Thanked: 156 times
  • life is too short to drink instant coffee
    • View Profile
Re: Can an experiment measure bias in creationists?
« Reply #1 on: 01/09/2013 10:26:09 »
Not sure what the object of the experiment is. My approach would be to cut to the chase and ask "are you a creationist?". If yes, then assume 100% bias, as with any other -ist (communist, fascist, flatearthist, methodist, scientist....). Why else do we adopt labels if not to proclaim our prejudices? 
« Last Edit: 01/09/2013 10:29:18 by alancalverd »
 

Offline Supercryptid

  • Hero Member
  • *****
  • Posts: 606
    • View Profile
    • http://www.angelfire.com/sc2/Trunko
Re: Can an experiment measure bias in creationists?
« Reply #2 on: 01/09/2013 17:46:21 »
Not sure what the object of the experiment is. My approach would be to cut to the chase and ask "are you a creationist?". If yes, then assume 100% bias, as with any other -ist (communist, fascist, flatearthist, methodist, scientist....). Why else do we adopt labels if not to proclaim our prejudices? 
Because that would prove nothing. Creationists claim that their observations and conclusions in regards to evolution are based on an objective observation of available evidence, so the goal of this would be to see if this claim is true (that is, that their conclusions involve objective thinking). More importantly, if they are biased, then it is to prove to them that they are biased.
« Last Edit: 01/09/2013 17:48:01 by Supercryptid »
 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4729
  • Thanked: 156 times
  • life is too short to drink instant coffee
    • View Profile
Re: Can an experiment measure bias in creationists?
« Reply #3 on: 01/09/2013 18:23:06 »
Aha! Well then, let's just pick on one simple design: nipples on men. Here's a feature that has no function but can go wrong and kill the owner (male breast cancer is rare but frequently fatal). In what way is the assertion of optimality either true or derived from observation?  Or are we talking about an idiosyncratic definition of optimal? 

Not sure how you can distinguish between an intentional change (evolution by design) and an accidental one (evolution by mutation) by looking at the end product. The minimum assumption is "chance" because we know that all sorts of mutations occur all the time. To prove "intention", you need to demonstrate  foreknowledge of the outcome. So you could take a rapidly mutating organism and demand that your creationist tells you (a) what the dominant genetic sequence will be after ten generations and (b) why the designer, who is capable of making everything in perfect form,  needed ten generations to get there from here.   

Creationism is useless because it adds complication but has no predictive value. On the other hand the assumption of bounded randomness does give us a good statistical prediction of the outcome of breeding peas or cats. All we mean by evolution is that the process that results in intergenerational change, has been going on for a long time.     
« Last Edit: 01/09/2013 18:48:40 by alancalverd »
 

Offline Supercryptid

  • Hero Member
  • *****
  • Posts: 606
    • View Profile
    • http://www.angelfire.com/sc2/Trunko
Re: Can an experiment measure bias in creationists?
« Reply #4 on: 01/09/2013 23:02:40 »
The goal is not to prove creationism wrong or evolution right. The goal is to demonstrate that they have bias in such a way that they cannot deny it.
 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4729
  • Thanked: 156 times
  • life is too short to drink instant coffee
    • View Profile
Re: Can an experiment measure bias in creationists?
« Reply #5 on: 02/09/2013 00:28:36 »
The practical evidence of bias is making consistently wrong predictions. But since you will sometimes be right by chance, you need an infinite series of experiments to distinguish between bias and a run of bad luck. However, being unable to make a prediction (or in the case of male nipples, explanation) at all because your bias demands a single step to perfection, is an absolute mathematical proof of that bias.
 

The Naked Scientists Forum

Re: Can an experiment measure bias in creationists?
« Reply #5 on: 02/09/2013 00:28:36 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length