AuldSoul wrote:
: AlanF: Natural selection is by definition a process of trial and error, and if biological systems can be described as the products of trial and error, one would need other information to decide which was the real manner in which the systems came to be.
: What other information would one need? I am not disagreeing, I simply note that you have not described what would characterize such information.
I'm not proposing any specific information. I only meant to show that when you have two or more possible causes for the same result, you need information in addition to the result to decide among the causes. Try to answer my above rhetorical question about all humans splashing red on their artifacts and you'll see what I mean.
: Also, I disagree that natural selection is a process of trial and error.
Of course it is. Trials are done every time a mutation results in new behavior or a new structure. "Error" comes about when the modified life form dies before reproducing.
: "Trial and error" implies an attempt to achieve a desired objective that natural selection does not allow for. There is no evidence, for instance, that single-celled organisms have ever intended to survive. They simply have survived. So, survival is not the objective, it is simply the result.
I think you don't really understand natural selection. Natural selection has no a priori goals, unlike some trial and error experiments done by humans. But not all trial and error experiments done by humans have clearly predetermined goals. For example, as I write, I'm performing experiments via simulation on an electrical circuit. The circuit is so complicated and has so many interactions that I often can't predict what the outcome of any particular minor design change might be. So I make a change and look at the simulation result. If I like it, I keep the change. If not, I discard it. Most of the time I discard the changes. So I often don't quite know in advance just how a design is going to look in the end, or even how it will perform, except in general terms. And very often, my goals are by nature rather fuzzy, because I'm always pushing the envelope of what's technically possible, and have to perform many experiments to see what works and what doesn't. What works becomes a money-making product. What doesn't work is discarded. Natural selection works the same way, except that the fuzzy 'goal' is survival long enough to reproduce. So in that sense, if you insist that trial and error experiments must have goals or objectives, then the objective and the result are the same thing -- survival. In his books Climbing Mount Improbable, The Blind Watchmaker, The Ancestor's Tale, The Selfish Gene, and the latest, The God Delusion, Richard Dawkins writes about these things far more clearly than I can. I suggest you peruse them.
: AlanF: That does not mean that systems that display evidence of trial and error in their construction were designed by intelligent humans.
: I won't be drawn into a discussion of the nature of the designer(s). I am looking for specific indicators of design. "Systems that display evidence of trial and error in their construction were designed." An objective was being sought out in the design and testing process if trial and error is present.
Again, this is simply false. My everyday work proves it in terms of human design, and Dawkins explains it in terms of natural selection extensively and clearly.
AlanF