Can Natural Selection Produce Intelligent Design?

In my opinion, evolution is an intelligent designer.There is some disagreement with the meaning of "intelligent", and often people associate intelligence with the use of logic. In my opinion, we should be associating intelligence with pragmatic judgement.Evolution is a system based on pragmatic judgement. It uses trial and error (random mutation) to generate variants of successful organisms. And then it evaluates the results of that trial and error on their success, using natural selection.

Well, fair enough.We are looking at this from different perspectives. You apparently want there to be a conscious agent before you will say that judgement is being exercised.I see that as somewhat problematic. For if you look at a single neuron, then on the same basis you would have to say that there is no judgement there either. And human decisions are merely the combined effect of all of those neurons. So the logical conclusion would seem to be that there is no judgement and no intelligence in people, either.I'm looking at judgement as the outcome of a process, rather than the decision of a conscious agent. I suppose this comes from my interest in artificial intelligence and cognitive science. I am wanting to be able to consider intelligence as arising from a community of simple judging processes.

I'm not sure what problem you are seeing. It seems to me that Darwin started with the ending point (a diversity of species), and worked backwards to come up with his theory of Evolution. Do explain which invalid conclusions you are referring to.Apparently we both agree that evolution is directed. You admit that reluctantly with "except for the filter of natural selection," while I embrace that directedness in natural selection as a simple example of natural intelligence. No, I am not assuming that. I wonder what gave you that idea.If your idea of intelligence is based on the planning and such like, coming from conscious agents, then intelligence and consciousness becomes difficult to explain. In my opinion, it's that way of looking at intelligence that gives rise to creationism, ID and similar theories. I'm trying to avoid that. I want to judge intelligence purely by the behavior, without any assumption of a conscious agent and without any requirement of specific planning for the future.

Since you are replying to me, I guess I should comment No, I don't see that at all. My posts have been related to the idea that intelligence can and does arise naturally. And if simple forms of intelligence can arise naturally, then it is surely possible that natural selection can produce more complex intelligence.If the moderators think I am taking this in the wrong direction, then I guess I will have to stop posting and see in what direction others want to take it. My understanding of the OP is that humans are intelligent designers, and that if humans are the result of evolution, then intelligent design results from evolution. As I see it, the OP was wanting a discussion of whether that is possible.It seems that you have a different understanding of what was being asked by the OP.

I don't agree with that. DominionSeraph was questioning me with respect to my use of "judgement". In the case of the NAND gate, I would not say that there is any judgement. A NAND gate is just a simple deterministic device. Likewise, I don't see any judgement in a multiple input gate, nor in the formal analysis of a logical argument. When we ordinarily use logic in our reasoning, we don't restrict ourselves to a formal analysis. We also evaluate the assumed premises, and that is where I see us using judgement. Sure. Except for "all the premises are true", this is completely mechanical and deterministic. No judgement is required, and no intelligence is required. The hard part is with "all the premises are true". When this has to do with real world matters, as distinct from formal logic propositions, the question of whether the premises are true can require considerable intelligence and judgement. But that judgement is outside what any NAND gate or other logic circuit can achieve. You seem to see intelligence in a NAND gate, and want to use that to prove that there can be intelligence in a neuron. By contrast, I see intelligence in a neuron such as we shall never be able to replicate with NAND gates alone.

Neither a NAND gate nor a multiple input logic gate will come up with an indeterminate truth value.If you are doing what a NAND gate could do, perhaps you are using some judgement. But the use of judgement is superfluous, for the decision can be reached mechanically, as by a NAND gate. The conclusion is completely determined by the input, and anybody else could, and normally would, reach the identical conclusion. Your judgement is not needed. If, however, you are doing something where there could be an indeterminate truth result, then you are doing something other than logic, and you are exercising judgement. If the conclusion could be at an indeterminate truth value, then you are not doing what NAND gates can do, and you are not just applying simple logic.In Message 20, DominionSeraph said "I don't see why a trial-and-error process couldn't come up with a trial-and-error brain." That's a good example of what I am suggesting. In many cases, a NAND gate cannot tell you whether the result of the trial was an error. Whether the result was an error is an empirical judgement, but logic is abstract, so does not make empirical judgements. This doesn't alter the fact that the gate will produce a determinate truth value.

I'm perhaps not seeing your point there. Formal logic deals with form, not with what is signified. If there is a way to harness logic chips to deal with what is signified, rather than with the form (the output level), then how to do this has not yet been discovered.I'll skip a couple of your other comments, and get to this: I think you are saying that, for a trial and error experiment, the sensor would report whether success or error, and the logic circuit would only have to deal with the signal from the sensor.That's fine. But I would say that the sensor is exercising judgement in that case, while the logic circuit is still just doing a deterministic mechanical operation on its input. The sensor is what interacts with the world, and the sensor is in some sense tuned to what its output signifies. The logic circuit is just acting on form (output level).A book report was just posted as Message 1. In that book, Andy Clark is arguing for the role of interactions, and that just logic isn't enough. What I have been saying is along the same lines.I think we are drifting a little too far from the thread topic. We should be discussing whether natural selection can produce something like human intelligence. Let's not get too involved in speculation about what could work to achieve AI.

The you are talking about trial and error testing circuits, where I was talking about trial and error testing behavior (interactions with the world). I think we might be talking past one another.