Allen MacNeill and Rock are two of the more thoughtful commenters at Telic Thoughts. Allen recently wrote a comment containing these remarks:
The presence or absence of design or purpose (technically teleology) in nature is not a "finding", it is an inference. That is, the presence or absence of teleology cannot be directly observed, nor can it be indirectly observed (as is the case for, say, atoms).
Rather, the presence or absence of teleology can only be indirectly inferred on the basis of its necessity. If teleology (i.e. pre-existing or "foresighted" design) were necessary to bring about those objects and processes we observe in nature, then it would be necessary to infer their existence and operation. Hence Dr. Behe's arguments for "irredicible complexity" and Dr. Dembski's arguments for "complex specified information", both of which they assert are arguments for the necessary intervention of a supervening "designing force" in evolution. In brief, if "you can't get here from there" without teleology, then teleology becomes an inference that flows from your analysis.
This means that the inference to teleology is an inference by exclusion. Only when one has conclusively shown that the observable objects and processes in nature cannot have come about via non-teleological processes is it legitimate to infer teleology. To be as clear as possible: teleology is a post hoc inference, not a propter hoc assumption. This is why scientists (at least those who practice in the "natural" sciences, such as biology, chemistry, and physics) do not assume the existence of teleology. Rather, we attempt to explain those objects and processes we observe without resorting to post hoc inferences, such as teleology.
That comment induced this response by JAD containing these remarks:
It’s true that teleology is a metaphysical inference, but so is dysteleology (to use the term the Ernst Haekel coined.) And, I think that both sides should be free look at the evidence and argue for their respective interpretation and/or point of view. However, such interpretations are not strictly empirical interpretations, rather they are top-down interpretations that are based on presumptions and assumptions that are themselves unproven and perhaps are un-provable. If we limit natural science to things that can be studied empirically, then both teleology and dys-teleology fall outside those limits.
I do think that there are some minimal metaphysical assumptions that are necessary to do natural science. For example, I think that we need to assume that there really is real world out there, and that the laws of nature are for all intents and purposes universal, acting the same everywhere throughout the whole history of the universe. These are a couple but not all the assumptions, that we need to do science, at least if we extrapolate our study into the past or into the future.
However, I don’t see why it is necessary for a scientist to assume that the universe, or the natural processes we observe in the universe have no plan no purpose or no ultimate meaning. If that is your philosophical or theological interpretation, fine. Argue for it in that way. But, it is not, in itself, an empirical argument, and neither is it a position one needs to accept metaphysically or methodologically to do science.
In other words, dysteleology is a post hoc inference, not a propter hoc assumption. And, it is unnecessary to do science.
History is full of examples of scientists who metaphysically saw God in nature and yet have made major discoveries and contributions. My favorite is, Johannes Kepler, who said about his scientific work:
"I was merely thinking God's thoughts after him. Since we astronomers are priests of the highest God in regard to the book of nature, it benefits us to be thoughtful, not of the glory of our minds, but rather, above all else, of the glory of God."
...as well as a comment of my own:
If you can't get there except through teleology then a teleological inference is imperative but you might be too exclusionary in your your application of a necessity standard. Direction is sufficient and bias toward outcomes of stochastic processes, which are inconsistent with probability assessments, can infer teleology. The difference can be subtle but important. A strict necessity paradigm requires a concurrent standard of demonstrating impossibility. Not only is that requirement too stringent, it may be impossible. But if observed outcomes deviate from expectations (but fall within physical possibilities) then biased outcomes are evident. If a biased stochastic process is not attributable to physical causality then a prima facia case is made for teleology.
The downside to this is a teleological predisposition toward expectations of non-teleology. We see this with some "gap" issues like the origin of life. If you cannot empirically derive non-teleological explanations based on finding suitable causes for mappings of codons to amino acids you maintain that necessity is implicated but not yet pinpointed. This is the flip side to exclusionary requirements. If some demand the demonstration of impossibility for teleology they likewise apply an always possible standard to expectations of non-teleology. That representation of a specific amino acid could not have been the causal result of conscious assignment even indirectly. It must have incidentally resulted from unexplained physical necessity coupled to a stochastic process. Conscious symbolism is unintentionally mimicked by a blind code maker.
JAD was spot on when he pointed out that "dysteleology is a post hoc inference, not a propter hoc assumption. And, it is unnecessary to do science." The meaning of concepts is framed by contrasting concepts. Old earth creationists are contrasted to young earth creationists. The contrast highlights the uniqueness of respective points of view. Dysteleology, as the default position for science, requires a rationale justifying its preferred status or it ceases to offer a meaningful contrast to teleology. Teleology is intrinsic to life. Survival and reproduction are obvious goals. Why are causal trails associated with them considered dysteleological when complete causal trails remain unelucidated?