You are wrong on that point. Science relies exclusively on naturalistic phenomenon. To go outside those bounds, means you are leaving science behind. It is definitely not circular. ID is the circular arguement where the question of design gets pushed back because ID fails to account for the origin of the necessary complexity of the designer. Who designed the designer of the designer?All historical theories depend on making inferences based on indirect observations in an attempt to reconstruct past conditions or causes from present facts.
If you take a look at modern science, there is very little that is directly observed. Even in the hard sciences, data are indirectly inferred from instrument measurements. Theories based on historical sciences are tested by further observation which is equally valid, scientifically, to replicate instrument measurements on a single sample.
The fundamental probel with ID is that it does not offer a test that can falsify it. There is a flaw in your example of a homicide investigation. Tests are done to determine cause of death and the mechanism of death. A murderer leaves physical evidence. In ID, there is no evidence of intelligent intervention. It is possible to philosophically say that ID 'shadows' evolution, but there is no evidence of direct physical interaction of a designer. The other problem is that you say either hypothesis my be true. Evolution is not an hypothesis. Evolution is built upon millions of observed facts and is the model that explains these facts. ID is not even an hypothesis because it contains no testable postulations.
You are asserting that to be scientific a theory must be naturalistic. Why must a theory be naturalistic to be scientific? To answer that such theories are not scientific because they are not naturalistic simply assumes the very point at issue. What independent metaphysical criteria of methodology disqualifies theories that invoke nonnaturalistic events—such as instances of agency or intelligent design? What non-circular reason can you offer?
The fundamental probel with ID is that it does not offer a test that can falsify it.
If the same standard is applied impartialy to evolutionary theory then it doesn't meet the strict criterion of testability by direct direct verification or repeated observation of cause-effect relationships either. There are a host of Darwinian theoretical postulations of past, unobserved and unobservable events that purport to account for present classes of facts and present biological data. that cannot be directly tested. In either case the "testabilty" lies in the putative explanatory power rather than verification by direct and repeated observation. In the case of a theoretical postulate of ID, the past action of an unobservable agent could have empirical consequences in the present just as an unobservable genealogical connection between organisms does. It is only a hegemony of metaphysical bias cloaked in a lab coat that denies the possibilty that the actions of an unobserved and observable agent (as opposed to an impersonal mechanism) in the past could have left empirical evidence in the present.
A murderer leaves physical evidence.
Exactly my point.
In ID, there is no evidence of intelligent intervention... there is no evidence of direct physical interaction of a designer.
"There is no evidence" is a pretty strong statement. Are you sure about that? I also notice your use of the terms, "intervention", and "interaction". Both of these terms imply assumptions, dare I say, "theological" assumptions, about the nature of or the way a putative designer should have or would have designed that may or may not necessarily be the case. Darwin did the very same thing, but it was very odd for him to support his scientific theory in such a way. I do know this, if you assume there is no designer a priori, you would never find one even if there really was one, and had left a trail of blood and fingerprints all over the place.
Cordially,