Science is a method for learning, a way toward understanding. The goal is to develop an explanation for an observation (or a series of observations), keeping in mind that there could be any number of potential explanations. It is the role of the scientist to formulate possible explanations and to devise and perform tests to determine the validity of various explanations. In science these explanations are known as hypothesesA proposed explanation for an observation.. Each hypothesis must be subjected to a variety of experiments as scientists try to eliminate explanations and settle on one explanation as “true.” For this reason, they must be falsifiable (testable).

Skeptical non-scientists often dismiss scientific explanations as a "mere" hypothesis or theory, as if it were a “guess”, or at best, an “educated guess”; speculation. Nothing could be further from the truth. A hypothesis may not have much empirical support at first, but that can change. As a hypothesis is subjected to scrutiny, tested in a variety of ways, it can gain support to the point at which it is effectively a fact. If a hypothesis is accepted as fact, it is because an overwhelming body of evidence has consistently failed to discount it. It is seen as “true” even if it can never be “proven” true.

Scientific theories, too, are not unsupported conjecture and are they not just well-supported hypotheses. A theory is a testable explanation for a broad range of related phenomena or observations[#](Futuyma 1998), a group of interconnected ideas that accounts for a great variety of phenomena. And a theory must be strongly supported by evidence. Theories are a pretty big deal in science.

There are a lot of scientific theories: gravity, plate tectonics, and the atom, to name just a few. Evolutionary theory is the body of interrelated ideas about the causes of evolution. It is not conjecture. In fact, the main tenets of evolutionary theory are so well supported that nearly all biologists accept them as fact.

Like all theories, evolutionary theory is subject to revision; as we learn more we may find that some details may be wrong. We expect theories to change as they are tested; Experiments may reveal that parts of the theory must be discarded, modified or supplemented. For example, before 1960 or so, almost all geologists believed the continents were fixed in position. Now all accept the theory of plate tectonics and continental drift. Darwin had no knowledge of genetics when he wrote The Origin of Species[#](Darwin 1859). Consequently, some of his theory had to be revised as scientists learned more and more about genetics and heredity. As theories are tested and revised, they are honed. With each new test, we get a deeper understanding of the phenomena a theory explains. The revision of a theory is not a bad thing, but a step toward a more complete understanding of the world around us.

There are at least three reasons that ID is not a scientific theory:

ID is not falsifiable.
The "central tenet" of ID says: "The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection"[#](Discovery Institute 2006). But there is no conceivable way to test (i.e., try to falsify) this claim. We could point out the many aspects of the natural world that would seem to argue against the existence of a designer, but this approach overlooks one problem: We can't identify the intelligent cause[#](Discovery Institute 2006) [**], and without knowing the designer's limitations, motivation and rationale, it is not possible to test this theory. It may be true, but there's no way we'll ever know. Which means it is not science.

For example suppose we ask:

-Why did the designer give bats the same respiratory system as ground-dwelling mammals instead of the much more efficient respiratory system of birds?

-Why don’t birds have arms in addition to wings?

-Why did the designer make animals unable to photosynthesize?

Because "intelligence" does not necessarily mean "omnipotenceHaving absolute power over all.," or even
"omniscienceHaving having infinite knowledge, understanding, and insight.," one can always claim that bad design does not rule out a designer. We can never rule out a designer because we cannot identify the intelligent cause nor its motivation or rationale. As Karl Popper pointed out, "Irrefutability is not a virtue of a theory (as people often think) but a vice."[#](Popper 1963)

ID makes no testable predictions. (see the illustration How science works)
ID simply asserts that the complex structures are consistent in appearance with the way a designer would make them. That is, ID says they look like they were designed by an intelligent entity, therefore they were designed by an intelligent entity. But many things bear the mark of intelligence when in fact they are something else. Talking parrots, for example. Without a theoretical framework from which to make testable predictions, ID is useless to science.

The main argument for ID is a process of elimination based not on experimental results but on a logical fallacy.
ID advocate Michael Behe[#](Behe 2006) has developed the idea of “irreducible complexity,” arguing (incorrectly) that complex organic structures cannot be “the product of an undirected process such as natural selection”[#](Discovery Institute 2006). Therefore, he says, we must conclude that they are the product of an intelligent designer. This "Argument From Incredulity" - I can’t see how it’s possible, therefore it can't be so (see Intuition) - also suggests that there only two possible explanations for the origin of species: Evolution by natural selection vs. conscious design by an intelligent being. A lack of imagination is no substitute for scientific rigor. You cannot simply accept a hypothesis because others fail. You must test and evaluate every hypothesis independently and on its own merits.

(In attempting to avoid that criticism, Dembski[#](Dembski 2001) proposed a method of inference to “recognize” design. He argues that we can conclude that an object is intelligently designed if it shows “specified complexity”, complexity that matches an “independently given pattern.” But he misses or ignores an important point: in evolution there is no “independently given pattern.” Species do not evolve towards a predetermined target; evolution has no goal. Yet his conclusions, including those based on his use of so-called No Free Lunch theorems[#](Wolpert & Macready 1997), critically depend on evolution having a preset (“specified”) goal.)

And even if we accept a designer, all we have then is a giant black box (see cartoon above right); we can't identify the intelligent cause, and without knowing its limitations, motivation and rationale we can never know how or why species were made, nor can we understand their physiology, ecology or behavior. This puts some serious limitations on the utility of ID for understanding biology. We can never know the reasons for anything, if we use this theory.

Imre Lakatos noted that the crucial element for a new theory to replace an existing one "is whether the new theory offers any novel, excess information compared with its predecessor and whether some of this excess information is corroborated."[#](Lakatos 1970) ID decreases our understanding of our world and makes no predictions.

Until ID can can be tested on its own merits, it does not warrant inclusion in a scientific discussion of evolution.

Readers wanting a more detailed criticism of ID can visit talk.origins.org. Information on ID from ID advocates can be found at the Discovery Institute website. And as you consider the issue, keep in mind that evolution is not proposed to explain the origin of life, just the origin of species. It's an important distinction and one that is often blurred. See the quote by Darwin on the home page.

Futuyma, D. 1998. Hypotheses, Facts, and the Nature of Science. pp. 9-12 in Evolutionary Biology, 3rd ed. Sinauer Associates, Sunderland, MA

Lipton, P. 2005. Testing Hypotheses: Prediction and Prejudice. Science 307: 219-221