I've heard many people ask whether evolution exists. Here is the answer from the philosophy of science's perspective.
When Darwin first suggested evolution, it was a theory, based on survival of the fittest and separation of breeding colonies. Darwin made observations to substantiate his theory, but it still wasn't a really good scientific theory.
It wasn't a really good theory, because a good theory can be directly refuted by experiment compared to a 'controlled scenario' (where the theory does not apply). For Darwin's theory, there was no way to make an experiment with a control group, so it wasn't a really good theory.
However, while it was not a good theory, It was a very useful theory. It was a very useful theory because it created avenues of new thought , opening the door to many possible predictions. Nonetheless, at the introduction of Darwin's theory on the origin of species, there were theological debates which overshadowed a purely scientific response, so the relative merits of the theory, from the perspective of the philosophy of science, was largely ignored.
Since then, our understanding of DNA has evolved, and we are able experiments directly on bacteria colonies to test the theory of evolution. Many additional observations on both genetics and behavioral traits of social animals have been made, for example. So in total, evolution is therefore no longer just a theory. It is now a model. The model explains molecular, cellular, sexual, social, behavioral, historical, and paleontological phenomena.
To say whether one 'believes in' a model is rather meaningless from a scientific point of view. It is simply a model. Science can produce no more than theories and models. A model that explains observations is considered useful. When a model can produce many unfalsified theories, it is considered a good model. Some models are more useful, and some are less useful. Some are models are good, and some models are less good. Just as for theories, models are evaluated based on their merits.
Some experimental results may invalidate the model for a particular domain, but that does not 'disprove' the model. Experiments can corroborate theories derived from the model, or disprove theories derived from the model. If an experiment disproves a theory, the model is invalidated for that application domain, but the model still exists, and if it explains other phenomena, it can still be a good model. The only way to 'disprove' a model is to show that it contains some tautological error which causes it to be inconsistent with its own axioms.
You can't choose to believe whether or not the model exists. The model exists. It is not questionable whether the model exists or not. The evolutionary model exists. From a scientific perspective, you don't choose to 'believe in' the model or not. The model may produce some predictions that you don'ty believe will happen, but that doesn't mean the existence of the model itself is questionable. The model is known to exist.
Due to the nature of science, alternative explanations are often compatible with those provided by models. If you want to believe God invented the universe in seven days, you could say all the fossil records and all the genetic evidence and all the other evidence was created simply to tease scientists. There's no way to disprove it. You are free to believe in any creation myth you like. You can believe the universe was vomited up by the titan Ouroboros after he got drunk, like the ancient Greeks did. However the universe actually was created, it could have been made just the way it is, including all the phenomena that are in accordance with the model of evolution. However it was actually created, it isn't really relevant to the model of evolution. The model can still be used to form substantiated theories, it is still a good model. and the model still exists.
As explained above, asking whether one 'believes in evolution' is a semantic confusion. It's nonsense. You can't choose to believe that the model exists or not, because it does.
Similarly, asking whether the creation in seven days means evolution is 'wrong' is also a semantic confusion. It's nonsense. There is no right or wrong, or good and bad, about a scientific model, A model exists to explain observations in a simple way. That is all a model is, and this framework, of defining theories based on models, is the basis of how all scientific inquiry works.
As stated above, these are not really arguments 'against' evolution. Rather, they are facts that the standard model cannot explain easily.
The first fact is that the number of fossil records showing gradual evolution of different species is far more incomplete than the original Darwinian theory would lead a rational person to expect. While some records are notoriously complete, such as the often cited evolution of the horse species, others are conspicuously absent without any explanation.
The second fact is that there are sudden jumps in species difference which are difficult to explain, for example:
- Until wings are developed enough to permit gliding or flying, it is difficult to explain how rudimentary wings could develop. This is because, until they are sufficient to provide a competitive advantage, rudimentary wing stubs would reasonably be considered to be only be a hindrance.
- Sudden genetic mutations are shown as possible--for example, the human Turner syndrome occurs quite frequently (~1 in 3,500) and could have evolutionary benefits in other circumstances. However, the Turner syndrome, like almost all extreme genetic mutations, is infertile.
- Some variations are difficult to explain by evolution due to the parallel requirement of several unrelated biological functions. The Poison Dart Frog, for example, excretes lethal poisons from skin glands when under threat. The poison is extracted and purified from eating poisonous ants and then stored in the skin gland. Other frog species simply belch up the poison. This Poison Dart Frog must have developed three mechanisms simultaneously, first to extract and purify the poison; second, to make the modified skin glands for its storage; and third, the neural control to excrete it only under threat, otherwise it would kill its mate and not be able to breed. It is difficult to explain how three unrelated biological subsystems could 'evolve' concurrently and simultaneously by genetic variation, especially as without all three working at once, the frog would kill itself with its own poison.
To explain sudden jumps in evolution, there are two main schools of thought. The first is that mass extinction by a global-scale event creates a void into which existing species evolve very rapidly. Experience from forest fires leads one to think the available resources are populated in a series of separate ecological stages. However, recent experience in population swings from oil disasters, such as the Exxon Valdez, for example, suggests in fact the new populations oscillate rapidly back and forth, with different species becoming repeatedly dominant--exhausting food sources and shrinking again; and then becoming dominant again as the food source repopulates; and so oscillating back and forth, creating many cycles of deaths and population surges, and so accelerating evolution. The continuing and repeated rise and fall of each species creates greater opportunities for genetic differentiation to become significant in a small span of time.
The other addition to the Darwinian theory is 'quantum evolution,' and its derivations. These theories state reasons why, in an isolated population, a marked genetic difference may become dominant enough to cause species separation. For example, in an isolated group of humans, redheads may be rare, but desired so much in the reduced population group that redheads create significantly more offspring. This is one basis of 'quantum evolution' theories.
Nonetheless, while evolution clearly can be shown to take place now at the bacterial level, it is still impossible to demonstrate how such a mechanism has worked over a period of time sufficient to cause the evolution of different animal species, because it is not possible to construct a full experiment, with a control group, to demonstrate that the Darwinian theory is necessarily a causal force in all animal species development.
Even so, the integrated model of evolution is useful enough that it is likely to remain, much as it is, for the foreseeable future.