We can say evolution of the human species is a fact because it is a fact.  Just because no one was there to see humans evolve (and it wouldn't be possible to do so because evolution, despite what you might think, doesn't take huge leaps and bounds through phenotypes) doesn't mean that it didn't happen.  No one alive today was there to see the signing of the Constitution of the United States but that doesn't mean that the Constitution wasn't signed. "Wait!" you cry. "But we can see the Constitution and it had to come from somewhere" and you might even think I have delivered myself into your hands because the Constitution was designed. However, the point I am making is that you can do historical science. It's very possible and evolution is a historical science.  But I get ahead of where I should be.

I'm going to assume, for the moment and against all experience, that you are an intellectually honest person who simply has not read any evolutionary theory. If you've never taken an upper-division university course on biology I would not be at all surprised if you knew next to nothing about evolution.  It simply is not taught well at the high-school level.

Before I begin explaining what the theory does and does not say (and I apologize, at the outset, that this will be a series of long posts) I'm going to define some terms.  This is a scientific discussion and so the definitions I'm going to stipulate are the ways these terms are used in the physical sciences.  I recognize that there are colloquial usages but they should not trouble us here. 

Theory:  In science a theory is a well-tested explanatory framework that models a set of phenomena in the natural world.  It is built up by accretion of hypothesis, which seek to explain an observation.  In order for a theory to be considered robust, it must withstand falsification.  Denial is not falsification.  Falsification can happen when the theory makes a prediction and an observation is taken and the prediction is shown to be false in a critical (not trivial) fashion.  If a high number of anomalies appear that the theory is entirely incapable of dealing with then the theory is weakened. However, the mere presence of anomalies do not weaken the theory if the theory can find a way of dealing with them in a manner that does not require the needless interjection of Deus ex Machina constructs.

Species:  A species is a reproductively isolated population, sharing a common genotype with normal variation and having a common phenotype.

Reproductive isolation:  Reproductive isolation means that populations of two groups cannot interbreed. 

Genotype:  Genotype is typical genetic motif for a species.  Within a genotype there will be alleles (variation from the main sequence) but it will be variation on a theme.

Phenotype:  Phenotype is created by genotype. DNA is the genotype, the body that is built by DNA is the phenotype.  Phenotype can extend beyond mere bodyplan and into behavior (bees dancing is a phenotype, human language is a phenotype)

Selection:  Selection is the process whereby certain genotypes are 'favored' to reproduce.  When a genotype is selected for, there will be a statistical rise in its prevalence within a population.  A population is a group of a particular species living in a given geographic location or space.  Selection can be talked about in three ways (I'm oversimplifying this greatly): 

Artificial selection:  Selection by humans for a particular trait(s), the nature of which may or may not directly enhance the genetic interests of the organism but which provides some benefit to humans, thus inadvertently enhancing its reproductive chances.

Natural selection:  Selection without outside intervention for a particular trait(s) which directly enhance the reproductive prospects of the organism by enhancing its ability to compete, survive or mate.

Sexual selection:  Selection within a species by potential mates based upon some (usually) arbitrary characteristic which serves as a means of 'advertising' genetic fitness.  The coloring of birds and/or their songs is one example. The coloring of fish is another example.  The tails of the peacock is the stereotypical example.

Evolution:  Evolution is the differential reproductive success of phenotypes, based upon genotype difference, within a population over generations by a process of genetic reproduction with variation of genes (alleles) leading to a change of frequency of certain genes. 

Speciation:  Speciation is the process by which populations become distinct enough to no longer be capable of interbreeding.  For this to happen, there must be some period of separation of the genetic strains, otherwise crossover will lead to continued mixing of the genotypes.  The usual means by which this occurs is geographic isolation but it is by no means the sole one.


Analoger wrote:

How can anyone say that evolution of the human species is fact.

Having gotten some of the nomenclature out of the way, I'll get into that matter.

Firstly, what does Darwinian theory actually say.  Here is what I believe to be the best axiomatic explanation for evolutionary theory.

1. Evolution:  Organisms change through time.  Both the fossil record of life's hisotry and nature today document and reveal this change.
2. Descent with modification:  Evolution proceeds through the branching of common descent.  As every parent and child knows, offspring are similar to but not exact replicas of their parents, producing the necessary variation that allows adaptation to the ever-changing environment.
3. Gradualism:  All this change is slow, steady and stately.  Given enough time, small changes within a species can accumulate into large changes that create new species; that is macroevolution is the cumulative effect of microevolution.
4. Multiplication:  Evolution does not produce new species; it produces an increasing number of new species.
5. Natural selection:  Evolutionary change is not haphazard and random; it follows a selective process.  Codiscovered by Darwin and the naturalist Alfred Russel Wallace, natural selection operates under five rules:
   1. Populations tend to increase indefinately in a geometric ration: 2, 4, 8, 16, 32, 64, 128, 512, 1024...
   2. In a natural environment, however, population numbers must stabilize at a certain level.  The population cannot increase to infinity--the earth is just not big enough.
   3. Therefore, there must be a "struggle for existence".  Not all of the organisms produced can survive.
   4. There is variation in every species.
   5. Therefore, in the struggle for existence, those individuals with variations that are better adapted to the environment leave behind more offspring than individuals that are less well adapted.  This is known as differential reproductive success.

(From "Why Darwin Matters: The Case Against Intelligent Design" Shermer, Michael 2006)

Now, one of the most immediately apparent things is that none of the above can be denied by a reasonable observer.  It is plainly obvious that there is diversity of organisms and that this diversity has a long provenance in that we find diverse organisms going back deep into the geologic record.  It is also obvious that descent with modification is a real phenomena.  The axiomatic rules of natural selection also are difficult to deny.  Whenever you get a flu or a bacterial infection, you are witness to geometric increase.  If you have had bacterial infections, you will realize that over the course of your lifetime (if you are over the age of, say, 25) stronger antibiotics have been required because they have evolved immunity to older antibiotics.  The older forms of, say, penicillin do not work not because they are old, per se, but because there are precious few strains of, for instance, streptococcus that are still vulnerable to it.  The strains that were vulnerable have all died out, leaving only those that were immune to it.  Because the bacterial defense against antibiotics is a protein based one, the linkage between a change in gene frequency and its affect on phenotype is direct and easily observed because it happens in so short a time.

Slower breeding organisms (like us) will not necessarily show large genotype change (although our phenotype is very different than it was, say, 75,000 years ago) in the lifetime of a single individual because 100 years is simply not a very long time. It appears to be because we can pretty much imagine the lifespan it would take to encompass four or five generations at maximum.  We can sort of grasp the idea that something that happened 500 years ago, really happened.  Beyond about 5,000 years and suddenly time no longer seems 'real'.  No one has an intuitive grasp of just how long 100,000 years really is.  One can accept the idea that it is possible for time to continue over a period of millions of years or you can reject it.  If you reject the idea, however, you need to come up with a reasonable explanation for why there cannot be a period of millions or billions of years.  This is one of the problems people have in understanding evolution is that we are simply not used to thinking in very long time scales.

So let's take a couple of case in points.  This is not meant to recapitulate the evolution of any particular species.  Rather, it is intended to give you an examples of evolutionary thinking.  How evolutionary theory goes about dealing with given problem sets.  I'll start general and get more specific.  Let's take, as an example, an animal that most people are familiar with.  The common house cat.  Because most house cats are still border-line feral we can see much of what their behavior and survival strategies are in the wild because we live with them day-to-day.  If you have ever seen your cat catch a mouse, you will understand exactly what I mean.

Take two cats. They are in the same colony of wild cats and are pretty much otherwise similar.  Let's also say that they are both female (it simplifies the example again).  The cats are pretty much the same except that one of the cats, we'll call it Fast Cat or FC, is just a little faster in running and reflexes than the other cat which we'll call Slow Cat or SC.  The rodent population is variable based upon how good a job the entire population of cats are at hunting, the presence of disease and the food supply.  Let us say that the cats are doing very well indeed and the food supply is somewhat stressed meaning that there are fewer rodents for the entire population of cats.  Because there is a limited food supply being a better hunter will mean that a cat will catch more rodents than other, less adept hunters.  Since speed is a factor in the kind of hunting that cats do, which would you expect to do better in any given period, FC or SC?  Let's say that one of the consequences of doing better at hunting is that the FC-mother will be able to maintain a slightly larger, more healthy litter of kittens.  Say, FC has five kittens four of which survive.  SC has four kittens, three of which survive.  Which do you think will have more descendants, twenty generations down the line?  Over time, what we would expect is that FC would have more descendants who would, in turn, have more descendants until whatever gene caused FC to be the fastest cat around in its day would be the predominant gene, equalizing the benefit through the population.  This change in gene frequency is known as differential reproductive success.  The criteria of speed is a selection pressure.  Speed is the trait being selected for. 

This, of course, leads to the question of why there aren't cats that move at the speed of light (I KNOW it sounds ridiculous but people who know nothing about evolution routinely ask 'if we descended from monkeys why are there still monkeys).  The reason is that it takes energy to build bodies and there is a point of diminishing returns where the energy outlay to build a much faster cat is no longer justified by the returns that would be gained. A house cat that could move at even the speed of sound would never be able to catch enough mice to build a body that was capable of such a feat. 

So, does something like this happen?  Can we observe this kind of thing? As it turns out, genes really do work this way and we can take as an example, the domestic dog.  Humans have created the domestic dog, selecting for certain traits.  Most of the breeds were created by selecting for particular traits.  There may have been other effects as well, but simply by selecting for, for instance, docility we get the Retrievers.  Irish wolfhounds were selected for those traits that helped them hunt wolves, etc.  Now, of course, you will cry 'yes but humans did this and this is what god did' but the example is not about the agent of selection, merely the power of selection itself.  I've already demonstrated that if a given trait were favored in the wild, one would expect it to increase in the population over time and this mechanism requires no intelligent intervention to work.  So, unless there is a compelling reason to believe that changes in genes when selected by an intelligent agent somehow have different kinds of consequences than when there is no such agent or it can be demonstrated that no such selection is possible without an agent, we can certainly say that in potential the above scenarios are plausible.

Cheers
DGG