a list of lists

9 errors in talk.origins' responses to creationism

6 September 2005
At talk.origins there is a list of creationist claims with links to accompanying responses. talk.origins's responses attempt to refute the creationist claims, but there are errors in some of the refutations.
Each item below starts with a link to the section of the talk.origins site that responds to a particular creationist claim.

The problems that I have discovered in these responses fall into various categories:

  • Arguing a point which is not necessary to make the case against the creationist claim.
  • Omitting an argument that may be necessary to disprove the creationist claim, in effect conceding that the claim holds if the arguments that they have provided do not stand up.
  • Failing to clarify the meaning of a creationist claim before attempting to refute it.
  • Failing to refute the strongest possible version of the creationist claim.
1. Survival of the fittest implies might makes right. The talk.origins response talks about the naturalistic fallacy (confusing what is with what ought to be), and about cooperation and treating other people well.

None of which really refutes the claim. Probably because the claim isn't wrong. That is, might is right.

How can this be?

Let's do an evolutionary analysis of morality. Morality is a fairly important and distinctive aspect of human behaviour. Very likely, there are morality genes, i.e. genes which cause moral behaviours, and which have been selected for because they cause moral behaviours. These morality genes have spread throughout the human population by contributing to the reproductive success of the people who have them. Directly, or indirectly, the morality genes have exterminated the alternative genes ("immorality" genes?), and this necessarily implies the direct or indirect extermination of immoral people by moral people. (When I say "direct", think of warfare and killing, and when I say "indirect", think of hogging all the resources and not letting your opponent have any.)

This seems a bit contradictory, because morality is supposed to be about being nice to people and not hurting them. Well, whatever it is that morality is supposed to be about, the evolutionary analysis tells us that it is really about something else. (In fact, we will see that it is about being nice to some people, but not about being nice to all people.)

This is certainly different to the religious view of morality. The religious view, and in particular the Christian view, is that morality is a set of rules provided by God, and applied in judgement by God, and the only issue for the people is to decide whether or not to obey the rules. In the scientific approach to understanding morality, there is no God. Certainly morality does consist of rules (some of which are about not killing, and some of which are about being nice). But where do the rules come from? If morality evolved, there are two main possiblities:

  • The genes for morality encode the rules.
  • The genes for morality encode the process for deciding what the rules are.

The correct answer to the question is probably a mixture of both, however a variety of moral principles have been observed across different societies and across different historical eras, and this suggests that the second option contributes more than the first one.

When we attribute all aspects of morality to human nature, and not just the decision to follow the rules or not, then we must consider a whole long list of moral activities:

  • Inventing and advocating new moral rules, or advocating changes to existing moral rules
  • Deciding whether to accept new or changed moral rules advocated by others
  • Judging the behaviours of others according to moral rules
  • Discussing these judgements with others
  • Judging one's own behaviour according to moral rules
  • Deciding whether or not to act in accordance with said rules
  • Deciding an appropriate response to the moral infractions of others

Each of these activities is an essential part of the moral process, and each must make some contribution to the reproductive success of the individual if the genes that cause the activities to occur are to be selected for.

The bottom line has to be that those who engage in these moral activities must succeed reproductively, at the expense of those who don't. This is where the "might is right" comes into it: the minority who attempt to ignore the moral rules of a society will be collectively punished by the majority who defend the moral code, and the punishment will result in reduced long-term reproductive success for the punished. The "might" comes from the moral consensus of the majority, and it is the "might" that determines the "right".

This picture suggests an extreme moral "relativism", since it seems that the individual has to go along with the moral code defined by the society that they live in. However, there are many different societies, which compete against each other, and there are societies within societies, which also compete against each other, although to a lesser extent. Moral systems can vary from one society to another, and they can vary between different groups within a single society. Those societies with moral codes which enable individuals within those societies to function more effectively will be stronger when the time comes to compete against other societies. For example, a society that prohibits random killing between individuals will be stronger, and because it is stronger, it will be better able to wage war against another society when the need (or the occasion) arises. A society that enforces fairness in commercial dealings, and prohibits stealing, will tend to become richer, and this will also help it to wage war, when the need arises. Thus competition between societies will be a strong determinant of which moral rules survive in the long run.

According to many religions, killing is very wrong, but in practice the religious establishment often finds a way to justify a war and the plundering that goes with it. If the other side belongs to the "wrong" religion, then this may be sufficiently immoral to justify whatever action is necessary to prevent the practice of that religion by its followers, which might just happen to include invading their country and stealing all their gold.

There are those of us who do not feel inclined to go round attacking and invading other countries, and many of us live in countries whose leaders profess a desire for peace. Much of the aggression between countries is caused by the nasty people – Nazis, Communists, or whatever – who happen to be in charge of other countries. It seems that if we could get rid of those nasty guys once and for all, then the problems of war would go away. The only thing is, there has never been a world that was only ruled by "nice" guys, and it is pure speculation to believe that having only "nice" people in charge would be a guarantee of peace in the long term.

2. Evolution is atheistic. Well, yes and no. The talk.origins response emphasises the existence of religious people (even Christians) who believe in God and in evolution.

Unfortunately, the co-existence between religion and science is not so relaxed as some people want it to be:

  • Most religions contain a belief in some magical aspect of reality. The most significant magical aspect is the human soul, a hypothetical non-material entity which is responsible for all or part of what the human mind does, and which is usually supposed to survive death.
  • Modern science has been slowly but surely removing the need for "magic" from our understanding of the universe. The theory of evolution is a major agent in this anti-magic campaign:
    • It is no longer necessary to invoke a magical process to explain the origin of species. (It might still be necessary to invoke some kind of "miracle" to explain the origin of the very first species, but that is a separate issue.)
    • Normal physical causality is found sufficient to explain the existence of purposeful behaviour in living organisms.
    • Evolution establishes a continuous lineage between very primitive single-celled organisms and very complex human beings. Bacteria and other simple organisms are much more obviously non-magical; the argument by continuity suggests that all other organisms are non-magical as well (and yes, that does include you and me).
Why is talk.origins so keen to accommodate the religious beliefs of those who disagree with creationism? Is the argument against creationism a search for truth, or is it a search for popularity?

It is perhaps OK to document the existence of religious believers who also believe in the theory of evolution by natural selection, but one should be wary of implicitly claiming that those same religious beliefs are consistent with the theory.

3. If man arose by chance, life would have no purpose or meaning. talk.origins responds to this by saying that (1) "Purpose can come from anyone" and (2)"Purpose is not determined by origins". Unfortunately, given a technical definition of purpose as "reverse" causality (where the consequences of an action "cause" said action), this is (1) wrong and (2) wrong! If our origin is that we evolved by natural selection, then the purpose of life has to be long-term reproductive success. There is a secondary question of psychological purpose, which is a function of internal selective processes occurring within the brain, where these processes are themselves the result of evolution by natural selection. But in either case, purpose is only ever scientifically explicable if it arises from a selection process, and the nature of the selection process must determine the nature of the purpose that it creates.

(The talk.origins response also mentions that evolution does not imply that life arose "by chance", which I do not disagree with, although it might be more accurate to point out that random chance is part of the evolutionary explanation, and it is natural selection which is the non-random part of the explanation.)

4. The odds of life forming are incredibly small. The talk.origins response is to point out that the creationists always calculate their odds by making overly-specific assumptions about how life originally formed.

Now there is no hard evidence that the probability of life forming is very small, but there is also no hard evidence that the probability of life forming is not very small. Life has never been observed to arise more than once, and there is no known bound on the number of occasions on which life might have had the opportunity to arise (and I think there is no possibility of ever determining such a bound, since there is no guarantee that we can ever observe all of physical reality).

We might feel that the origin of life is only scientifically explicable if the odds of its formation is not "too" small. Applying a scientific optimism (like Einstein's assertion that the universe is comprehensible), we might conclude that there has to be an explanation involving processes that are not unreasonably improbable.

However, the smallness of the probability of life forming by purely natural processes is not by itself a proof that life did not form by purely natural processes, because, with a sufficient number of opportunities, even the most unlikely event will probably occur. This type of reasoning leads to the weak anthropic principle, which states that an observation of anything necessary to our own existence cannot by itself be treated as an observation of an impossibly unlikely event. (There are other statements of the anthropic principle, but this is the form most relevant for the current discussion.)

One can further speculate as to what physical mechanisms might be responsible for the existence of a number of opportunities large enough to account for the occurrence of highly improbable events, but such speculations are secondary to the argument based on the anthropic principle.

Because the anthropic principle is a principle of reduced falsifiability, and because no situation has yet been found where it has proven necessary to invoke the anthropic principle to explain something, (for example, considering the origin of life, the odds of life forming might not be so small), it is seen as somewhat controversial, and perhaps "unscientific". This might be why talk.origins leaves it out of the response to this claim (even though it is discussed elsewhere on their website). But the anthropic principle is based on a straightforward statistical theory of conditional probability, and it should not be left out of any discussion of the probability of the origin of life.

Addendum: The last item in the talk.origins list for this claim says:,

"The calculation of odds ignores the fact that innumerable trials would have been occurring simultaneously."
This comes close to the anthropic principle argument, however the "would have been occurring simultaneously" seems to implicitly refer only to trials occurring on Earth, and the number of such trials is not large enough to overcome the potentially enormous odds that might be required to originate life. There needs to be more detail about where the "innumerable trials" would or could have been happening.

5. Why is new life not still being generated today? The response to this claim leaves out one obvious possibility (perhaps because it has already been "refuted", for which see the previous item): that life is not being generated today because the odds of life forming are incredibly small.
6. Most mutations are harmful. talk.origins deals with this claim by talking up the rate of beneficial mutations and talking down the rate of harmful mutations. However, such argument is not strictly necessary to refute the claim, because, for any given ratio of harmful to beneficial mutations, there exists some non-zero overall rate of mutation which results in the long-term accumulation of beneficial mutations. Exactly how many mutations are beneficial is of interest, but as long as there are some beneficial mutations, and as long as the overall rate is low enough to prevent harmful mutations from accumulating faster than natural selection can remove them, evolution in a positive direction will occur.
7. The genetic code is a language. talk.origins starts its response to this claim by attempting to prove that the genetic code isn't really a language. However, although the genetic code might differ in practical details from natural human language, it is very much a language in a more general sense of the word, and it is pointless trying to prove that it is not a language. The only necessary part of the response is the third item, which points out that all real languages have a material implementation. (One could perhaps add that this creationist claim is just silly philosophising – language is called "non-material" because it is an abstract concept, and this is equated with "non-material" in the sense of being supernatural.)
8. The speed of light has changed. Before anyone attempts to refute this claim, they should ask what it means. Specifically, "changed" compared to what? The speed of light is not an absolute number. Any claim about the speed of light changing must explicitly state what speed it is being compared to. For example: "the ratio of the speed of light to the top speed of a Rolls Royce Silver Phantom is changing over time." Such a claim can just as easily be regarded as a claim that the other speed is changing, i.e. that the top-speed of a Rolls Royce Silver Phantom is changing over time, as compared to the speed of light.

As luck would have it, the current SI units for distance and time actually define the speed of light to be a constant. So, measured in SI units, the speed of light can't change. It's also worth noting that creationists are not the only ones guilty of making potentially meaningless statements about changes in the speed of light. For example, see Speed of light may have changed recently at New Scientist, and read the fine print, where you learn that it's not actually the speed of light which is being measured, it's the fine structure constant (which is a number).

9. The second law of thermodynamics prohibits evolution. For this popular creationist claim, and various sub-claims, the following basic questions are not answered by the talk.origins response:
  • Does the entropy of the genome decrease when it evolves?
  • If so, where does the required "negative entropy" come from?
The precise moment of evolution consists of a "good" mutation, but mutations can in principle be caused by thermodynamically reversible processes. A full explanation of the thermodynamics of evolution is that evolution by natural selection is a natural example of Maxwell's demon. (Follow the link to a full article by myself on this topic. This is a case where anti-creationists haven't properly refuted the creationist argument, because the creationists themselves have failed to state the claim as clearly and strongly as it could be stated.)
Note: I have compiled this list against the "index to creationist claims" on the talk.origins site. There is a lot of other material on that site, some of it derived from postings to the talk.origins news group, and some of that material expresses alternative opinions on some of the issues I have mentioned. For example, there are discussions about the anthropic principle, fine-tuning and multiple universes.

However, the list of responses in the "Index to Creationist Claims" pages do appear to be given as an official best answer to the creationist claims in each case, so I think it is reasonable to criticise any responses from the list which are not the best possible ones that could (or should) be given.

Update (24 Sept 2005): I have dropped the 10th item from this list, as I notice that the 5th item in the list for The Anthropic Principle does mention the possibility of multiple universes.

p.s. if you want to read about an evolutionary theory of music, read my book What is Music? Solving a Scientific Mystery.