Black Canadians: outcomes, attitudes, and evidence »« Black Canadians: Making it work

The Watchmaker Analogy: not an argument

The ‘watchmaker analogy‘ has been around for quite some time (about 209ish years by my count), and it was refuted shortly after it’s explication (in fact, Paley was refuted by Hume before Paley was born). Several folk have gone after it, in a variety of ways but the damned thing just keeps showing up. To be fair, it’s not that the argument won’t die, it’s that people ignorant of it’s failure simply won’t stop trotting it out, as if restating it over and over again somehow means that the previous refutations didn’t happen.

Quite recently, Fazale Rana (a member of Reasons to Believe) directed me to his claim that “Kai ABC Proteins Re-invigorate the Watchmaker Argument for God’s Existence” with the invitation to ‘explain how is reasoning is faulty’.

Ask and thou shalt receive.

So this argument suffers from a number of critical flaws, the biggest flaw being a failure to understand why no amount of empirical evidence will support Paley’s Watchmaker Analogy. This stems from a failure to understand how analogies work: analogies are not arguments. Analogies illustrate arguments, and insofar as one only makes an analogy (but fails to sketch out the meat of the argument), then one is failing to make an argument.

But I’m willing to be a little flexible on this point: insofar as a good argument is one that is clear and unambiguous, and insofar as an analogy is less clear than a list of premises followed by a conclusion, then an analogy is a bad argument. Sure, good rhetoric has implicit/hidden premises, but good arguments don’t.

I said above that the Watchmaker Analogy will never demonstrate that design is true, or that belief in design is justified, no matter the evidence. Allow me to provide an analogy to illustrate my point. Imagine, if you will, that we have a painting by a particular artist. The artist has admitted to creating the painting, and people witnessed the creation. The artist has a distinctive style and technique, prefers to use certain unique materials (which are generally not used by other artists). In short, there are a set of characteristics that are associated with this particular artist.

Now suppose that we find a second piece of art. The artist is silent as to whether or not they created this new piece. We start to investigate all the materials and techniques that went into creating this picture, and every characteristic we identify in the second picture, matches a characteristic in the ‘set of characteristics’ mentioned above. Are we justified in concluding that the same artist also created this picture? If not, if we keep accumulating more and more ‘characteristics’, will our conclusion eventually be justified?

Absolutely not.

Because I’m not a theologian, I’ll attempt to make the argument clear:

  1. There exists a painting (P1) known to have been painted by an Artist (A1)
  2. The construction of P1 consisted of certain steps (S1) known to be associated with A1.
  3. If a painting (P2) is constructed according to S1, then P2 was created by A1.
  4. P2 was construced according to S1.
  5. P2 was created by A1.

The flaw in this argument lies in Premise 3. Premise 3 fails to account for any alternative hypotheses, such as the existence of another artist (A2, A3, … An) who also utilises S1. Changing Premise 3 to the more weak “If a painting (P2) is constructed according to S1, then P2 was probably created by A1” doesn’t resolve this issue. Once we arrive at the conclusion that it’s possible the painting was created by either A1 or A2, we now need to compare A1 and A2 (themselves) to see how likely it is that they created the painting.

Paley’s argument is that a designer (A1) is known to have created a watch (P1), and the marks of design (S1) can be found in the watch. By analogy, Paley claims that life (P2) also exhibits these marks (S1), ergo a designer (A1) is responsible for the creation of life. This argument fails because it fails to take into account an alternative explanation, namely that the processes of Evolution (A2) also exhibit S1.

What Fazale is doing with his article is simply increasing the size of S1, the number of steps involved in the creation of [insert object of choice here]. You can make S1 consist of 10 points of similarity, 1000 points, or 1,000,000 points of similarity: so long as those other points are likewise explained by evolution, one is not justified in simply declaring “alright so, they were designed”. Merely shoveling in more data into S1 is irrelevant.

At this point, anyone acting in accordance with intellectual integrity will move their investigation up a notch, to discuss whether or not A1 (god) or A2 (evolution) exists. As there are only self-contradictory definitions of god, and as there is no evidence for god, and as there are no non-question-begging arguments for god, one cannot assert that god (A1) is a viable explanatory mechanism.

As the arguments for god collapses, the argument for theistic design collapses. The argument is fatally flawed not because of a lack of empirical data, but due to the insufficiency of the arguments for god.

Science is based on analogical reasoning

“But wait!” declare the creationists of the world, “all of science is based on analogical reasoning!”

This is the point where my hand starts moving at high speed towards my face, where ostensibly educated adults start pointing at something else and crying out “but… but… but…”. Look, I am 100% in favour of the “Intelligent Design” crowd adopting the methods of science. I really wish they would adopt those methods, so we could finally drop the bullshit that is “Intelligent Design”. It would be a refreshing change from the pseudo-scientific methods they employ, adopting merely the veneer of science/philosophy without employing the actual reasoning necessary to be legitimate. Let’s compare.


  1. See something weird.
  2. Notice that weird thing is analogous to other known/explained things.
  3. Posit that said weird thing has an analogous source to other known/explained things.
  4. Go check if the analogy is correct.


  1. See something weird.
  2. Notice that weird thing is analogous to other known/explained things.
  3. Posit that said weird thing has an analogous source to other known/explained things.
  4. Declare that the analogy is true/real/justified/whatever.

Step 4 is kinda the critical step here, the step that seperates people who are genuinely interested in enquiry (scientists and philosophers), from those who are interested in merely pushing an ideology/dogma (theologians).

Fingerprints and Law

But what about finger-printing? Isn’t finger-printing an argument of analogy? Don’t we send people to jail on the basis of finger-printing?

Finger-printing works on the basis that we all have fairly unique patterns on our fingers. Finger-printing analysis works on the basis that if we find a print, and match a certain number of similar points, then the number of people that could have that same pattern is reduced to the level where it’s probable that only a certain number of people have that print (ideally, the odds should be less than 1 in 7 billion).

So it’s true that finger-printing is an argument of analogy, but (as per usual) ignores the counter-argument: let’s say that someone was found dead in Los Angeles, and the time of death was (let’s say) 7pm, on Thursday the 23rd of February, 2012 (local time). The murder weapon was found, and my finger-prints were all over it. Does this mean that I am the murderer? Am I immediately sent to jail? Absolutely not: this is merely sufficient evidence to investigate whether or not I committed the crime, and once that it’s determined that at the time of the murder I was in Japan teaching an English class, with multiple witnesses corroborating my alibi, the fingerprint evidence is thrown out as irrelevant to the case. In all cases, empirical evidence (and logical inconsistencies) trump analogical arguments.

The Key Point

See, where the Watchmaker Analogy ultimately fails, is that we know as a matter of fact that watches have been made in the past, and that watches don’t spontaneously pop into existence (or self-organise). For watches, this is a settled question. To assert that this is analogous to the origin of life is to illegitimately extend the analogy: we have no evidence to support the claim that ‘physical objects were organised by a non-physical designer’, and that claim is bogus on it’s face. To restate my point: we know that watchmakers exist, ergo it’s legitimate to declare a watchmaker to be the source of a surprise watch. We have zero evidence to support the foolish notion that gods exist, ergo it’s illegitimate to declare ‘a god’ to be the source of a surprise universe.

An argument-by-analogy is, by itself, never evidence that something exists. It is merely an argument in favour of something existing. Once the buried premises are revealed, it becomes clear how much work the rhetoric (rather than the reasoning) is doing the work. The purpose of an argument, in intellectually rigorous fields, is to motivate us to go look for actual evidence: the argument, itself, proves nothing.


(I’d like to believe that I’m done hearing about Paley’s argument from design, at least from Fazale Rana, but… That would be a foolish and vain hope: as outlined here, neither Creationists nor Theologians typically care two whits about the arguments against their position.)




Like this article? Follow me on Twitter!