The Problem with the ID Argument

No, you’re not wrong at all. I tried to make that clear in my preface to what I teach my students. There isn’t any universal science standards body that defines the terms and methodologies across the sciences. What I said was just a fairly common way for people to think about laws and theories.

My point was 2-fold.

  1. Using physical laws as an example of theories without mechanisms is probably not the best example.
  2. That if ID folks want to put forward a scientific theory they should be ready to give a mechanism and not just an observation or assertion. Scientists will expect both a description of what happened, but also how it it happened.

If the scientists are open minded they won’t expect unreasonable detail for the mechanism or for you to even necessarily defend it, that can come later. What they need is something to differentiate between other proposed theories.

What scientists do really well is take competing models and break them down into component parts and compare them to the data to see which of the models is closest to reality. If I could summarize what science is that’s pretty close, in my opinion.

So, when people put out a theory without mechanism or anything to “grab on to” with their tools, they can’t do science and eventually they will say that the theory isn’t science.

1 Like

No Bill it doesn’t ASSUME a single tree. A single tree is an empirical observation for which the most parsimonious explanation is common ancestry. You keep getting that wrong no matter how many times it is explained to you.


No, I am claiming both have purposefully arranged parts.

You just can’t demonstrate it with biological life. You’re just guessing again.

In what case would the model deliver a tree with more than one starting point?

Bill we’re talking about empirically observed reality here and now. The genetic data we have now shows a tree going back to a single starting point. Deal with reality, not your fantasized ID world.


It won’t. The model doesn’t deliver trees. It evaluates trees. We pick the tree that best fits the data. If it fits the data uniquely, much better than other trees, that’s evidence for common descent. You could (and this is occasionally done) evaluate two or more unconnected trees for the same taxa, using the same model. That’s what Theobald 2010 did. Do you remember?


Yes, and his analysis uses random change as the null. This is the point that @Paul_Nelson has been making.

What you are observing is not random but you are claiming a mechanism that is random with respect to changing sequences.

The two non random mechanisms being discussed are reproduction and mind.

I think that science should be first defined by its purpose before being defined by some methodology. In that sense, science is above all a quest for truth about the natural world.
It follows that the following questions are legitimate scientific questions, right?

  • is an intelligent cause responsable for some aspects of the natural world?
  • are there reliable signs of intelligence in the natural world?
    ID theorists argue that they have found such signs of intelligence. Now, would it be wrong to say that ID theory is precisely about « signs of intelligence »?

Actually, I will respectfully disagree, for two reasons:

  1. That gives science too much power. I firmly believe there are truths about the natural world that are found outside of science. Is nature beautiful? How did the universe get here? Why is the universe so consistent and uniform in it’s laws?
  2. Science is fundamentally a human endeavor and those who do science care a lot more about how science is done. If your hypothesis isn’t testable or falsifiable, if you don’t engage with empirical data or make predictions, if you don’t participate in peer-review, it’s just not science.

Neither Darwin nor Newton participated in peer-review.

That is categorically false. Their peers reviewed their work very closely, and their books were targeted at their peers, not the public.

1 Like

It doesn’t work that way.

The maser was invented around 1953, by people doing research on microwave technology.

I am connected to the Internet with optical fiber, where the signal is sent through the fiber with a laser driver. The laser is a stepchild of the maser.

Do you really want to say that the purpose of that 1953 microwave research was to bring me a high speed Internet connection?

Not sure that the publication of « the Origin of Species » went through such a different process than « the Design Inference », « Introduction to Evolutionary Informatics », «Darwin Devolves » or « Darwin’s Doubt », to name a few.

Sorry, but that doesn’t fit what the paper actually did. There is in fact no null hypothesis in that paper, just various hypotheses tested against each other. What it assumes is that the likelihood of two unconnected trees is the product of the likelihoods of each tree. How would you do it?

As you should know by now, mind is not a mechanism. Mind makes no predictions. There is no model of what mind might or might not do.

1 Like

What is the mechanism used here to test the likelihood of a tree?

How has he independently determined that mechanism was responsible for the pattern?

My guess is that a mind generating a eukaryotic cell has a higher likelihood of success if it does not have to start from a prokaryotic cell.

@Giltil, I can be sure Galileo and Kepler went through a different process than any book published today. Science and publishing worked very different at its inception. Galileo and Kepler, moreover, were not writing books for the public, but for other scholars. All the ID books I know of are targeted at the public, not other scholars.


Mechanisms aren’t used to test likelihood. Models are used to test likelihood. The assumptions of the model are easy to find if you just read. Here: " In these models the
primary assumptions are: (1) that sequences change over time by a gradual, time-reversible Markovian process of residue substitution, described by a 20320 instantaneous rate matrix defined by certain amino acid equilibrium frequencies and a symmetric matrix of amino acid exchangeabilities; (2) that new genetically related genes are generated by duplication during bifurcating speciation or gene duplication events; and (3) that residue substitutions are uncorrelated along different lineages and at different sites."

Wrong question. What do you think is wrong with the model?

On what is this guess based? How does a mind generate a cell?

1 Like

What is the chance based on his calculations of the prokaryotic to eukaryotic transition happening at all given the method of change he is using? Just for starters he has to insert hundreds of introns and make the splicing machinery to remove them.

If this number does not make sense then his paper does not. Certainly the probability is better if you start with some similar proteins but so what.

This was clearly not the case for Dembsky’s Book « The Design Inference ». Moreover, this seems a strange statement for someone who has written a review of « Darwin Devolves » in the journal Science!

The truth is that most ID books are targeted both to the general public and to scholars.