Perry Marshall: What is Random?

(Perry Marshall) #65

Evolutionary mutations are the ones you said were random and spontaneous, when you referred to the conversation between the man and Shapiro. (As described in my book and most recent blog post.) Joshua, Please define randomness.

I will post future replies on my blog.

(S. Joshua Swamidass) #66

@Perry_Marshall if you want a private conversation we can do that. If you want a public one, why blog posts? The exchange is much more difficult this way. If you can just devote a couple hours to an exchange I’m sure we can get on the same page. I can even make a thread for just the two of us if you don’t want to answer to others.

I might eventually get around to answering blog posts on another site, but they are a low priority now. There is a queue, @Agauger is ahead of you, and I’m not even sure the right way to respond to her. Also you are not answering questions directed your way. You are setting this up in a very adversarial, which is not necessary. Can you explain why?

So let me know how you want to proceed.

(S. Joshua Swamidass) #68

8 posts were split to a new topic: Mark: Are Mutations Random?

(S. Joshua Swamidass) #69

So, @Perry_Marshall, I cleared out this thread to focus on your questions. Waiting for you to map a way forward.

(Perry Marshall) #70

The way forward is for you to define your use of the term “random”. This is not an unreasonable request.

(S. Joshua Swamidass) #71

Are you going to participate in the conversation here or not @Perry_Marshall? Are you going to answer questions to or not?

If the answer is “no” and “no,” we will have to wait for another time.

(Perry Marshall) #72


(S. Joshua Swamidass) #73

@Perry_Marshall I find that post to be unnecessarily aggressive. It appears you are trying to create controversy where none need exist. When you are ready to communicate with me directly, let’s talk. I’m not willing to engage in a blog post war with you.


If you purposefully shuffle a deck a cards the result is still a random order of cards. That seems to be a concept that Shapiro and others fail to understand.

(Neil Rickert) #75

I’ll try a more detailed reply to @Perry_Marshall here.

I did look at your blog post, and my reaction is similar to that of @swamidass, but lets avoid repeating that and get to the question of randomness. In ordinary life, “random” is used with a range of different meanings. For biology, particularly evolutionary theory, the main concern is that mutations are random with respect to fitness. And that mostly means that mutations do not seem to be purposely oriented to improve fitness. If we found that mutations were usually beneficial (improved fitness), that would pose a problem for evolutionary theory, even if the mutations satisfied mathematical tests of randomness. So the use of the term “random” in evolutionary theory is mostly a way of saying that mutations do not appear to have any bias toward improved fitness.

If there is a locus on the genome where mutations are almost always fatal or seriously detrimental, then evolutionary theory already predicts that the mutation rate should be lower at such a locus. So variation of mutation rate is not consider non-random.

The traditional account of evolution says that mutations are copying errors. And it looks to me as if you are really arguing about that view of mutations. But random need not imply error. If I’m making a cake, I will stir up the cake batter. Stirring it up is a form of randomization, but it isn’t accidental or an error. Random is not the same as erroneous or accidental. I’m suspecting that you (and Shapiro and others) are making a mistake by arguing against randomness of mutations. If you were instead arguing that mutations are not accidents, but are instead part of how the system works, then I would probably be agreeing with you.

(S. Joshua Swamidass) #76

Except mutations are biased in some important ways towards beneficial mutations.

(S. Joshua Swamidass) #77

5 posts were split to a new topic: Mutation is Biased Towards Fitness

(Perry Marshall) #78

Neil, T_aquaticus and Joshua, thank you for your responses. Here’s my reply:

(Neil Rickert) #79

I’m not interested in cross-site debating.

(S. Joshua Swamidass) #80

Even worse yet, he is starting to make an Information Theory argument against evolution. Sadly, he hasn’t even read the answers we’ve given on this, or he has read them and wants to claim his questions were unanswered none-the-less. Poor form. It is probably for the best he won’t engage here right now.

Maybe he will reorient and try again someday. I hope so.

(S. Joshua Swamidass) #81

8 posts were split to a new topic: A Shuffled Deck of Cards is Random?


I would be happy to discuss your ideas with you on this site if you were willing.


An interesting paper to consider is the link between mutation rates and gene regulation:

In this paper they looked at the reversion of a faulty gene responsible for making leucine (leuB). In other words, they measured the rate of mutations that fixed the gene.

As part of the experiment they put bacteria on media that lacked certain amino acids, such as plates that lacked leucine. In these conditions the gene for making leucine is derepressed which is an awkward way of saying that the protein blocking the transcription of the gene is removed allowing the gene to be expressed.

What did they see in these experiments? The number of leuB- revertants increased on plates that lacked leucine compared to plates that had leucine which means that beneficial mutations in leuB- increased in environments where those mutations are beneficial.

What is the mechanism that is causing this? As it turns out, genes that are being actively expressed have higher mutation rates. All genes that are being expressed see an increase in mutation rate, not simply the leuB gene. For certain genes this may very well increase survival. However, you will also see an increase in deleterious mutations within genes that are functioning just fine and are being actively expressed. This is a case of an increase in random mutations (as the term is used by biologists) within actively expressed genes which produce beneficial, neutral, and deleterious mutations.


What do you mean an information theory argument AGAINST evolution? First of all, can you link to your arguments against this? I’m interested in reading them. Are you arguing that DNA is not a code?

Secondly, how is he arguing against evolution? He agrees with ID that the math against Neo-Darwinian evolution has some force but argues this means that evolution occured by a different means, not that it didn’t occur.

(Perry Marshall) #85

As Mark points out, I am not making an information theory argument evolution. I’m using an information theory argument that evolution is not random, it’s stochastic (and ergodic). Random and stochastic are far from the same thing in engineering:

Joshua please define what you mean by randomness.