rafe gutman

you know those puzzles where you are given a word and asked to change it into another word by making one letter substitutions, with each "intermediate" being an actual word in the dictionary? one website calls them "doublets", but other names include "Word Ladders, Ladderwords, Stepwords, Word Chains, Laddergrams, Transitions, Transformations, etc." here's an example:

one of the latest backpeddles in the ID debate is the concept of indirect darwinian pathways. initially, behe and company said that IC systems could not evolve. then they said that they could evolve, but only through an indirect pathway. they implied that this pathway was just the simultaneous assembly of all the parts in a single step, suggesting that as the number of parts increased, so did the difficulty in doing this. in fact, when dembski calculated the probability for the evolution of the flagella, he assumed that all the parts would have to come together at once, which is ridiculous. when pressed, dembski finally admitted that subsets of IC systems could be functional in their own right, meaning that an indirect pathway did not require simultaneous assembly of multiple parts. however, in defense of IC, dembski said that no examples of IC systems coming together in such a fashion have been demonstrated (to his satisfaction).

a lot of internet IDists like to equivocate between definitions of direct and indirect. i think a lot of the time, people tend to think that indirect means simultaneous assembly. for example, check out this quote from dembski:

(i should mention that dembski also tries to deny the existence of functional intermediates here, but he doesn't in other places, go figure). anyway, my point is that a lot of people are unclear as to what an indirect pathway is. an indirect pathway involves a change-in-function (though not all of the time). an analogy is the aforementioned word doublet. each of the intermediates from MORE to LESS is a real word (i.e. functional). notice that none of the words are synonyms to LESS (maybe "loss" is close, but you get the idea). so the word LESS may be irreducibly complex in the sense that if you change any of the letters to something else, the original meaning is lost. if the individual letters represent proteins, and the words are complexes, then its easy to see how the gradual accumulation of parts, each with changes in function, can produce an IC system. since the example above does not involve the addition of parts, here's another where each step involves either the addition or change of a single letter, with each intermediate being a real word:

I
IS
HIS
HISS
MISS
MESS
LESS

using a strategy like this, it's not hard to imagine how the formation of very complex words, like ANTIDISESTABLISHMENTARIANISM could occur, entirely through functional intermediates. the same could be said for the flagella, if we had the right dictionary.


for more info on word doublets, check out this page:
http://thinks.com/puzzles/doublets.htm


so does this analogy work for you?

Doubting Didymus

EDITED: That'll teach me to rush out the door without properly reading the opening post! You are of course, completely right, and it's so simple!

Who was it said "How absolutely stupid not to have thought of that!" Was it Wallace? Huxley? No matter.

Secular Pinoy

Huxley, of course. Wallace needn't have said that, since he independently discovered the idea of Natural Selection, and that both Wallace and Darwin had their papers presented at the same time to the scientific peerage, thus establishing co-priority.

pz

Yes, except that you killed it with your last example. Can you think of a valid English word that is only one letter different from ANTIDISESTABLISHMENTARIANISM? I think that's the heart of the IC argument, that if you start with the existing complex 'words', you can't step backwards to find a valid intermediate.

Of course, the solution is that your analogy has broken down, not that gene sequences have this problem. Your last example also illustrates that there are other forces than single 'nucleotide' changes that can occur -- we can swap in whole suffixes and prefixes, rather like the modules and motifs that get juggled around in protein evolution.

RBH

pz wrote Not to mention that once there is an inventory of words, recombination can produce what look like saltational jumps (referenced to a point-mutation fitness landscape). Of course, they aren't really saltational, since on the fitness landscape induced by the recombination operator they are one-step increments.

RBH

rafe gutman

i was thinking of forming the root words independently, then recombining them into the final sequence. i guess i should have specified that.

if antidisestablishmentarianism was the flagella, i would say that we can observe roots like ANTI- and DIS- as parts of other protein complexes. the type III secretory system would be ESTABLISHMENT. okay, maybe i'm pushing the analogy too far.

anyway, the point of all this is to try and come up with good analogies to explain these concepts to the layperson. if anybody else has some good analogies, feel free to post them here.

Nickle

From the Dembski quote:
I know I may be picking a nit here, but I dislike the wording Dembski uses. He acts like evolution has to go through a drudging indirect process to achieve its final goal. Evolution has no goals, no foresight. This whole talk of direct and indirect makes no sense to me.

Maybe a decent analogy is a person meandering through a field chasing a butterfly. A person observing her may say she is taking a needless indirect path to reach the end of the field when, in fact, she is just chasing the butterfly right in front of her. She may end up at the end of the field, but that wasn't her original goal.

I just realized I added an analogy to a thread about an analogy. Sorry guys.

Principia

Neat topic! I've always wondered to what extent etymology can be explained with evolutionary concepts. In fact, the language people have their own version of a monophyletic tree:
from here:,


and from here:
It looks like the phoneme is the unit of "genetic" change in linguistics, and the expressed meaning of a string of phonetic sounds is analogous to a phenotype, best understood in some local framework (syntax, grammar, etc.). So, the next logical question is whether or not linguistics is best understood in a teleological paradigm. In other words, were modern languages designed ab initio and then modified with a particular purpose in mind? Or perhaps, there is an IC core of words front-loaded so that all modern languages are evolved.

Nic Tamzek

I agree that antidis(whatever) is problematic unless we allow (which we should given biology) it to be broken into functional subpieces. This would get you most of the way there at least. 'establishment' would be tough, but:

establish + men + t
or:
establish + meant
subtract a:
establishment

I think you have to allow for somewhat-functional-but-suboptimal intermediates like this in order to make language and DNA have more "equivalent" flexibility.

establish:

I can get "tab" but other than that it's tough unless we start allowing other languages. It would be nice if we had some kind of "flexibility index" to compare language and DNA...

Of course DNA also has only 4 letters, codon degeneracy, etc.

rafe gutman

okay, here are the rules. aside from single letter substitutions, you can splice in bits of letters from other words provided that they are whole roots, like "anti-" or "dis", but not "abl" or "hes" or something like that. however, you must splice from an actual word. for example, you cannot generate "anti" separately (A - AN - ANT - ANTI) then attach it onto a word, since "anti" itself is not a word. you could splice it from ANTITHESIS, but you'd have to account for its generation as well (perhaps ANTIC would be a better choice). lastly, each round you can only perform one step, either the recombination of two words or the change/addition/subtraction of a single letter.

so with these rules in mind, can you generate the word ANTIDISESTABLISHMENTARIANISM with no unselectable steps?

(maybe we should start with something easier, like INTELLIGENT DESIGN, or IRREDUCIBLE COMPLEXITY)