RSS 2.0 Feed

» Welcome Guest Log In :: Register

  Topic: Vagueness and Explanatory Constraints, Why explanatory constraints are good.< Next Oldest | Next Newest >  

Posts: 319
Joined: May 2002

(Permalink) Posted: Jan. 19 2003,00:48   

I was thinking about posting this on this ISCID thread, and then I thought, "why bother?" Besides I am too busy to start a big debate.

Here is the thread, started by Mike Gene:

Topic: Brainstorming Lessons
link to thread

I quote the end of RBH's post:


If Mike Gene's reference to an "attempt to squelch discussions of design until we first extract a data base of definite information (i.e., actual mechanisms, identity of design, etc.)" means that speculation or questions about the mechanisms of design and the nature, identity, and/or properties of the designing agents ought to be out of bounds, then I think it is he who draws the wrong conclusion from the account of OOL he gives in the OP. I'll give him another example to think about.

Wegener's 1915 hypothesis of continental drift was not accepted for nearly five decades because there was no known mechanism by means of which continents could be propelled across the earth. There were suggestive data that it had occurred - e.g., geographical form matches across seas, distributions of fossils, and so on - but no mechanism, and hence no broad acceptance. It wasn't until the 1960s, when actual drift rates could be measured and a plausible mechanism for drift was offered that it became broadly accepted. (I have stood in the valley at Thingvellir, in Iceland, where the mid-Atlantic Ridge crosses the island, and have seen the lasers that measure the drift rate as the North American Plate drifts west and the Eurasian Plate drifts east. It is an eerie feeling to be there, knowing that.)

That example suggests ID itself ought not squelch questions and speculations about mechanisms and designing agents. No matter how many entries there are in the CCF, absent testable hypotheses about how they came to be IC, it will be merely a marginally interesting list of biological oddities and a set of targets for enterprising graduate students in molecular biology.


I agree that it is ID that is squelching hypotheses, namely the details in origins scenarios that make them testable (strengthenable or weakenable, not always strict true/false).

There is nothing wrong with going out on a limb and proposing hypotheses with specifity that goes beyond the data; this is how science proceeds into the unknown. This is why OOL researchers propose specific hypotheses, test them, and then revise -- e.g. RNAworld has become pretty well supported as a stage preceeding the origin of modern life, but difficulties in prebiotic syntheses of RNA are provoking studies of RNA precursors, e.g NA or PNA "worlds".

The way science does *not* proceed is by maximizing vagueness, e.g. "a designer did something somewhere sometime for unhypothesized reasons by unhypothesized means". With ID, not even the laws of physics are considered legitimate constraints on the hypothesized IDer(s). I would argue that every successful (e.g. archaeology/forensics) or viable (e.g. SETI) "ID-detecting" discipline has hypothesized far more details regarding the IDer(s) than any hypothesis put forward by Mike Gene or anyone else in the ID movement.

The problem with ultravague hypotheses is that they are explanatorily unconstrained; the problem with an unconstrained hypothesis is that there is no objective way to strengthen it or weaken it by consideration of further data.

E.g., with Mike Gene's front-loading via mutational bias idea (leaving aside questions of what the actual biases are, which Art and others will have to work out), it seems to me that front-loading via evolution is approximately the most difficult and clumsy possible way to design something that I can think of. It would be like trying to type with your elbows even though you had fingers. Trying to get to, say, multicellularity through a nonspecific mutational bias would be rather like trying to convert from the Articles of Confederation to the Constitution via a bias in the replacement frequencies of various letters.

Such a conversion could be accomplished either by intelligent or algorithmic selection of specific letters (in the case of biology we should convert this analogy to natural selection's *documented* ability to sweep specific beneficial nucleotide substitutions to fixation in the population, to avoid the usual Dawkins-METHINKS debates) -- but if these capabilities are in play, what's the point of the mutational bias? The mutations will happen slightly slower without the bias (well, assuming that the necessary mutations are those included in the bias, which seems completely unsubstantiated to me), but they will happen sooner or later and then can get selected. (In the case of an IDer, they would presumably not even bother with waiting for the mutations and just design straight-up whatever they wanted to design).

Do these considerations have any weight in weakening Mike Gene's hypothesis? Only if you hypothesize some things about the designer, which Mike Gene does not, because his hypothesis is basically "someone frontloaded something for no specified reason" and thus considerations of efficiency, effectiveness, etc. (even though these are often invoked by Mike Gene and others in support of ID in other situations) will just be brushed aside as "we don't know anything about the IDer".

IMO, this "unconstrainedness" of ID-movement "hypotheses" is their central weakness. This is a problem that supernatural hypotheses have, but is common to "superpowerful but unspecified aliens" "completely unspecified designer(s)", etc., as well. ("Unspecified natural processes" falls in the same boat, BTW) None of them predict or explain anything without further details. Full exhaustive detail is not necessary, but a least enough detail to make us expect some pattern in the data that we wouldn't otherwise expect, and which could be weakened by other patterns, is what it takes to get started.

Vagueness will insulate an idea from refutation but will also doom it to the land of non-explanation.

End of Saturday Night Sermon,

[edit: cross-posted to II evo board:

Vagueness and Explanatory Constraints


Edited by niiicholas on Jan. 19 2003,00:50


Posts: 319
Joined: May 2002

(Permalink) Posted: Mar. 13 2003,14:54   

Charlie D. has expressed the problems with Mike Gene's frontloading in about the most succinct way, ever, I think:;f=13;t=000644

Originally posted by Mike Gene:
The key is always to remember how ID critics view ID proponents - ID proponents are either stupid, ignorant, deluded, or evil. The ID critics argue with me not to understand my views, or even objectively compare views, but because they are looking for ways to rationalize my existence. Someway, somehow, I must be fitted into the precast stereotypes. That's their only interest in these debates (that is, when they are not politicking).

Talk about projecting and sterotyping, Mike! LOL.

Personally, I think your theories have muddled considerably in the past few months. As such, they may seem to you they "explain" more things, but in fact to me they are far less interesting.

Right now, your hypothesized designers seem to me completely schizophrenic. On one side, they seem like totally anal micromanagers: they gave their colonizing bacteria supersophisticated micromachines made up of (altogether) probably hundreds of complex proteins, to do things like syntesizing more proteins, degrading proteins through the ubiquitin system, replicating DNA and correcting replication errors, generating energy, etc etc, even allowing them to swim.

On the other side, the same alien bioengineers were supposedly complete laissez-faire evolutionists, providing bacteria with a minimalist compendium of protein domains (to the point of purposefully avoiding hydrophobic aminoacids!;), and a front-loaded genetic system that would introduce mutations at high frequency in these very simple protein domains to allow them to later develop more complex domains, so that those bacteria could evolve freely in a fully darwinain fashion.

So bacteria were both very simple, and programmed to evolve, and very complex, with all sorts of unevolvable, rigid micromachines. They had very simple protein domains and a system to mutate them at high frequency, and very complex proteins, with fully formed domains, in which changes in domain structure almost invariably kill the original function. Extreme mutability and error correction. On the one side the bioengineers spent countless alien-hours designing a system for bacteria to swim, and on the other their evolvable bacteria evolved probably a half dozen equally efficient motility systems on their own, thanks to their evolvability.

Whenever the evolution of something seems hard to figure out, it's because that something was designed; whenever it seems pretty straightfoward, it's because it was designed to evolve. Basically, your model right now it's not just indistinguishable from evolutionary theory, it's indistinguishable from pretty much anything. Pardon our lack of enthusiasm, if you know what I mean.
The non-teleologist also wants the teleologist to prove the impossible, namely, demonstrate that non-teleological processes couldn't produce some aspect of biotic reality.

Wrong, at least as far as I am concerned. I just want evidence that teleology can generate any aspect of current biotic reality. We have plenty of positive evidence of the existence and power of non-teleologic processes - none of teleology. It's up to the teleologicians to show there's something there, other than smoke and mirrors.


  1 replies since Jan. 19 2003,00:48 < Next Oldest | Next Newest >  


Track this topic Email this topic Print this topic

[ Read the Board Rules ] | [Useful Links] | [Evolving Designs]