Monday, February 23, 2009

Chance and Necessity

AUTHOR: Allen MacNeill

SOURCE: Original essay

COMMENTARY: That's up to you...

One of the main disagreements between evolutionary biologists and "intelligent design" supporters is the role of "chance" (also called "randomness") in the origin of biological objects and processes (especially adaptations). In many cases, it seems that this disagreement is exacerbated by a disagreement about what the antonym of "chance" might be. In my experience, many people on both sides of the EB/ID disagreement think that the opposite of a "chance" event is an event that was "caused". However, this ignores the fact that many scientific explanations now include "chance" or "random" processes in causal explanations of natural objects and processes.

In this context, therefore, I propose that the antonym for "chance" is "necessity", as was first pointed out by Democritus of Abdera in the 4th century BC, whose two famous aphorisms were:

"All things are the fruit of chance and necessity"


"Nothing exists except atoms and the void".

The phrase "chance and necessity" is often used as a descriptive term in modern science. It means essentially "all natural/physical causes", or what one might consider to be the "Newtonian" world view. The reason that evolutionary biologists consider that Darwin founded the science of biology is that he proposed a theory of descent with modification and the origin of adaptations that was based entirely on "chance and necessity", thereby "unifying" biology with the other natural sciences.

However, it is clear that there is a myriad of objects and processes in the universe that are not entirely the "fruit of chance and necessity", just as there are clearly "things" that are neither "atoms" or "void". In purely physical terms, we now recognize at least three "things" in nature: mass, energy, and information. Only the first of these three qualifies under Democritus second aphorism, as neither energy nor information qualify as "things". Energy, of course, is interconvertable with matter according to Einstein's famous energy/matter equation. However, in the form of "pure" energy (such as electromagnetic radiation), energy is not detectable unless and until it interacts with matter (this is why outer space appears black, even though it is filled with light).

The "detectability" problem of energy is even more difficult in the case of information. It seems clear that all forms of information involve some sort of "translation", in which matter/energy relationships are "translated" into information, which can be stored and transmitted in forms that are not entirely reducible to the original matter/energy forms which they represent. As Korbzybski famously said, "the map is not the territory"; the representation (in the form of information) is not the "thing" represented".

No one, including hard-core "naturalists", suggests that information doesn't exist. The problem (and this is where EB and ID run into serious difficulty), is how (and perhaps by whom) information can become "translated" in the first place. It is not even completely clear that simple "natural" processes (such as the photoelectric effect) do not include an exchange of "information" as well as an exchange of energy (in the form of a photon, for example). In quantum electrodynamics, does the exchange of a photon (or, even worse, a virtual photon) constitute an exchange of "information" between the interacting particles?

In classical physics, there is no "arrow of time". Newtonian mechanics can be run forward or backward in time, with no contradictions (and no way to tell which way the process is happening). However, in those branches of modern physics in which random processes play a part (statistical mechanics, thermodynamics, and quantum mechanics), randomness is necessarily tied to "time's arrow". The same is apparently the case with information, at least in its Shannon form.

Ergo, it seems to me that the problem of information is one that is necessarily tied with the concept of "chance". Indeed, I have come to think that information is a manifestation of one of the operations of "chance" in nature; without information, chance overwhelms everything and the universe disintegrates into permanent incoherence.

This is clearly the case in biology. The "fixing" of information in the physical form of the genetic material is the only thing that makes biology possible. Without such "physicallization" of information (in the form of DNA, RNA, proteins, etc.) biological systems would be impossible, as they would disintegrate into incoherence.

Ergo, the transition from purely physical (i.e. mechanical/Newtonian) processes that do not include the translation and transfer of "translated" information to biological processes that necessarily involve the translation and transfer of "translated" information is the central problem of both biology and the physical sciences. As I have written before, I am not sanguine about our ability to answer this problem using historical information, as the transition occurred at a time (and perhaps in a place) which has left no traces from which we can infer its dynamics.

This leaves us with theoretical models, which are of course based on metaphysical assumptions about reality. I believe it would be fair to say that IDers assume that information can exist without a physical referrent (i.e. something that "carries" or "transmits" the information), whereas EBers (along with most other natural scientists) assume that information must have a physical referrent (i.e. it cannot exist in purely "disembodied" form).

Furthermore, it seems clear from previous discussions in this forum that IDers assume that information can be "foresighted"; that is, it can somehow anticipate future outcomes, not by "induction" from the past but by some kind of "deduction" from the future. EBers (again, like most other natural scientists) assume that "time's arrow" cannot point backward, and that the future is therefore relentlessly driven by the past.

It seems to me that the foregoing lays out the problems which which any scientific theory of "origins" must come to grips:

• whether information can exist in purely "disembodied" form in nature, without a physical referrent

• whether the origin of the "translation" of information into physical form (i.e. the origin of the genetic translation machinery of living organisms) cannot take place without an input of "disembodied" information

• whether any form of information transfer can be genuinely "foresighted" (i.e. can be modified by events that have not yet happened, rather than simply predicting future events based on events that have happened in the past).

Evolutionary biologists (and the vast majority of all natural scientists) begin with a metaphysical world view in which their starting assumptions answer these three questions with NO. By contrast, most of the ID supporters with whom I have had such discussions begin with a metaphysical world view in which their starting assumptions are exactly the opposite: they answer these three questions with YES.

Personally, I believe that the metaphysical world view of most scientists is easier to work with, as it requires fewer "YES" answers to these questions (i.e. it requires fewer unverifiable assumptions about things that must exist for the universe to work). However, I freely admit that this belief is on a par with my acceptance of "Occam's razor" as a basic principle of scientific explanation. That is, "Occam's razor" clearly isn't "true", it's just useful as a rule of thumb in doing science.

So, where does this leave us? I think it explains why most scientists are uncomfortable with the word "design" being applied to biological objects and processes. As I have pointed out, Ernst Mayr argued for the legitimacy of the concept of design in biology, when what was meant by that term was the idea that organisms are "designed" by the information encoded in their genomes, interacting with the information obtained from interactions with their environments. This is because this view of biological "design" conforms to the three answers to the three questions listed above as answered by most scientists.

However, Mayr (and virtually all other evolutionary biologists) was uncomfortable with the idea that the process by which genomes and environments came into being was also "designed" - that is, that there was some foresighted process in which intention played a part in the bringing into existence of the physically embodied objects and processes in biology. Again, this is because this view of biological "design" does not conform to the three answers to the three questions listed above as answered by most scientists.

Where do we go from here?

As always, comments, criticisms, and suggestions are warmly welcomed!


Labels: , , , , , , , , , , ,


At 2/23/2009 01:58:00 PM, Anonymous Anonymous said...

I've always thought that the discussion between chance and necessity was a non-issue. They both seem to mischaracterize the constrained and unpredictable path that evolution takes.

Why not use "contingency" instead to reflect the interplay between "chance" and "necessity" in just one word?

Or, as Stephen Jay Gould said in The Panda's Thumb:
“Organisms are not billiard balls, propelled by simple and measurable external forces to predictable new positions on life’s pool table. Sufficiently complex systems have greater richness. Organisms have a history that constrains their future in myriad, subtle ways.”

Or Francois Jacob's expression from his 1977 paper "Evolution and Tinkering":
“It is hard to realize that the living world as we know it is just one among many possibilities; that its actual structure results from the history of the earth. Yet living organisms are historical structures: literally creations of history. They represent, not a perfect product of engineering, but a patchwork of odd sets pieced together when and where opportunities arose. For the opportunism of natural selection is not simply a matter of indifference to the structure and operation of its products. It reflects the very nature of a historical process full of contingency.”

That's just my perspective anyway. Great post as always Allen. :-)

At 2/23/2009 03:59:00 PM, Blogger Allen MacNeill said...

Here's a comment from my good friend, Will Provine:

Dear Allen,

What a thoughtful comment you made. I do have 2 serious drawbacks to it.

Many things in biology involve random factors. Random genetic drift is one of them. All evolutionists believe that there is a random parting of chromosomes in meiosis, and random breeding in the next generation. All senses of this randomness are based upon deterministic random number generators. In physics we have random processes for which we have no deterministic random number generators. There is no randomness in biology that is not based on deterministic random number generators. That makes a huge difference between physics and biology. In other words, the chance and necessity distinction means something in physics, but is meaningless in biology, where all chance is necessity.

My second point concerns your number one option for doing science:

• whether information can exist in purely "disembodied" form in nature, without a physical referent

In evolutionary biology, I rail against my colleagues, who treat natural selection as having no physical referent, or a made up one. Random drift is worse. All that is required is a small population of sexually breeding organisms. And almost everyone applies random drift to bacteria and other organisms that have no meiosis, and which makes absolutely no biological sense whatever. For this latter, biologists have no referent whatever, and merely invoke random genetic drift.

I agree with your point, but many working evolutionary biologists pay no attention to this problem. Then they treat ID’ers with zero respect when they point out that the evolutionists do about as they do.

Warm wishes, Will

At 2/23/2009 08:43:00 PM, Blogger Allen MacNeill said...

I have recently come to avoid the use of the word "chance" in describing non-quantum-level natural phenomena. I tend to agree with my friend and mentor, Will Provine, who rejects the entire idea of "chance" and "randomness" as causes in biology, at least in phenomena above the quantum level. For example, it is often stated that the outcome of meiotic independent assortment is the result of "chance" or "random assortment", similar to that produced by the "random" shuffling of a deck of cards prior to dealing out a round of hands. However, as a grandson of a stage magician who could perform a "perfect shuffle" (i.e. every other card from alternate hands), I am convinced that neither meiosis nor card-shuffling (nor coin-tossing, etc.) are genuinely "random" in the same way that radioactive decay or other quantum processes are "random" or the result of "chance". That is, all events in the macroscopic universe are the result of necessity, and what appears to us to be "random" events are actually the result of massive contingency.

Which, by the way, is a reasonably concise description of what I believe to be the origin and evolution of life on Earth, including my life (of course).

At 2/23/2009 08:45:00 PM, Blogger Allen MacNeill said...

As to the appropriate use of the term "design" in biology, I have come to have serious reservations about it as well. I tend to agree with Vrba and Gould that exaptation is perhaps as important as adaptation in biology. Indeed, I would go further and argue that virtually all characteristics of living organisms are exaptations, in the sense that they exist as the result of a long chain of contingent deterministic processes, rather than as the goal of an equally long chain of teleological processes. At first it would seem that the difference between these two is merely a matter of perspective, but I think the difference goes deeper.

If exaptations are outcomes, then they are produced by processes that do not require "disembodied" plans or programs at the beginning of the causal/deterministic chains which culminate in their coming into existence. However, if adaptations are goals, then this necessitates the a priori existence of the goal in "disembodied" form prior to the instantiation of the causal/deterministic chains which culminate in their coming into existence.

At 2/25/2009 02:28:00 AM, Anonymous Anonymous said...

Just to confuse things even more, I note that mathematicians can speak of different orders of randomness.

There are regions in the genome that are particularly susceptible to mutations ("hotspots"). There are some point mutations that are more likely than others in particular regions. This is nothing radical for biologists, but one UD poster made a hooplah about it last year, suggesting that such mutations are thus "non-random", which means evolution is being directed, which makes Darwinism false, which validates "front-loading", blah, blah, blah.

At 2/26/2009 11:13:00 PM, Blogger RBH said...

Hm. I always thought the only sense of "chance" or "random" that was invoked in evolution was the zero correlation between the distribution of variation in the selective environment (roughly, the topography of the nearby fitness landscape) and the distribution of mutations that occur. Have I missed something?

At 2/26/2009 11:19:00 PM, Blogger RBH said...

I should have added that I tend to regard "information" as a property, not a thing. Hence it makes no more sense to worry about the existence of disembodied information than it does to worry about the existence of disembodied temperature. The notion of "existence" gets pretty slippery there.


Post a Comment

<< Home