Last year Peter Jennings opened a segment of the ABC evening news, by stating that the DNA content of a single human baby would stretch from the Sun to Pluto and back, 15 times! But lately NASA seems to be telling us that all we have to do is find water in the universe, and the likelihood of life also being there with the water is high enough that we should allocate considerable sums of money to reach out and try to touch it.
How can life be the utmost definition of complexity in the universe, and yet be only a moist environment away from sprouting up all over? In the past three years, the number of articles relating to life in the universe in Sky and Telescope and Astronomy magazines alone is astounding, not to mention page after similar page in dozens of other science and nature magazines. The debate over life arising by chance compared to life being created continues to divide us, with little help seeming to be on the horizon.
How good are our estimates?
Scientists and engineers are generally quite good at making rough-order-of-magnitude estimates, most of the time. But how well do we do when it comes to our estimates involving very large or very small numbers? Often times our estimates and our subsequent reasoning/conclusions suffer greatly when we enter very unfamiliar territory way beyond our practical realm of experience. Such can be the situation involving cosmology and biogenesis, where the numbers themselves, and the statistics surrounding the numbers, very quickly can exceed the region of reliable estimation.
Most of us scientists find ourselves doing more preaching and believing than we do understanding and comprehending. The following black-and-white statements perhaps can help set the stage for the entire premise of this article:
1) Random-chance undirected processes always get the wrong answer; this is especially true when it comes to trying to increase the order and complexity of non-living matter to become living entities.
2) The 'Figure of Merit' of any living entity is so astronomically high with the coded information it contains, that there is not enough time or space in the entire universe to construct this merit function to produce life from non-life.
3) Because #1 above always gets the wrong answer when trying to achieve #2, one should ask: Where are all the wrong answers when it comes to the living creatures we have on this planet? Each living entity is peaked for staying alive in its niche on Earth, so where is the continuum of all the not-so-peaked creatures? The earth should be inundated with wrong answers (living and dead) - so where are they?
4) None of us can trust our intuition when it comes to extremely large or extremely small numbers, which are way outside our area of expertise. And even if you manage to compute the large or small numbers correctly, there is no way that you can fully grasp the meaning of the computations.
Probably a number of the readers of this article are boiling mad at this point, especially if the reader is a scientist or engineer. When I get a chance to speak to students about what we can really know, compared to what we think we know, many of the above items prove to be good discussion topics about our current state of understanding on a number of issues. Let's expand each of the above items in greater detail to see if we can perhaps learn something new. We will begin with how well we can trust our scientific intuition:
Number (4): Limited Range of Expertise
All of us understand the number 10, and that goes for the number -43 too most likely. NONE of us understands the number 10-43 that permeates many cosmological articles today. But we have scientists who purport to know and understand such concepts, to the point of writing hundreds-of-page textbooks about our supposed origin. They even produce hour-long PBS shows that expound on these ideas as if they were not "science fiction" but actually understood aspects of reality. The complete physics of the Sun's interior and exterior is poorly understood today, despite the fact that the Sun is a very-measurable laboratory right next door to us compared to the large dimensions of the universe. So if we don't have all the equations in closed-form for how the Sun behaves, how can we purport to know how the entire cosmos sprang from nothing, billions of years ago, in time and distance intervals that are ludicrously small? Just what is science fact and what is science fiction these days? It is very tough to know for certain in cosmology.
But I don't even try to wade into such unknowable territory when I am trying to get the attention of a young scientist eager to learn. I generally try to work through a simple example of how our scientific intuition gets poor quickly, when we depart only a little from our basis of experience. Sometimes we don't have to go even that far. Take for example the question "What would be the physical and biological impact for Earth if the viscosity of water were to change by a factor of two, up or down, from its current state?" None of us could answer that question without an incredible effort of research and calculation and experiment. We could imagine that the change in capillary action alone would take a devastating toll on plant and animal life, but that is only our gut-feel. Changing water's viscosity by only a factor of two would cause us to be way out of our range of experience, and our ability to comprehend the impact.
Oil, Oil Everywhere, and not a drop to burn.
But an even more basic example is easier to address. Rather than speculate about possible scientific effects of how a different water viscosity would change our world, I usually ask the class I am addressing, to consider a down-to-earth problem of oil consumption in cars. "How far could a car be expected to drive if it used up a single drop of oil per revolution of the engine?"
Well first, let's examine if a single drop of oil being used per revolution is even a reasonable sounding assumption. One drop doesn't sound like all that much, and our gut feel is that this amount of oil is small, certainly small in comparison to the engine containing it. A quick calculation shows that for a large-displacement engine, the stroked surface area of 8 pistons amounts to about five 8.5" x 11" sheets of paper area. Now, imagine one drop of oil spread out over this surface area. How thick would the layer of oil be?
If the class calculates the thickness of the oil film that one drop makes in coating the 8 cylinders, they find that the thickness is about 1/3 wave of light! I then inform the class that it is impossible for any of them to actually coat an equivalent flat surface area by hand, trying to spread a single drop out over the entire surface to a thickness of 1/3 wavelength of light. The class then usually agrees that this amount of oil usage is very small.
Before proceeding further with the example, it is worth noting that conclusions are sometimes drawn on seemingly very reasonable calculations or data, but in fact, further analysis shows a different conclusion is sometimes warranted based on additional deeper insight or analysis. Thus, just because one researcher performs an experiment or does a reasonable calculation, the conclusion might still be very much off the mark, even though the math or the research is correctly done. In our oil-use example, the intermediate calculation of a third of a wavelength of light of oil being used up per engine RPM seems very convincing, and it is based on good calculations. The trouble is, we probably have little practical experience with a number this small, when it comes to oil consumption in car engines, so we should continue the analysis somewhat further to try to validate if our initial assumption remains reasonable. For this example, that is fine. But how many technically-debated areas of the creation/evolution debates are halted at very plausible sounding conclusions based on good calculations and assumptions, only to be found out later that the debater's sphere of practical experience has misled people to accept answers which are false.
Going back to the drop of oil, an unconvinced class member might ignore the 'gut-feel' about oil films being 1/3 wave thick, and simply calculate how many miles the car can go if it uses a drop per revolution, given that there is something like four quarts of oil in the crankcase. This student arrives at the conclusion that the car could only get about 25 miles before running out of oil. In fact, typical cars today use less than 1/300 wave of oil per revolution; actually only a couple of molecular layers of oil are lost on each stroke. No one in the class really knows the difference between 1/3 wave and 1/300 wave, but they can all understand 25 miles compared to 2500 miles.
One might argue that it is a good thing that automotive engineers did not do a quick back-of-the-envelope calculation on expected oil consumption, before deciding on whether to attempt to build an internal combustion engine or not. Very logical reasoning could have been quite persuasive in concluding that it would look impossible to build an engine with such tight tolerances to allow the engine to propel the car more than a few miles at most. What would seem to be a good engineering estimate of 1/3 a wavelength of light of oil used per stroke (1 drop), is in fact over two orders of magnitude off in its prediction. Suffice it to say that too quick of a calculation based on a reasonable sounding assumption might sometimes lead good scientists to the wrong conclusions.
Carl Sagan liked to point out that the entire Earth was imaged as only a few pixels (Pale Blue Dot), when viewed by the Voyager 1 spacecraft from beyond Neptune. How petty all of our small concerns are when we realize our position in the huge cosmos, he loved to state. The next time you are flying at 35,000 feet, try to see how large the 200" Palomar mirror is as you look back from your plane window. It is a pinhole in the cosmos, virtually unresolved by your eye. We get almost nothing through that pinhole as we view the wonders of the heavens, and yet we make claims on the physics of the entire universe based on an incredibly small sampling of the available real estate. How quickly we can fool ourselves depending on which way we are looking, what perspective we want to take, and what sermon we want to preach.
So far we have only taken a few baby steps away from our "sphere of understanding". What happens when we head out into the cosmos, or perhaps peer down past the microscope into particle-physics territory?
Number (1): Random Chance ALWAYS gets the wrong answer.
The book Thinking Like a Physicist poses the following as one of its problems for the student: "Huxley is alleged to have said that 'six monkeys, set to strum unintelligently on typewriters for millions of years, would be bound in time to write all the books in the British Museum'. Could any array of computers acting at random within the time and distance scales of our universe produce even a single book?"
Generally evolutionists believe that given enough time and enough bio-chemical combinations, living matter evidencing complex order and structure can arise from non-living materials. In large part, such beliefs are kindled by the lack of examination of the numbers involved in the scenario being considered. Just like the oil-consumption example, how well can we expect Nature to do on its own in producing complex living organisms, and do the actual calculated numbers agree with our first estimates or conjectures?
The physics problem for the student poses a classic example of extrapolating our 'gut-feel' to arenas way beyond our area of expertise. With a scientific flare no less, we might find ourselves, like Huxley, declaring totally absurd garbage as fact. Did you notice how smooth and logical-sounding Huxley's statement was when you read it a few seconds ago? It is straightforward to work out the answer to the problem posed above, but I have found that it is even more revealing to determine the answer in a two-step manner.
I happen to have a 1-1/4" chrome-steel sphere (large ball bearing), that is not unlike the size of a normal telescope eyepiece. If we imagine filling this sphere with atoms, and we let each atom become a computer, and we let each computer-atom operate at the speed of light (one calculation in the time it takes for light to cross the diameter of the atom = 1,500 Billion MegaHertz), and we let all these atom-computers "type" with a 40-key typewriter (not even the ~105-keys on most PC's today), how many specified characters can this awesome computing power produce in the approximate age of the universe, 20 Billion years?
The answer is a whopping 38 characters!! 38 Characters; not even a long sentence.
But the real show-stopper comes if we scale the problem up to include the dimensions of the entire universe, not just a 1-1/4" speck. Fill the entire universe (20 Billion light year radius) with atom-computers, and let them calculate at the speed of light. How many more characters than 38 can we produce?
The shocking answer is only 90 characters, TOTAL.
Do yourself a big favor and take about 5 minutes of your life at this point to ponder the simple calculation in the last two paragraphs. If this doesn't cause you to take a serious re-evaluation of some of your time-honored ideas about life, and life in the universe, then I guess you will probably be forever criticizing the Kansas Board of Education for being such fools.
Do you see what is going on with this two-step example? Expanding the scope of the problem from a speck of dust to the entire size of the universe, does not even add a factor of three more to the information generated by a universe full of computers. In any reasonably complex system, random chance always gets the wrong answer. And it does not make any appreciable difference if we have 105 keys, or even a binary 2-letter keyboard. Go ahead and work out the numbers. We do not get the entire works of the British Museum as Huxley would have us believe. Not one book, not one chapter, not one page, not one paragraph. But Peter Jennings told us that our human information content went to Pluto and back 15 times! Something is desperately wrong with our gut-feel.
Let's not lose the point here. Our gut-feel, that given the dimensions of the universe in space-time certainly there must be the chance that life would occur somewhere, is just as poor an example of reasoning as our oil-drop usage example earlier. Life is infinitely more complex in its coded sequences, than a 40-character keyboard. The simple and inescapable conclusion is that random-chance processes are not enhanced by scaling up the linear dimensions and time dimension of the problem. Infinitely more "wrong" answers are obtained compared to barely lengthening the sequence of "good" instructions.
As an analogy, it might help to think of the Power-Ball Lottery at this point. Here, only 6 balls with less than 50 "keys" per ball produces an 80,000,000 to 1 set of odds that you will lose. As the dimensions of the random-chance scenario increase beyond 6 balls and 50 possibilities/ball , the odds diminish exponentially to the point of absurdity. Who would trust such odds to produce anything useful - to produce any winning combination? Apparently Biology does, because the most incredibly bad odds imaginable (actually, unimaginable) are what evolution has as its foundation. It is hard to define NEVER in a better fashion than to think of how absurd it is to hope that random chance will ever produce anything complex. How far out of our area of expertise are we at this point, and in what are we putting our faith? Hopefully not in scientists like Huxley, or Sagan, who have their non-creator axe to grind, and never did bother to work out some simple numbers like the example above for their readers.
At this point someone usually makes the counter argument that maybe we can't get a specified coded sequence by random chance processes, but we certainly can get a vast number of randomly coded sequences, and one of those is what arose to become 'life'. Perhaps then this is the fallacy in the preceding conclusion about the infinitely-poor odds against life arising by chance.
Well, let's say life somehow did occur, by chance. And life flourished evolutionarily so well that it eventually produced a very complex animal, like a horse, for example. If we start at the very end of this line of thinking, namely the perfectly functioning animal which is incredibly fine-tuned as a living organism, and we back up just a little in its coded sequence of life instructions, say by only 90 "instructions", and then we let Nature randomly try to find the correct pathway of these last 90 instructions, the answer is what we have already seen, that random chance will not be able to find the required 90-instruction solution in the lifetime of the universe.
As an example, let's say the horse is perfectly functioning in all respects, except that it is missing the ability to see, because the very last set of instructions (or characters from a 40-key keyboard) for producing sight is missing:
CONNECT NEURON No.178 TO NEURON No.255 ONLY AFTER CONNECTING NEURON No.101 TO NEURON No.102.
Unfortunately, this instruction sequence is >90 characters long, and thus Nature could never get this needed sequence by random chance. And it certainly does not help the problem to let Nature change all the other finely-tuned good-working life-facilitating coded sequences of the horse to become variables now, in the hope that by randomly changing these sequences along with the missing sight-producing sequence, that a new combination will result that keeps the horse a horse, keeps it a live horse, and gives it sight along with preserving all the other life-facilitating coded sequences. That is, the problem of searching randomly for the correct sight-producing sequence is not helped by allowing all the other 'good' sequences of life to be randomized again at the same time. RANDOM2, or RANDOM129, or whatever, does not help solve the sight-loss problem in the horse. More random combinations produce infinitely more wrong answers, which hurt the horse rather than help it see. Basically, the horse goes from being only blind, to being blind, and lame, and diseased, and dead.
But the problem is even worse than that for the horse. Nature does not have a universe full of combinations to try anymore, with very complex living creatures like the horse. Nature actually has a very finite population to work with, on a very finite spec of dirt in the universe (The Earth). As Nature starts to play with the finely-tuned merit function of the horse to try to get it to see, all the wrong answers begin to emerge rather than the right answer we are hoping to get. The finite population of almost-seeing horses dies off quickly while Nature fumbles with even more disastrous combinations it is trying as it rearranges the horse anatomy.
Evolution can't have it both ways. If one wants to think that "simple" cells can come out of a universe full of random chance chemical interactions, fine. But at the other end of the upward complexity walk, the highly-evolved living organisms are of a very finite number, nowhere near filling the universe, or even a planet. To alter a finely-tuned living organism by randomly changing its genetic code, will result in de-tuning that organism in a life-threatening manner infinitely more times than enhancing the organism's genetic makeup. Unfortunately, there are too few of the finely-tuned organisms in the population to try to randomly experiment with, in the hopes of evolving a more-complex organism.
It has been argued that some microbiologists have conducted experiments where certain genetic sequences were "surgically" removed in simpler life forms, and that after a dozen or so generations, these life forms "reconnected" the missing genetic information by evolutionary random processes. This kind of reasoning is exactly the situation described by the blind-horse scenario. No one doubts that the reconnections were formed after a few generations. The problem is, what process should we assign the gathering of the 'new' information to? Random chance? Or is it that the incredibly complex genetic information already possessed by every living being (remember Peter Jennings?), has the ability to heal/repair/problem-solve itself to keep life flourishing on this planet? Is it the situation that random chance eventually finds the answer, or does the living organism already have the coded information self-contained to afford the best chance for life to adapt and to survive? The question to the microbiologist is: Work out the probabilistic mathematics on the missing sequence of information, and then ask if random chance attempts at regaining this specific sequence can happen in the lifetime of the universe? How about the chance of it happening by chance in a dozen generations? It would seem incumbent upon such researchers to work out the probabilities of their events happening by chance, and report those numbers along with the evolutionary claims, to try to put a reality check on whether or not evolution has happened at all. More certainly, what is termed "the miracle of life" by evolutionists and non-evolutionists alike - that is what really accounts for the ability of genetic sequences to "re-establish" themselves in short order. If the organism had to wait for Nature to shake the dice correctly to repair the genetic damage, Nature would get the wrong answer infinitely more often than it seemingly does.
I don't expect students to pick up on this tremendous failure of random chance to produce meaningful results, in just a few minutes of discussion time in class. I find a better way is to roll a single six-sided die and count up the number of times the rolls allow us to correctly answer a simple problem such as: How much is 1 + 2? There is a 1 out of 6 chance that by rolling the die we would get the number 3 as the correct answer. And let's say that when we get the number 3, that correct answer facilitates "life" and helps to add to the increasingly complex genetic code in a beneficial manner. All other die combinations either hinder that process, or virtually leave it alone. The further away from the correct answer 3 we go, the more the hindrance.
If the students record on sheets of typing paper the number rolled by the die, we then can start stacking up the results as piles of paper. If after a 1000 throws of the die we look at the results, there will be close to 167 sheets of paper with the number 3 on it. Let's call these "living". The 2's and the 4's are close to the right answer, and their total will be about 333 sheets. The 1's, 5's, and 6's are further away from the right answer, and they total about 500 sheets of paper. In total, there are only 167 living sheets compared to 833 non-living or quasi-living sheets. Thus, for a 6-sided die, there are a lot more poorly-living or non-living answers generated by random chance, than there are correctly living answers. 833 wrong answers makes a large pile that might even look somewhat overwhelming in comparison to the much smaller pile of 167 correct answers.
The question is then posed, where are all the wrong answers in Nature's rolling of the living dice? Why don't we have a few right answers of very well adapted living creatures, and an overwhelming pile of poorly-living wrong answers? But absolutely everything that lives on this planet is the epitome of a very complex right answer, and there doesn't seem to be a continuum of "less-right" answers in Nature. One person gets the PowerBall jackpot "right", and 79,999,999 others get it wrong. Where is the pile of wrong organic attempts by Nature that illustrates the random chance evolutionary process? This pile doesn't exist. EVERY living creature is a masterpiece of coded information that defies all probabilistic computations of how it could ever get to function at that level by random chance. In the end, the large pile of wrong answers helps the students tangibly understand how utterly ridiculous it is to expect miracles of life to arise out of garbage-producing random-chance events.
Certainly, only those answers very close to the one correct answer will allow less-than-ideal life to continue, while the overwhelming majority of the wrong answers will terminate life. Nature, by throwing incredibly complicated many-sided dice in the random-chance evolutionary process, will produce vastly more wrong answers than right answers. That's the way probability works. And the genetic sequence is so much more complicated than any of these simple examples of dice, that random chance processes will never get the right answer as compared to always getting the wrong answer.
Back to the real question in the above discussion: Where are all the "almost right answers" in the living world? They are no where to be found. Every living organism seems exactly and ideally tuned for having all of its bodily functions working in complete harmony with keeping it alive. Listen closely to the next PBS nature broadcast, and it will state this same exact conclusion. Random chance would demand vast populations of "almost right" answers that are close to perfection, but missed the mark measurably. Where are the living organisms with lots of appendages that have unknown or useless functions, that don't kill off the organism per-se, but are rampant in numbers compared to the "correct" answer for life? Where are all the random organisms that lack bi-lateral symmetry? Why all the symmetry in living things anyway? All we seem to have on this planet are untold kinds of living organisms that are peaked for performance. Probability demands that there should be vastly more less-than-peak organisms that random processes produce on a far more frequent basis than the "right answers" we actually have.
And remember, the premise here is that Nature has only a limited supply of highly-complex highly-evolved organisms to try its random re-tuning process on. The coded-life sequence of living organisms is so much more complicated than a single 6-sided die, or than a PowerBall drawing, that random changes in the genetic sequence of living organisms by Nature will kill off the finite population infinitely more times than it will ever get the "right answer" that enhances the function of the organism. Every living organism seems to attest to the fact that it is the peak of efficiency in dealing with its environment, that it is not burdened with tons of wrong genetic answers, and it has exactly the right parts of construction that make up its framework to allow it to survive.
Why is this the case for all living creatures on this planet? Random chance and natural selection cannot account for all these finely-tuned living organisms. The populations of each 'kind' are far too small to allow evolutionary tampering to try to eventually produce a 'right answer' out of an infinity of wrong answers. If this is not obvious, then just try to use random chance processes to solve any problem you work on in your daily routine. For example, try to type out today's date on the typewriter by randomly hitting the keystrokes. Or program your computer to do this, and you will find it will take it years of CPU time to get the correct result. In 20 billion years you will never be able to replicate a single page of your checkbook, even if the whole cosmos is filled with computers at your command. ANY process we would attempt to solve or enhance by randomly trying all combinations, is doomed to failure because we get the wrong answer vastly more often than the right answer.
In the last century, astronomers have taken hundreds of thousands, if not millions of photographs of the heavens. If random chance works so well, perhaps some astronomer can find a photo that spells out the name 'Carl Sagan' in the stars. Or maybe someone can find this name in the myriad of craters we see on planetary landscapes. Maybe we could settle for even a sans-serif 'Carl' if you wish. The devastating conclusion for random chance is that randomness does not produce complexity. If we can't find even a simple organized word in the Billions and Billions of star images we have taken, why does anyone think that DNA can arise out of nothing, beating all the odds?
In my own field of optical design, I like to illustrate this scenario by selecting a straightforward lens, the double-Gauss, as an example of how random chance can never produce a right answer. A double-Gauss is a 6-element lens made of various glass types, that can be designed to produce a good quality image (the "right" answer). If a designer, or a computer programmed by the designer, were to try to randomly examine all the possible combinations of glass types, surface radii, lens thicknesses and spacings, focal lengths, and fields of view, the number of possibilities easily exceeds 1080, a staggering number. And that is not even including additional parameters such as non-spherical surface types, gradient-index lens profiles, diffractive and binary optical elements, tilts or decenters, mirrors, and the list of dimensions could go on even further.
It is interesting to note that there are tens of thousands of double-Gauss designs that would be useful lens systems. So it is not the case that there is only one right answer. The point we dare not lose in our thinking about the lens example, is that random chance will never find any of the possible solutions. There are simply too many possibilities to search through, that a universe of computers doesn't have enough time to try them all. We can understand this for this easily-calculated example, but the evolutionist refuses to apply the same logic to the incomprehensibly complex assemblage of life. The only way to fail to see this incredible contradiction of logic is to choose to believe in something other than the probabilities show.
That random chance fails miserably to arrive at solutions to even the simplest of organized systems, is certainly not news to anyone that tries to design systems to do certain tasks. The only way a lens designer can make any sense out of all the possible combinations is to direct the effort at finding solutions. This is so universally understood that it seems silly to even think otherwise. Yet biologists are notorious in their belief that somehow Nature can randomly produce the most complicated structures in the universe, living creatures, by completely undirected random-chance purposeless processes. The 'merit function' that describes any living creature - that is, all the items that have to be operating correctly and simultaneously correctly - is so astounding that none of us can even construct it, let alone comprehend it. We have no idea how living things actually keep on living, they are so complex. In the much simpler parameter-space of optical design, we can get a good handle of all the items it takes to build a merit function to obtain peak performance out of a lens system. And even that very simplistic example is beyond the scope of the time and distance scale of the universe to have random chance arrive at a solution for a lens. The next time you take a hike in a National Park, see how many Nature-produced double-Gauss lenses you find laying on the ground. How about that plant you see on the hike? Did random chance produce it, even though it is infinitely more complex than a double-Gauss lens?
The challenge for the evolutionary crowd is to take some simple examples in biology, like we have done here with the examples in this paper, and work out the evolutionary mathematics and probabilities that support the evolutionary claim that random chance can actually produce increasing complexity, and ultimately life. Let's stop being philosophers and preachers, and let's start cranking out some actual examples for our peers to evaluate.
The common person in Kansas, and every other state, knows that NO scientist uses Darwinian evolution to solve physics problems in the laboratory. Random chance always get the wrong answer. And the public knows by common sense that this is what is at the heart of the debate. By ordinary real-life experiences, the public knows that the probability of anything complex self-organizing is vanishingly small. Only in Biology are we to believe that somehow random chance can actually defy the odds. No researcher, nor Nature, by random chance alone, can produce any meaningful improvement in a complex process.
For those who don't believe that random chance produces garbage rather than ever-increasingly organized systems (e.g., life), let's debate the science, and the math, and the probabilities in the open forum. Let's do it in the schools too. That, fellow readers, is what Kansas is trying to do. If the odds are with evolution scientifically, then science has nothing to fear from the open debate and the open learning process. If the odds show that the current evolutionary thinking is poor science, then the scientists can continue to search for better explanations without ridiculing others who think they have found one. Either way, the debate produces a better educated student and public. And isn't that what we all want?
David E. Stoltzmann
Optical Engineering of Minnesota
368 North Ninth Street
Bayport, MN 55003-1145