New documentary: Dismantled: A Scientific Deconstruction of the Theory of Evolution
The online premiere has ended, but you can order the DVD or Blu-ray here.

Historical Science, Chaos Theory, and the sliding scale of trust

by and

Published: 1 October 2020 (GMT+10)
Image for commentary (Fair Use)jeff-goldblum
Jeff Goldblum as Dr Ian Malcolm

The question of historical vs operational science is raised often in the evolution-creation debate. But evolutionists often cry foul, saying there is no distinction between these two supposed forms of science—that this is something the creationists made up. Nothing could be further from the truth!

Historical science is fundamentally less reliable than operational science, primarily because unlike operational science, historical science is unfalsifiable in the strict sense. This is directly contrary to the views of Dr Carol Cleland, a secular philosopher of science who has attempted to blunt this argument by claiming that the two are on the same level.1 But it is not just the lack of falsifiability; there is also a lack of reliability, and this gets worse the farther back in time one tries to go. Sure, it is possible to find fossilized remains of an ancient animal that clearly show the reason it died, and it is possible to find remains from a recently killed animal that are less well preserved and thus the reason for death is more mysterious, but, in general, time obscures evidence. This is a major problem with historical science.

Briefly put, operational science deals with the ongoing operations of nature, and is therefore subject to direct experimentation and—this is key—repetition. The fact that we can repeat experiments and consistently confirm the results is a major factor in how operational science works at a fundamental level. Operational science was an amazing development in human history, and we mainly developed through an application of Christian theology to the natural sciences.

Historical science, on the other hand, lacks this factor of repeatability (since we can’t repeat the past). Historical science is all about looking at evidence in the present and trying to piece together a picture of the past based upon those clues. This process is naturally laden with conjecture.

The sliding scale of reliability

Frequently, trying to draw a distinction between operational and historical science will generate a complaint in reply: “What about forensic investigations, like crime scene investigations? Should we not trust the results of forensic science just because the crimes are not repeatable or because they were not directly observed by the investigator?” But forensics does not operate in a vacuum. It is usually combined with eyewitness testimony and relevant circumstantial evidence. Neither of these are available when extrapolating into the distant past. The exception to this is that we have a book that claims to be eyewitness to the history of the universe: the Bible. In the secular version of historical science, it is the amount of time about which we are speculating that matters. The further back in time they attempt to go, the less confidence we should have in their speculations. Forensic investigators are looking at evidence on a very short timescale, comparatively speaking. Secular paleontologists, however, make claims stretching into the millions or even billions of years. Hence, they are introducing a giant amount of uncertainty. And their speculations have no bounds because they are unconstrained by any eyewitness testimony. This has huge implications for the creation/evolution controversy!

Why should historical science be viewed on a sliding scale of reliability, with more recent events being more certain than more distant ones? The primary reason is that more time equals more uncertainty.

“The Malcolm Effect”

We are going to use an illustration from a 30-year-old book, Jurassic Park, that helps to illustrate what is going on and why historical science is important to understand. Spoiler alert: If you enjoyed the movie, you may not like the book! Or vice versa, since the two stories play out very differently. The print version of Jurassic Park delves deeply into philosophical ideas, much more than the movie does, and these musings are usually voiced by the character of mathematician Dr Ian Malcolm. Make no mistake: there is no hint of creationism to be found in the pages of Jurassic Park. It portays evolutionism from cover to cover. Nonetheless, some of the ideas that Malcolm brings up, particularly relating to so-called ‘Chaos Theory’, are highly relevant in this context.

In the book, the term ‘Malcolm Effect’ is used in reference to how things tend to go wrong when people incorrectly assume they’ve got it all figured out. That is the ultimate moral of the story in Jurassic Park: the hubris of mankind leads to our downfall. That hubris may not be displayed anywhere more fully than attempts by naturalists to explain everything in the world without God. In reference to historical science, the ‘Malcolm Effect’ can be thought of this way: the further back in time someone wants to stretch the evidence, the more likely they are wrong about their conclusions. To quote Ian Malcolm (or, more accurately, Michael Crichton):

“Chaos theory says two things. First, that complex systems like weather have an underlying order. Second, the reverse of that—that simple systems can produce complex behavior. For example, pool balls. You hit a pool ball, and it starts to carom off the sides of the table. In theory, that’s a fairly simple system, almost a Newtonian system.
Since you can know the force imparted to the ball, and the mass of the ball, and you can calculate the angles at which it will strike the walls, you can predict the future behavior of the ball. In theory, you could predict the behavior of the ball far into the future, as it keeps bouncing from side to side. You could predict where it will end up three hours from now, in theory …
But in fact … it turns out you can’t predict more than a few seconds into the future. Because almost immediately very small effects—imperfections in the surface of the ball, tiny indentations in the wood of the table—start to make a difference. And it doesn’t take long before they overpower your careful calculations. So it turns out that this simple system of a pool ball on a table has unpredictable behavior.”2

Historical science is much like trying to predict what will happen in the future—only reversed! Just as it gets harder to predict the future as you try to go further and further on, the same applies in the other direction. The pool ball is a small-scale example of a larger scale phenomenon. The point is, our ability to predict and extrapolate is imprecise because we are not omniscient and we are always working on limited information. The more you stretch your predictions, the more this imprecision and lack of omniscience starts to come into play.


This is why we are justified in being reasonably confident when we look at forensic science in the short term, like with crime scene investigations, especially when it is corroborated with eyewitness accounts that can constrain the possible conclusions. Yet, we feel equally justified when we are skeptical of the conclusions of scientists who are drawing conclusions (contrary to the Bible) spanning millions or billions of years.

We suspect this ‘Malcolm Effect’ plays a very big role in questions such as radiometric dating. We can test the rate at which the unstable elements change into stable daughter elements today, but trying to extrapolate precise predictions of such a process over alleged millions of years is fraught with problems for exactly the reasons that Crichton explained above. What seems straightforward and “simple” is almost never as simple as it may appear. In the case of radiometric dating, depending on the method being used, one has to assume the decay rates never change, that there has been no leaking of these water soluble or gaseous elements into or out of the system, that the temperature of the same over the course of its existence is known, that we know the starting concentrations of the radioactive element and its daughter products, etc. We human beings are notorious for overstating our own ability to ”know things for a fact”.

Things change—some quickly, and some gradually. Some things change so gradually we don’t even notice over the course of our lifetimes—or in all of recorded human history! Other things are subject to short bursts of sudden change followed by long periods of stability. Considering the relatively short history of modern science, how can we claim with any certainty that we know enough about present-day processes to successfully extrapolate them millions of years into the past? The Ian Malcolm character should have been appalled at this, but he accepts evolutionary theory and deep time in the book without any apparent question. That’s realistic, since most modern-day academics are guilty of the same thing!

Assumptions, hubris, and more assumptions!

How can evolutionary scientists rule out the possibility that things have happened in the past (e.g. Divine Creation or a worldwide Flood) which are no longer happening today? Secular scientists generally make a blanket a priori assertion that ‘the present is the key to the past’. That is, they assume they can look at the gradual changes happening today and extrapolate backward as far as they like. This ignores any event (like a worldwide flood) that would interfere with their calculations.

Imperceptible changes

The history of science is full of examples where new discoveries forced a wholesale revision of established ideas. This is why we no longer believe in phlogiston theory, but other examples are not hard to find.

This brings up another question: how many ongoing processes exist which would affect earth over long timespans, but which we are currently unable to measure or witness? Just like the tiny grooves on the surface of the pool ball affect our attempts to extrapolate its motion, many tiny factors of which we are unaware almost certainly will affect our ability to extrapolate things like geological processes, radiometric processes, etc., over deep time. And major perturbations (like a sudden burst of radioactive decay) cannot be ruled out. Because we cannot rule out any such possibilities, we should view all long-term extrapolations into the alleged deep past with a large degree of skepticism. All such extrapolations are based upon an assumption of complete knowledge, and that is never a good assumption for humans to make.

Missing evidence?

Another of the big reasons to take historical science with a progressively larger grain of salt the further back it goes, is that with the passage of time comes the destruction of relevant pieces of evidence. Oftentimes it is hard to know at the outset of an investigation exactly which pieces of evidence will turn out to be the most important. But the longer you wait, the more chances there are for important pieces of evidence to become obscured or lost. ‘Old earth’ scientists claim to be able to reconstruct the distant past based only upon the clues we have available to us today (and their own imaginations). How can they be at all confident that important pieces of evidence have not been lost?

In one study, it was determined that physical (forensic) evidence was only used in solving about 13% of cold cases, compared to eyewitness testimonies at about 63%!3 If it is so difficult to maintain the usefulness of physical clues to solve riddles only a few decades old, how much more difficult is it to keep hold of evidence over timespans of millions of years or more?

Let’s examine a couple of geological examples. First, rock arches are highly interesting and enigmatic formations that present major problems for old-earthers. They look more like a relic from the fast erosion of the Flood.4 They are also disappearing quite quickly. In Utah’s Arches National Park, they are collapsing at a rate of about one per year, giving a maximum of 2,000 years before the entire supply is depleted.5 Of course, one cannot extrapolate out that far with any degree of certainty. The arches most likely to collapse have probably already done so, yet one large earthquake in the region could topple many arches simultaneously. On an old earth, they would undoubtedly all be gone by now (ignoring the problem that no mechanism on an old earth could account for their original formation anyhow).

In another example, the cliffs of Joggins, Nova Scotia, are replete with fossils, some of them being huge polystrate lycopods that demonstrate the catastrophic and rapid nature of the Flood.6 Yet, these cliffs are eroded twice daily by some of the highest tides in the world!7 New exposures are regularly created through erosion. This implies a corollary: fossils are being lost on a daily basis to the sea, likely never to be seen again. What if some of these fossils turned out to represent ‘Precambrian Rabbits’? Evolutionists often claim they would abandon their theory if we found something like a rabbit in a Precambrian stratum (layer); but even granting there are none to be found today, how can they know with any certainty that such evidence did not exist in the past, and is now lost? With evidence being constantly destroyed all around us, the picture we see of the past gets dimmer as time goes on.

Fiat creation

Historical science eventually runs into the hard wall of fiat creation. It is possible to extrapolate past that wall, but there is no real history in that realm. Saying, “If these rocks erode at a rate of 1 mm per century, it would have taken 1 billion years to make them look as they do today,” does not mean that one billion years has actually passed. First, you cannot know that this rate was maintained over all that supposed time. Second, a historical event like Noah’s Flood could have caused millions or billions of years’ worth of erosion in a short time. Also, unanticipated forces could have contributed to your calculated rate. As scientists found out the hard way a few years ago, lightning erodes rocks orders of magnitude faster than rainfall and maybe even faster than frost cracking.8 One lightning strike can cleave off a flake of rock the size of a dinner plate or larger. And geologists are only recently becoming aware of how frequent landslides are, and how much material they bring downslope instantaneously. So how fast do mountains erode? If you factor in every real (but possibly unknown) factor, you would conclude that the presence of ‘young’ mountains like the Alps, Andes, and Rockies are excellent evidence for the youth of these vast ranges. But if you minimize the erosion rates, you might calculate a formation time before they actually existed.


In conclusion, we discovered that historical science gets progressively less reliable the further back in time it goes, as a result of three primary factors:

1) Chaos—the working of unknown factors that are imperceptible in the short run (or the immediate present) that would invalidate hypotheses based upon the extrapolation of present-day processes that are imperfectly and incompletely known.


2) The loss of relevant clues to the sands of time as physical evidence is destroyed or hidden by innumerable different factors that come into play over time (or, generally speaking, entropy).


3) The fact that historical science eventually runs into the hard wall of fiat creation, causing one to draw historical conclusions that predate actual history.

It is for these reasons that creationists are justified in being skeptical of the way forensic scientists make claims about the ultra-distant (alleged) past. In fact, we are far more justified in being skeptical about these claims than we are when we see forensics playing a deciding role in murder convictions. In the long run, testimony from reliable witnesses is highly preferable to speculative reconstructions. And the Ultimate Witness, the most reliable Being in the universe, gave us a witness and a testimony that the universe came into being through divine, fiat creation. This is found in the pages of the Bible.

References and notes

  1. Price, P., Examining the usage and scope of historical science—a response to Dr Carol Cleland and a defence of terminology, Journal of Creation 33(2):121–127, 2019. Return to text.
  2. Crichton, M., Jurassic Park, Ballantine Books, New York, 1991, p. 75. Return to text.
  3. Davis, R.C., Jensen, C.J., III, Burgette, L. and Burnett, K., Working Smarter on Cold Cases: Identifying Factors Associated with Successful Cold Case Investigations, J Forensic Sci, 59: 375-382, 2014. doi:10.1111/1556-4029.12384 Return to text.
  4. Oard, M., Many arches and natural bridges likely from the Flood, Journal of Creation 23(1):115–118, April 2009. Return to text.
  5. Batten, D., The age of arches, Creation 40(4):23, October 2018. Return to text.
  6. Price, P., How the Joggins polystrate fossils falsify long ages, 16 April 2020. Return to text.
  7. Juby, I., The Joggins Polystrate Fossils, in: Oard, M. & Reed, J., eds., Rock Solid Answers, ch. 13, Master Books, Green Forest, AR, 2009, p. 217. Return to text.
  8. Knight, J. and Grab, W.W. Lightning as a geomorphic agent on mountain summits: Evidence from southern Africa, Geomorphology, 204:61–70, 2014. Return to text.

Helpful Resources

Rock Solid Answers
by Michael J Oard, John K Reed
US $20.00
Soft Cover
Evolution's Achilles' Heels
by Nine Ph.D. scientists
US $17.00
Soft Cover

Readers’ comments

King T.
Perhaps my understanding and knowledge of the forensic investigation process is lacking but I think that most of the procedures and conclusions reached are based on repeatable events - things that are known to happen and that could happen under a given set of circumstances, e.g. blood stains becoming visible or detectable when certain chemicals are used or the time it takes for maggots to develop etc.. Hence anyone immediately jumping to point out how forensic investigations are examples of historical science that work are also jumping the gun - in most cases it's not in the same ballpark as the non-repeatable pre-historical conjectures.
Paul Price
Historical science is not repeatable. If it were repeatable, it would be operational science. The practice of forensics in crime investigations overlaps both realms, since it involves operational science (repeatable experiments on how guns leave traces, etc.), and historical science (interpreting clues in order to build stories about the past). For all the reasons discussed above, the process of extrapolating present-day processes into conclusions about the past gets more shaky the further back you try to go. And we cannot repeat the most important formative episodes in our Earth's past: Creation and the Flood.
James H.
I don't know why creationists or anyone else even uses the term, 'historical science.' You can't do science on the past. As you said, it's not repeatable and not observable. Any study of past events is history, and really it's only historical speculation. No one can be certain of past events (without reliable written records), as they can't be repeated. Even in astronomy, we're not looking back in time, regardless of how old the universe is. We're only seeing the light and aged remnants from the past, not actually observing the events. As you said, the further back you go, the more likely your description of events is wrong. Unless you want to redefine what science is, there is no science involved. CMI should stop talking about historical science.
Paul Price
You are not the only person who holds that opinion. Have you read this essay yet?
Stephanie S.
Yesterday’s article of the day was about forensic evidence that was wrong.
Paul Price
Yes, that certainly is interesting timing. Yesterday's article was a demonstration of how even recent forensic evidence can be badly misinterpreted, especially when biases are involved (evidently an anti-religious bias played a role in this case). How much more true is it for evidence allegedly millions of years old?
Dean D.
My wife and I like to watch The First 48, a television show about solving murders. I’d love to ask one of the investigators a question, “If you couldn’t investigate this murder for 20 years, would you likely be able to solve it?” I imagine a look of unbelief would suddenly appear on the investigator’s face.

Imagine a bloody glove being blown around by the wind or taken off by animals. Or footprints left by the assailant being washed away by rains. Evidence readily available at the time of the murder is now gone. Some of the eye-witnesses have disappeared or died. Where would the investigator start?

The whole premise of the show is that the investigators must get a lead within the first 48 hours or the chance of solving the murder is cut in half. Extrapolating that to millions of years and it shows the folly of attempting to come up with the correct answer without an eye witness to tell us. In this case, we have one; God Himself.
Richard P.
Several years ago, when I was a schoolteacher, the Head of Physics in the school mentioned the "Big Bang" during a conversation over lunch. I asked him what evidence convinced him that the Big Bang had occurred. He replied that redshift in cosmic observations shows the universe to be expanding, and so we can work out when it must have begun.
I then pointed out that, if any of his pupils made some observations in a Physics experiment, plotted them as points on a graph, and then extrapolated that graph to derive conclusions millions of times beyond the scale of the plotted points, he would surely fail them. He had no reply!
It is well known in statistics, and in most applications of the natural sciences, that extrapolation is tenuous at best, and such extreme extrapolation is altogether worthless (or worse: deceitful). The points are well made in this article.
Stephen N.
Thanks for the excellent article about so-called historical science. Using the examples of forensic science and eyewitness testimony, we were given a tragic example of a wrongful conviction in Australia in the reprinted article from yesterday. In Canada. there have been several famous examples of wrongful convictions over the years. Not only can so-called forensic science be used to reach false conclusions, but even eye-witness testimony can be notoriously unreliable at times. If this is true of the human justice system, how much more is it true in trying to determine exactly what happened in the distant past. Human methods of investigating what happened in the past can be unreliable. You have made a point that is difficult to argue against.
Jean L.
A good quote (admission) on this topic comes from Ernst Mayr: "Evolutionary biology, in contrast with physics and chemistry, is a historical science—the evolutionist attempts to explain events and processes that have already taken place. Laws and experiments are inappropriate techniques for the explication of such events and processes. Instead one constructs a historical narrative, consisting of a tentative reconstruction of the particular scenario that led to the events one is trying to explain."
So much for creationists making up the idea. Anyone can make up a good story, but that doesn't make it true.
[Mayr, Ernst. 2010. "Darwin's Influence on Modern Thought." Scientific American 283, (July): 78]
Paul Price
Yes, although unfortunately Mayr in the same speech made the claim that historical science can be "tested", which is plainly false. He equivocated on the meaning of the concept of "testing" in order to obfuscate the point.
Andrew S.
Meteorologists use a concept called Shannon Entropy to describe how numerical weather prediction models lose information through time. Also, the difference between historical and observational science can be boiled down to the problem of induction, as outlined for example by David Hume. If science is based on experiment and direct experience, then projecting such observations across distance and time raises a problem - one must assume it is acceptable, but we cannot prove it scientifically. Such inductive inferences may work for gravity, but what of dating rock layers? This is far more of a problem for science.
Geoff C. W.
"Another of the big reasons to take historical science with a progressively larger grain of salt the further back it goes..."
Could you have meant "a pregressively SMALLER grain of salt"?
Paul Price
I meant a larger grain. As in, the longer back in time the historical science deals with, the less we can be confident of it.
Tim N.
A good example of the shortcomings of historical / forensic science is the JonBenet Ramsey case. It has been said that the investigation was botched because the crime scene wasn't preserved. As a result, there were too many variables to make a clear forensic determination.
Roy N.
Tracking down the claim that arches in Utah are collapsing at a rate of one per year leads ultimately to a now-removed anonymous isolated unreferenced Wikipedia edit from 2006. There is absolutely no reason to believe it is true, and it this article should be revised to eliminate it.
Paul Price

Thanks for bringing this oversight to our attention (the lack of citation was unintentional in the article on arches collapsing). You will find this has now been corrected; according to park rangers, and allowing for a small number of losses due to vandalism from tourists, we still arrive at an average collapse of one per year between 1977 and 2015. FYI, the missing citation never had anything to do with Wikipedia. :)

Comments are automatically closed 14 days after publication.