web space | free hosting | Business Web Hosting | Free Website Submission | shopping cart | php hosting

Chapter 1


[Part 1]

The three questions which closed the last chapter are actually closely interrelated. For if Moses wrote the Pentateuch, then he certainly would have had access to the correct information concerning the Exodus, and it is reasonably probable that his information on Creation and the Flood are accurate as well. Also, if the information on Creation and the Flood and the Exodus are accurate, there is no good reason to deny the Mosaic authorship of the vast bulk of the Pentateuch. On the other hand, if the Pentateuch and Joshua were written late, there is little reason to expect them to be accurate, and if they are inaccurate, it is virtually certain that they were written late.

It may surprise some, but I think the easiest approach for now is to deal with the question of the historicity of Genesis 1-9 and its relation to the theory of evolution. [112]

Evolution, Creation, and the Flood

Numerous attempts have been made to relate the early Genesis account to the theory of evolution. These can be divided into four groups. First, there is mechanistic evolution. This theory holds that the universe and life in it evolved by purely naturalistic means, without any outside interference. In this view the geologic column represents millions of years of time (currently felt to be 4.3 to 4.5 billion years total, of which 550 to 600 million years, called the Phanerozoic, have undisputed traces of life). The adherents of this theory commonly hold that the early Genesis account is entirely mythical and thus unreliable.

Second, there is theistic evolution. This theory holds that mechanistic evolution is essentially correct except that at certain points (or perhaps continually) God helped the process along a little. This theory usually deals with the early Genesis account in a manner similar to, but not usually quite as harsh as, mechanistic evolution.

Third, there is the theory of multiple creations, sometimes called progressive creation, which holds that there was a creation, or multiple creations, over a period of millions of years. These successive creations are usually thought of as being destroyed catastrophically, creating the geologic column. Then a few thousand years ago, there was a special creation and a fall, whose details more or less fit those of Genesis 1-3. From the multiple creations viewpoint the Flood is usually interpreted as a local phenomenon that did not leave any unequivocal traces.

Finally there is special creationism,1
1The term originally came from the idea that God created each individual species as it is today, but by now has evolved (pardon the expression) into the definition given in the text, with the term "special" becoming less technical and more an expression of a specific unique supernatural intervention.
which holds that God created a chaotic world and then created life on it for the first time a few thousand years ago. In this view the geologic column (more properly the Phanerozoic to at least the Cretaceous and possibly to the Pleistocene) is the result of the Biblical Flood. What happened before creation week is not completely agreed upon. Our planet may have been created on the first day, or it may have been in a chaotic state for millions of years before creation week. The same holds true for the stars outside our solar [113] system. From the point of view of modern physics, this distinction may not matter, as the aging of the universe without observers is in one sense irrelevant. 2
2 As a logical option, the day-age theory can be ignored. It is neither Biblical (the days in Genesis 1 had an evening and a morning) nor is it adequate to explain the geologic column (for example, the reptiles of the 6th day precede the birds and whales of the 5th day in the geologic column). It is a hybrid born of desperation.

All of the above theories accept minor evolutionary changes (sometimes called microevolution) today. Only mechanistic evolution and one brand of theistic evolution wholeheartedly accept large evolutionary changes (macroevolution). Nobody knows the precise line to draw between microevolution and macroevolution, except that macroevolution would bridge the gap between the phyla and between the classes, and between other animals and humans, while microevolution does not. Therefore differentiation between these theories on the basis of macroevolution is not as useful as it would seem at first.

The first alternative, mechanistic evolution, appears at this point highly unlikely to be correct. As we have noted in chapter 2, mechanistic evolution has no explanation for the origin of life. The evidence we have indicates that this problem is becoming more acute rather than less so.

There are a number of other criticisms of "evolution" which are actually criticisms of mechanistic evolution. For example, there is the problem of the "missing link". For almost all phyla the problem is actually a missing chain—all the links are missing. Stephen Jay Gould’s "hopeful monsters" might as well be relabeled "miracles". One of the few testable predictions Darwin made was that intermediate forms would be found as the geologic record was more fully examined. At present this prediction appears to be dead wrong. Then there is the problem of evolving complex structures like the eye—and not just once, but twice (squids, octopi, etc., and vertebrates). Such problems make it reasonable to discontinue consideration of mechanistic evolution until more evidence compels its re-evaluation. This is in spite of the fact that it and special creationism are the two most satisfying positions from a theoretical point of view (the most elegant).

But none of the foregoing objections touch either theistic evolution or multiple creations. For if God was there to help the process, in whatever way He did it, then the fact that it was a miracle would not prove theistic evolution or multiple creations wrong, and in fact is not unexpected. Hence the above arguments fall well short of proving special creation. Therefore we must use [114] another approach to decide which scenario is most likely.

We should now note the major advantages and disadvantages of the three remaining groups of theories relative to each other. Theistic evolution has the advantage of not having to challenge the scientific evidence for long ages of earth’s history. It is also able to incorporate any evidence for macroevolution, and yet is not mechanically dependent on macroevolution. However, it must deny the historicity of the Genesis creation and flood accounts (and thus must assume that Jesus’ theology was incorrect, which is difficult if Jesus was really the Messiah). It is also basically an ad hoc theory. It can accommodate almost any evidence, which from a theoretical point of view is a disadvantage. One would prefer a theory which had more predictive power.

On the other hand, special creation has a great deal of predictive power, and allows for the historicity of the Genesis creation and flood accounts, making it an elegant theory, on a par with mechanistic evolution. However, it must deny the validity of the standard arguments for the existence of life on the earth for millions of years (it bears repeating that special creation may allow for our planet, and the stars, to be in this age range. It is only when unmistakable fossils exist, usually believed to be in the Cambrian, that the difficulties become acute).

The multiple creations model is a compromise. It allows life on the earth to be as old as usually believed, while at the same time being more or less faithful to the Genesis creation account. This has some theological advantages; it can allow Jesus’ theology, and Paul’s theology, to be more accurate than can theistic evolution, as Jesus’ theology makes use of the Genesis account of creation, and Paul’s theology depends on the Fall.

However, the multiple creations model is unable to find any traces of a Flood in the geologic record. It is thus in the awkward position of insisting on the basic historicity of Creation and the basic nonhistoricity of the Flood.3
3Remember that according to the Genesis account the flood is supposed to have lasted for a year. Noah is supposed to have built a large boat in preparation and supported a veritable zoo. Perhaps most striking, the boat is said to have landed in the mountains of Ararat. This implies an event of global scale. If there are not obvious geological evidences of such an event, then it is wildly exaggerated if not fictitious. Theology based on a non-existent event is baseless.
The multiple creations model therefore does not completely exonerate Jesus’ theology, as Jesus also makes use of the Flood as a parallel to the time of His second [115] coming. The insistence on a Creation and a Fall for what are essentially theological reasons while denying another theologically significant event in the same document because this time it is scientifically testable seems incongruous to me. Thus I see the multiple creations model as theoretically the least satisfying option, an option to be used only if we must eliminate both theistic evolution and special creation from the competition.

So for now we can attempt to choose between theistic evolution and special creation. It should be noted that with the disappearance of mechanistic evolution4
4In fact, even mechanistic evolution may not fit some of the more important qualifications of a scientific theory. Karl Popper had trouble with the scientific aspect of mechanistic evolution as noted in the following passages:

". . . I intend to argue that the theory of natural selection is not a testable scientific theory, but a metaphysical research program; and although it is no doubt the best at present available, it can perhaps be slightly improved." (Popper KR: Unended Quest: An Intellectual Autobiography. Glasgow: William Collins Sons & Co. Ltd., 1976, p. 151)

"It is metaphysical because it is not testable." (p.171)

". . . it suggests the existence of a mechanism of adaptation, and allows us even to study in detail the mechanism at work. And it is the only theory so far which does this.

"This is, of course, the reason why Darwinism has been almost universally accepted. Its theory of adaptation was the first nontheistic one that was convincing; and theism was worse than an open admission of failure, for it created the impression that an ultimate explanation had been reached.

"Now to the degree that Darwinism created the same impression, it is not so much better than the theistic view of adaptation; it is therefore important to show that Darwinism is not a scientific theory, but metaphysical. . . ." (p. 172)

Popper did see one prediction, and therefore a possible scientific test of evolution: "Gradualness is thus, from a logical point of view the central prediction of the theory. (It seems to me that it is its only prediction.)" (p. 172)

Thomas Kuhn also noted that it is difficult to derive testable conclusions from mechanistic evolution. When discussing his differences with Karl Popper, he noted that a modified Popperian approach might be that "For a field to be a science its conclusions must be logically derivable from shared premises. . . . But in this form, at least, it is not even quite a sufficient condition [for a field to be a science], and it is surely not a necessary one. It would, for example, admit surveying and navigation as sciences, and it would bar taxonomy, historical geology, and the theory of evolution. The conclusions of a science may be both precise and binding without being fully derivable by logic from accepted premises." (The Essential Tension, Chicago, University of Chicago Press, 1977, p. 250, n 21, italics his)

this is no longer a choice between science and religion (it does, of course, have scientific and religious repercussions). Both theories postulate a God Who intervenes. Both theories claim that they can explain the scientific evidence if given enough research. What this dispute is actually [116] over is history. What matters is not what should happen, or what could happen, but what did happen.

Once the dispute is seen in this way, two considerations come to the fore. First, the early Genesis account taken at face value purports to be historical, and therefore should not be ignored until it has been shown to be false. Special creation gains an edge (possibly a slight edge, but an edge) in the discussion, since there are no early historical documents supporting theistic evolution. Second, chronology is the backbone of history. And the essential difference between the two theories (other than their theological implications) is time. In this case absolute physical and chemical dating methods (primarily radiometric dating) are the backbone of chronology from a theistic evolutionary point of view (or any other point of view espousing a long age of life on earth). So it becomes incumbent on us to examine the reliability of these physical and chemical dating methods. We may wish to avoid getting into nitpicking details, but if we are to be honest and careful we really have little choice.

There are several books on dating methods. Perhaps one of the better ones for an initial survey is Absolute Dating Methods by Mebus A. Geyh and Helmut Schleicher.5
5Berlin: Springer-Verlag, 1990, hereinafter cited as Geyh and Schleicher. This book has an excellent bibliography in the back.
This book lists 76 physical and chemical methods used to date the earth, the moon, meteorites, or fragments thereof. The list at first seems overwhelming. But we are not looking for pat answers. And so perhaps the best way to begin is at the beginning. Potassium/argon dating is listed first, and is often considered the most reliable dating method demonstrating a long chronology, so we will begin there.

Potassium/argon dating

Potassium/Argon dating is one kind of radiometric dating (dating using radioactive material). Radioactive materials, like the starting materials of many other physical and chemical processes, transform in proportion to time and the amount present at the beginning.6
6 Some readers may find this introductory discussion unnecessary. Others, however, will find it too brief. Those who do may consult standard physics, chemistry, and calculus texts and/or the introduction to a geochronology text. Two good geochronology texts are Dalrymple GB, Lanphere MA: Potassium-Argon Dating: Principles, Techniques, and Applications to Geochronology. San Francisco: W. H. Freeman, 1969, hereinafter cited as Dalrymple and Lanphere, and Faure G: Principles of Isotope Geology (2nd ed). New York: John Wiley and Sons, 1986, hereinafter cited as Faure.

If you start with an amount, say 10 kilograms or 22 pounds, of an unstable substance (we will call it substance A), at the end of a specified time, say 1 year, you would have only part of it left, for example 8 kilograms. In this case, if you started out with 5 kilograms instead of 10, you would have 4 kilograms at the end of 1 year. The general formula would be A1 year = ABeginning 0.8, for 1 year’s wait (in line with conventional usage we will use A0 for ABeginning). But if you started with 10 kilograms and waited 2 years, you would not have 6 kilograms. At the end of the first year you would have 8 kilograms. This 8 kilograms becomes your starting point for the second year, and the amount at the end of the second year is 8 0.8, or 6.4 kilograms. By the same token at the end of 1/2 year, the amount will not be 9 kilograms but slightly less than 9 kilograms. This makes the formula A1 year =A0 0. 8 awkward to use.

A more convenient set of formulas are ln (A0 /A) = kt and (essentially the same formula) A = A0 e—kt. (For those whose eyes glaze over at the mere mention of calculus, it may be of some help to note that only standard formulas are used in this text. These formulas are included for the benefit of readers who want to go into the subject more thoroughly).7
7These formulas can be derived by using a very small time interval (theoretically infinitely small) dt and writing –dA/dt = kA or kA. That is, the amount lost in a very small amount of time (therefore the minus sign) is directly proportional to (= a constant k times) the amount at that time (in our original example k would be 0.22314/year. This means that during a very small amount of time, say 1/100,000 of a year or 5.259 minutes, our 10 kilograms would lose 0.000022314 kilograms). This can be rewritten dA/A = –k dt. This formula can be integrated to yield loge A – loge A0 = –k(t – t0) = –kt (we will define t0 = 0), or ln (A0/A) = kt (t is now the time from the beginning of the period). Taking the exponential of both sides we have A0/A = ekt or A/A0 = e-kt or A = A0 e-kt. Many texts use the Greek letter instead of k for the time constant.

Most readers are acquainted with logarithms to base 10. Some may not be familiar with logarithms to base e. Logarithms to base e

(e = 1 + 1/1 + 1/[12] + 1/[123] + 1/[1234] + . . . = 2.718281 . . .)

are slightly more difficult for common usage, but the natural result of calculus (thus the name natural logarithm and the abbreviation ln), simpler theoretically, and much easier for computers. We may compute
ex = 1 + x/1 + x2/[12] + x3/[123] + x4/[1234] + . . .

If 0 < x < 2 we may compute

ln x (= loge x) = (x-1) – (x-1)2/2 + (x-1)3/3 – (x-1)4/4 + . . .,

whereas if x > 1 we may find ln x by using ln x = –ln (1/x). There is no such simple formula for log10 x or 10x. The two systems are related by the formulas log10 x = ln (x) / ln (10) and 10x = ex ln l0.

We will see these formulas and variants again and again. They can be graphically represented by the following:

Note that the inverse of the constant k gives the time at which all of the substance would have transformed if it had kept up its initial rate of transformation. This is sometimes called the mean life. There is another constant, the half life (or t1/2), which is the time at which half of the material is gone. It is equal to ln 2 (= 0.693147 . . . ) times 1/k. Also note that the amount of material transformed at a given time (the daughter product D) can be found by the formula D = A0 (1 – e–kt). If one knows the starting amount A0 one can find the time needed to leave only a given amount of unchanged A by the formula t = (ln [A0/A])/k. If A0 is not known one can calculate it by the formula A0 = A + D. But this is only valid if there was no D present at the beginning and there has been no D or A gained or lost since the beginning (other than by spontaneous transformation of A to D). If there has been D present at the beginning (D0), or if D has been added (DA) or lost (DL) since the beginning, then the formula for finding the D formed from A, D*, is D* = D – D0 – DA + DL, and A0 = A + D*. If there has been gain or loss of A (other than spontaneous transformation) since the beginning, there is no easy universal formula for correcting the time estimate for such gains or losses. We shall find these conditions particularly important. [119]

We will now review the theory behind radioactivity. Atoms are nearly digital entities. That is, each atom has a whole number of protons, neutrons, and electrons, and its weight (or more properly, mass) is equal to their combined masses minus a very small mass called the binding energy. An electron has very small mass compared to a proton or a neutron (which have roughly equal mass), and for most purposes its mass can be ignored. This means that the mass of each atom is a nearly perfect function of the number of protons and neutrons in its nucleus. The number of protons in an atom determines the number of electrons it has when electrically neutral, and thus almost all of its chemical properties. All atoms of a particular element have the same number of protons. Thus even though technically it would be proper to write 1H for hydrogen and 2He for helium, the abbreviations H and He already contain the information in the subscript and it is not necessary to do so. However, the number of neutrons in hydrogen is not specified by the symbol H, and so to distinguish the different kinds of hydrogen we write their digital mass (protons + neutrons, nucleons) either in the upper left or the upper right corner. Thus deuterium, hydrogen with one neutron and one proton (two nucleons), is written 2H or H2 (the former is more common). Hydrogen without any neutrons is written 1H, and hydrogen with two neutrons (tritium) is 3H.

Some isotopes are unstable and spontaneously break down, or transform, into other elements. Thus 3H will eject an electron and turn into 3He (with two protons and a neutron). This transformation process is called radioactivity. There are 4 different major (for our purposes) kinds of radioactivity. First, a nucleus can eject an alpha particle, or 4He nucleus, and thus lose 2 neutrons and 2 protons. This happens mainly with nuclei that are too big to be stable. Second, it can eject an electron, as 3H does, and turn a neutron into a proton. This happens mainly with nuclei that have too high a proportion of neutrons. Third, it can eject a positron (a positive electron) which then annihilates an electron, sending 2 gamma rays (an electromagnetic radiation, related to light but more energetic than x-rays) in opposite directions. In what gives the same final result, except for a different kind and amount of gamma rays (or x-rays), it can capture one of the electrons orbiting it. This is called K-capture. Both of these processes turn a proton into a neutron, and happen to nuclei which have too high a proportion of protons. Fourth, certain nuclei which [120] are too heavy will spontaneously split into 2 comparable (usually not equal) halves, along with usually a few leftover neutrons. This is called fission. (In addition, a nucleus can be made in an excited state which emits a gamma ray, or in some cases gamma rays, and thereby loses a very small amount of mass. This does not affect the number of protons or neutrons and so will not be further considered here.) In all these cases the mass of the end products is slightly less than that of the starting material. The excess energy is transformed either into gamma rays or into motion of the end products. For example, 3H weighs slightly more than 3He.

What governs which atoms are stable and which are unstable (and how unstable they are) is not completely understood, and the part that is understood is complicated to explain. Perhaps the only additional observation we should make here is that nuclei seem to prefer to have an even number of protons and an even number of neutrons. Thus potassium-40, or 40K,8
8The abbreviation is for Kalium.
is unstable, even though it has a good balance of neutrons (21) and protons (19), because there are odd numbers of both. It will decay to either 40Ca (calcium) or 40Ar (argon).

One of the unusual things about radioactivity is that except for K-capture, which is very slightly influenced by the chemical environment and pressure,9
9Apparently because pressure creates a higher density of electrons, particularly K-electrons, near (actually in) the nucleus. The deviation in the half-life is 0.6% for 7Be at 270 kbar (Hensley WK, Bassett WA, Huizenga JR: “Pressure dependence of the radioactive decay constant of beryllium-7.” Science 1973;181:1164-5). This is far too small to account for the discrepancy between the time frames under discussion. A change in chemical environment makes an even smaller difference in the half-life (<0.2% in the case of beryllium-7, the most highly influenced isotope). There are theoretical reasons for expecting a slight effect of chemical environnment on other isotopes, but the expected effect is so small that if it exists we are unable to measure it. See Dalrymple GB: The Age of the Earth. Stanford: Stanford University Press, 1991, pp. 86-90 for a good summary of the available experimental evidence and theory.
10Via K-electron capture with a gamma ray (11.0%), K-capture without a gamma ray (0.16%), or positron emission (0.001%). The mechanism does not really matter for our purposes.
the rate of decay (the constant k) is not measurably influenced by any known environmental factor. Neither temperature, electric or magnetic field strength, light, x-rays, nor any other variable is known to influence the rate of decay. This makes radioactive decay the best physical or chemical method of measuring time. [121]

Potassium has at present a uniform mixture of 399K (93.2581%) and 41K (6.7032%), both of which are stable, and 40K (0.01167%). This ratio has been the same wherever it has been measured. As noted before, 40K is radioactive. Its decay constant is 5.543 10-10 /year, which corresponds to a half life of 1.250 109 years. It decays to either 40Ca (88.8%) via beta decay, or to 40Ar (11.2%).10 The ratio of production of 40Ar to 40Ca is called the branching ratio. The radiogenic 40Ca is hard to distinguish from 40Ca already in the environment. (The distinction can sometimes be made. We will come to that later.) But radiogenic 40Ar can be distinguished from atmospheric argon (about 1% of air is argon) by the presence of 36Ar (0.337%) and 38Ar (0.063%) in atmospheric argon (which leaves 40Ar at 99.600% and a 40Ar/36Ar ratio of 295.5 to 1). This makes it possible to devise a dating method which is valid if the following assumptions are satisfied:

1. The rate of decay, and the branching ratio, of 40K have not changed.
2. The material in question lost all its argon at an identifiable time t0.
3. No argon has been lost since time t0.
4. No argon except atmospheric argon, with today’s 40Ar/36Ar ratio, has been gained since time t0.
5. No potassium has been gained or lost since time t0, except by decay.
6. The ratio of 40K to total K is constant.
7. The total K, 40Ar, and 40Ar in the material in question can all be measured accurately.

For situations in which these assumptions are satisfied, we may derive a standard formula for potassium/argon dating:

t = ln (40K0/40K) / k (Assumptions 1a,5)
= ln [(40K + 40Ar* + 40Ca*) / 40K] / k (Decay products)
= ln [1 + (40Ar* + 40Ca*) / 40K] / k (Algebra)
= ln [1 + (40Ar* / 0.112) / 40K] / k (Assumption 1b)
= ln [1 + ([40Ar - (36Ar 295.5)] / [0.112 40K])] / k (Assumptions 2,3,4)[122]
t = {[ln (1 +[40Ar – (36Ar 295.5)]/[0.112 K 0.0001167])] / (5.543 10-10)} years (Assumptions 1a,6)
According to assumption 7 we can measure total K, 40Ar, and 36Ar. This formula uses units of moles per gram of sample. A slight correction is necessary if units of weight (mass) are to be used.

Potassium/argon dating has been used extensively, so there is a large amount of evidence regarding its fit with evolutionary11
11We will use the term evolution, rather than theistic evolution. This is for brevity, to avoid awkward phrases, and because the time scale is common to all theories of evolution. We will also use the term creationist instead of special creationist for brevity and to avoid awkward phrases, even though theistic evolutionists are technically creationists.
12For example, Evernden JF, Savage DE, Curtis GH, James GT: “Potassium-argon dating and the Cenozoic mammalian chronology of North America.” Am J Sci 1964;22:145-98. For evidence of their selectivity, see their discussion on pp. 171-4 of why all but one potassium/argon date for the Rusinga Island biotites was discarded. Then note their continued apparently uncritical use of biotite in other areas where the dates obtained matched their expectations. Note also that “Unfortunately many of the samples that passed field inspection for suitability and were laboriously collected, later proved unsuitable for dating. . . . Thus, of some 65 samples collected by M. Skinner only 10 could be used.” (p. 174) It might have been interesting to know why such samples proved unsuitable for dating, and what their potassium/argon dates were.

It is interesting to speculate what would happen if an article in chemistry or medicine were submitted with perhaps 1/6 of the data reported. It is difficult for me to believe that the article in question would have become a classic, as the article by Evernden et al. apparently has.

In point of fact, the selectivity in this article may be even greater than noted above. Sometimes the whole rock basalt date is reported, and sometimes a mineral fraction from the basalt is dated instead, such as biotite or sanidine. Why one type of date is used at one time and not at another is not specified. If there are 3 mineral fractions per basalt sample, there are 4 different possible dates for that sample. Thus one could pick the dates that fit one’s expectations and create a very impressive list of dates with close agreement without there being more than a general correlation of most dates with one’s expectations.

13See Geyh and Schleicher, p. 374 chart.
14This does not mean that all statements are presumed false until proven true. Statements whose basis can reasonably be believed to be experimental have some weight. Statements which are based on theory which is not being challenged have some weight. But statements which depend on the theories which are being evaluated cannot themselves support those same theories. That would be circular reasoning. These statements can only be used to help determine the internal consistency of a theory, or to suggest plausibility to the one who made the statement.
theory. It actually fits fairly well. Some studies give the impression that it fits perfectly, but such studies often use filtered data (that is, the data that fit best).12 The boundaries of the geological time periods have been moved to fit potassium/argon dating.13 And many minerals are not felt to be suitable for analysis; they do not give the expected dates. For these reasons the fit is not quite as good as might be thought. However, for [123] certain minerals the fit is quite good. Any creationist explanation of potassium/argon dating must account for its relatively good accord with the evolutionary time frame.

From an evolutionary perspective biotite and hornblende give the best dates. Dates on hornblende are most often in accord with the evolutionary time scale, but biotite is more widespread and retains its potassium/argon age under fairly severe weathering conditions. Many other minerals such as sanidine, anorthoclase, plagioclase, leucite, nepheline, muscovite, phlogopite, and lepidolite (all igneous and/or metamorphic minerals) can be dated by the potassium/argon method. Whole rock basalt (lava) can also be used. Only one sedimentary rock, glauconite, can be dated by this method and the results are not always considered to be reliable. Several sedimentary rocks which contain large amounts of potassium, particularly sylvite (KCl), which is over 50% potassium by weight, are not considered satisfactory.

The extensive use of potassium/argon dating also provides a fair amount of evidence bearing on the underlying assumptions. We should turn our attention to those assumptions now. In doing so, we should keep in mind several considerations. First, when we read any statement, we should ask, “How does the author know?” Statements without adequate evidence should not be determinative in our inquiry.14 There is a place for a certain kind of scientific skepticism. Second, we should not assume the evolutionary time scale when evaluating potassium/argon dating. Since the explicit purpose of our inquiry is to evaluate whether potassium/argon dating supports the evolutionary time scale, it would be circular reasoning to assume the evolutionary time scale at the outset of our inquiry. On the other hand, we will not assume a creationist time scale either (However, we may use evolutionist or creationist assumptions as limiting cases). Finally, we cannot use another dating method to calibrate potassium/argon dating until we have examined the other method and established [124] its validity. At present we have not done so for any radiometric dating methods. For now, correlations with other methods will not be used unless both evolutionists and creationists agree on their validity.

We now turn our attention to the underlying assumptions. The last assumption, number 7, is one of the safest. Measurements of potassium that have been made in different laboratories, and with different methods, are repeatedly in agreement to within experimental error. The isotope dilution method of measuring argon has a firm theoretical basis, and in appropriate specimens it yields results which match those obtained from volumetric and neutron activation analyses. The limitations in the accuracy of the various methods of measurement are fairly well-understood. We can accept the the “raw” data as basically accurate.

Assumption 6 is similarly secure. The isotopic composition of potassium from many sources has been measured, and the results are always essentially the same. Natural isotopic enrichment effects can be safely ignored.

Assumption 5 is fairly safe. In most situations where potassium has been either gained or lost from a mineral, the mineral has been noticeably altered (it would be difficult to do this without affecting the argon to an even greater extent). Replacement of the potassium in a rock with potassium from other sources, so long as the isotopic concentration is not significantly altered, would have no effect on the apparent age derived by the above formula. And if there were a problem of this nature, to effect the changes needed to explain the time difference between evolutionary and creationist models (up to 5 orders of magnitude), a creationist would need up to 5 orders of magnitude increase in the initial potassium content of our specimens, a physical impossibility (it would require more than 100% potassium). Nor can isotopic enrichment and then depletion effects bridge this gap in any reasonable manner.

Assumption 4 is probably satisfied for most samples. It would only be incorrect for materials which are heated in the presence of argon from the earth’s mantle, which apparently contains almost entirely 40Ar, or perhaps in primordial argon, which may have had a higher concentration of 36Ar than the present atmosphere. We will tentatively accept it, keeping in mind that it may be challenged. [125]

Assumption 3 is fairly commonly violated, according to most texts on potassium/argon dating. That is, according to the standard interpretation of potassium/argon dating, many rocks lose argon. Specifically, most sedimentary rocks are supposed to lose argon because their crystal structure cannot retain it. Glauconites appear to be the only sedimentary minerals from which an appropriate age (from the evolutionary perspective) can (sometimes) be obtained. Certain minerals such as sylvite appear to lose argon in recrystallization (or perhaps cannot retain argon); at least their ages are consistently much too young for the evolutionary time scale. Rocks that have been heated after formation can be demonstrated to have younger potassium/argon ages than similar rocks from the same formation which have not been heated. Several processes are listed in standard texts as explanations for this argon loss, such as metamorphism, weathering, and reheating. We will return to this assumption later. We will only note for now that the violation of this assumption would cause the rock to date younger than its age of formation. Depending on the loss of argon, this date could be as low as recent (< 5000 years).

Assumption 1 is often challenged by some creationists. They reason that radioactivity could have speeded up during the Flood, possibly providing a contributory cause of the Flood, and producing erroneously high apparent ages. For every order of magnitude that one increases the decay rate, one increases the apparent age of the rock by the same order of magnitude. The relationship is mathematically perfect. The only way to tell that anything unusual took place is to note whether daughter products have escaped as expected. There is some evidence which can be interpreted as a disequilibrium of helium and of argon.

The major problem with this creationist view is the absence of a mechanism to explain or predict the change in the half life. Theoretically radioactive decay could be caused by some mechanism such as neutrinos, rather than being random from the point of view of the atom, but no evidence of a decrease in any half life has been noted during recent supernova explosions, for instance.15 [126]
15A minor problem is determining which radioactive decay processes are affected. Presumably it would have to be all of them, or all of one kind, or one particular isotope, or else this creationist hypothesis is just another ad hoc hypothesis.

A systematic change in radioactive time constants is still a theoretical possibility, but until there is direct evidence for it we will use it as a last resort, only if radiometric dating is otherwise secure, but compelling non-radiometric evidence requires a short span for the history of life on this earth.

Assumption 2 sounds logical at first, and is usually stated in texts as self-evident.16
16For example, Geyh and Schleicher, p. 56: “What is special about the K-Ar method is that the daughter nuclide is a noble gas, which is normally not incorporated into minerals and is not bound in the mineral in which it is found.” Dalrymple and Lanphere state on p. 46: “. . . a silicate melt will not usually retain the 40Ar that is produced, and thus the potassium-argon clock is not “set” until the mineral solidifies and cools sufficiently to allow the 40Ar to accumulate in the mineral lattice.” Dalrymple (see note 9) states on p. 91, “The K-Ar method is the only decay scheme that can be used with little or no concern for the initial presence of the daughter isotope. This is because 40Ar is an inert gas that does not combine chemically with any other element and so escapes easily from rocks when they are heated. Thus, while a rock is molten the 40Ar formed by the decay of escapes from the liquid.”
17 Karpinskaya TB, Ostrovskiy IA, Shanin LL: “Synthetic introduction of argon into mica at high pressures and temperatures.” Isv Akad Nauk S. S. S. R. Geol Ser 1961;8:87-9.
18 Karpinskaya TB: Synthesis of argon muscovite.” Internat Geol Rev 1967;9:1493-5. This is approximately 2,500 times as much argon as is naturally found in the usual muscovite, and it is mostly liberated again at over 300° C. A linear interpolation would seem to indicate that the usual potassium/argon dates could be obtained with 40Ar partial pressures of as little as 2 atmospheres. I have found one reference on the introduction of argon into glass, Roy DM, Faile SP, Tuttle OF: “Effect of large concentrations of dissolved gas on properties of glasses.” Phys and Chem of Glasses 1964;5:176-7. The argon introduced (under 1/2 to 10 kbar) was not quantified, but was noted to be dissolved rather than in bubbles. The alert reader may wonder why I have not cited data for biotite or hornblende. The reason is because I am not aware of any such data. All the experiments on potassium-bearing minerals I have found in the literature are cited in this and the previous note.
19Dalrymple GB: “40Ar/36Ar analysis of historic lava flows.” Earth Planet Sci Lett 1969;6:47-55.
20Dalrymple, see note 19, citing Funkhouser JG: “The determination of a series of ages of a Hawaiian volcano by the potassium-argon method”. Univ of Hawaii Ph.D. thesis, 1966. Dalrymple’s citation is accurate. For those who are going into the subject in depth I recommend the thesis.
21A plagioclase phenocryst from Surtsey that was 1 cm in diameter gave an 40Ar/36Ar ratio of 298.9, which was not statistically different from the value of 296.1 which Dalrymple’ mass spectrometer gave for air. However the Mt. Etna 1792 basalt and the Mt. Lassen plagioclase both dated high, and both had large phenocrysts but no xenocrysts. In addition, Dalrymple cited the work of Damon et al. (Damon PE, Laughlin AW, Percious JK: “Problem of excess argon-40 in volcanic rocks.” In: Radioactive dating methods and Low-level counting. Vienna: IAEA, 1967, pp. 463-481). Damon et al. cited several instances of phenocrysts with potassium/argon ages of 1 to 7 million years over that of the whole rock, and one potassium/argon date on olivine phenocrysts of greater than 110 million years in a recent (<13,000 year old) basalt. They also state that “Coarse [phenocryst] minerals (x > 1 mm) may take more than 100 years to completely degas at lava temperatures.” (p. 478) Unfortunately, they do not give the evidence for this statement.
22Dalrymple GB, Moore JG: “Argon 40: Excess in submarine pillow basalts from Kilauea Volcano, Hawaii.” Science 1968;161:1132-5. These basalts were 60-90% glass, with phenocrysts. See also Noble CS, Naughton JJ: “Deep-ocean basalts: Inert gas content and uncertainties in age dating.” Science 1968;162:265-7, where the basalts dated up to 21 million years old, and also retained helium.
23There was 1.7-123 million years’ worth of 40Ar found in the lava if one ignores the 36Ar.
But it is one of the few testable assumptions (along with assumption 6 and 7), and so it should be checked.

I am aware of very few direct experiments in which rocks are heated to see if the argon is all driven off under realistic geologic conditions to reset the potassium/argon clock. Of course, rocks are heated routinely in a vacuum to drive off their argon for measurement. But no one would argue that the rocks in an igneous intrusion, for example, were intruded under vacuum conditions.

In one series of experiments, muscovite was heated to 740° to 860° C under high argon pressures (2,800 to 5,000 atmospheres) for periods of 3 to 10.5 hours. The muscovite absorbed significant quantities of argon (producing potassium/argon ages of up to 5 billion years), and the absorbed argon appeared like ordinary “radiogenic” argon.17 In another series of experiments, muscovite was synthesized from a colloidal gel under similar argon pressures and temperatures. The muscovite synthesized in this way contained up to 0.5% argon by weight!18 These experiments show that under [127] certain conditions argon can be incorporated into rocks that we are told are supposed to exclude argon when they crystallize. This makes me uncomfortable accepting assumption 2 without further evidence. One might even argue that minerals should not lose argon without someplace for it to go. But such conditions are not likely to be realistic geologic conditions either.

Perhaps the best way to test assumption 2 is to find formations that everyone can agree were formed within the last 5 to 10 thousand years, date them, and see if they date to essentially zero. This has been done by Dalrymple.19 He dated several lava flows which are known to have erupted in modern times. Most of the lava flows had essentially zero potassium/argon ages. However, about 1/5 of the flows had excess ages. The flows that dated oldest all had ultramafic xenoliths and xenocrysts (small rocks and crystals of foreign material) mixed into the lava. The excess argon, and the extra apparent age, was attributed to these foreign materials, which themselves could date over 1 billion years old.20 Doubt was also expressed about the resetting of phenocrysts (crystals which apparently crystallized from the lava itself), although all the lavas dated had phenocrysts, and some phenocrysts had only argon whose isotopic composition matched that of air.21 From this Dalrymple concluded that basalt can have its [128] potassium/argon clock reset, but this is reliable only if there are no xenocrysts or xenoliths in the basalt. The xenocrysts apparently can retain most of their argon even when heated to the temperature of molten lava. Furthermore, tests on basalt which flowed into the ocean showed that although the lava which hardened above the water dated to essentially zero age, basalt which cooled under the water could date as high as 43 million years old.22 This would of course be relevant for a creationist who believes that the world was covered with water, ocean water to be specific, during much of the Flood. Certainly if one is to avoid obviously erroneous dates in basalt, one will avoid pillow lava.

But there was another phenomenon which was noted in Dalrymple’s article. Some modern lavas had 40Ar/36Ar ratios of less than 295.5. According to a straightforward interpretation of potassium/argon dating, this should be impossible. Dalrymple was not willing to write these ratios off to experimental error. Thus the straightforward interpretation has to at least be modified.

Dalrymple suggested two possible explanations for the excess 36Ar (He rejected the possibility of significant 36Ar formation in situ from nuclear reactions). The kinder one (from an evolutionary point of view) was that when argon from the air diffused back into the lava,23 36Ar diffused in preferentially. But this would mean that the “zero age” lavas actually had an apparent age, and that most lavas do not degas upon eruption. In fact, depending on how strong is the preference for 36Ar diffusion, it could even be that all lavas do not completely degas. [129]

His other explanation was that the lavas with the anomalously high 36Ar actually came from an area of the mantle that had primordial argon which had not been diluted with radiogenic 40Ar and had not completely degassed. But this means that there is no reason to assume that lava whose argon matches that of the air has degassed either. It may have simply started with argon which matched air argon.

Thus the evidence is that lava does not completely degas on eruption. The precise amount of gas lost cannot be easily quantified using the data we have on hand. It would be very helpful to expose hot lava with a known argon content to 38Ar or 39Ar to see how much argon actually is lost and/or gained and how fast, and what its isotopic composition is.

When we turn to how basalt is dated in the geologic column, we find statements like “basaltic glass, in contrast to acid glass, has a very poor argon retentivity and is unsuitable for K/Ar dating.”24
24Geyh and Schleicher, p. 61.
25Mankinen EA, Dalrymple GB: “Electron microprobe evaluation of terrestial basalts for whole-rock K-Ar dating.” Earth Planet Sci Lett 1972;17:89-94. In one case the glass in question was unaltered, and still gave a potassium/ argon age of 1.6 million years rather than 7.4 million years. These are still not creationist dates, but if lava does not routinely degas, they are easily explainable from a creationist perspective.
26Fechtig H, Kalbitzer S: “The diffusion of argon in potassium-bearing solids.” In Schaeffer GA, Zähringer J (eds): Potassium-Argon Dating. New York: Springer-Verlag, 1966, pp. 68-103. It is worth quoting p. 101: “This section concludes that diffusion at room temperature is always so small that no appreciable argon losses occur.
Mankinen and Dalrymple25 noted that two basalts containing glass dated much younger than expected whereas the phenocrysts in one of those basalts gave the expected dates. They concluded that basalts containing glass should be rejected and in the latter case accepted the phenocryst age. This would seem to indicate that the workers in the field trust old samples that they would be reluctant to trust if they were recent, and vice versa. This is particularly striking in view of the experimental evidence that argon diffusion in glass is negligible under ordinary geological conditions.26

It would seem that at least the data tends to undermine the validity of the potassium/argon dating of basalt. It could even suggest that the conventional time scale is incorrect. Perhaps these basaltic glasses don’t lose argon. Perhaps they simply were [130] more completely degassed at the time of the eruption and the basalt is really as young as or younger than the indicated age. Certainly no potassium/argon date for basalt should be accepted as secure until we know whether the basalt matches the characteristics of recent basalt that is consistently dated at zero by the potassium/argon method.

It might be revealing to date recent and geologically old basaltic lava, glass, phenocrysts, and xenocrysts blinded to their geological horizon, and report all the results. This is the procedure that would be done in, for example, a controversial medical research area.

(Some may object to this comparison. However, there are parallels between geology and medicine. Both are not exact sciences in the sense that physics and chemistry are. They both deal with situations with many variables, not all of which can be precisely controlled. Both have a practical aspect—finding oil, and helping patients. And both make use of multiple branches of “basic” sciences.)

Perhaps we can place greater trust in granitic intrusions. There are unfortunately no historically witnessed granitic intrusions which can be used for a baseline. So we really don’t know whether or not granitic intrusions reset their potassium/argon clocks. One hint comes from a granitic xenolith from a pleistocene basalt (conventional age 60,000 years). This xenolith was estimated to have been at 1,100° C during the basaltic lava eruption, and yet its sanidine had a potassium/argon age of 2 million years (the biotite age was not given).27
27 Dalrymple and Lanphere, p. 143, citing Dalrymple GB: “Argon retention in a granitic xenolith from a pleistocene basalt, Sierra Nevada, California.” Nature 1964;201:282. The granite was 10 cm in diameter and 3 m below the surface of the lava.
28See note 20.
29Walter RC, Manega PC, Hay RL, Drake RE, Curtis GH: “Laser-fusion 40Ar/39Ar dating of bed I, Olduvai Gorge, Tanzania.” Nature 1991;354:145-9. The 40Ar/39Ar dating method is a variant of the potassium/argon dating method which uses neutron irradiation of the sample to produce 39Ar from 39K. Note that these results would appear to invalidate tuff dates.
30Pp. 121-144, especially pp. 126-8 table.
31P. 105, citing Ashkinadze GS, Gorokhovskiy BM, Shukolyakov YA: “40Ar/39Ar dating of biotite containing excess 40Ar” Geochem Int 1977;14(3):172-6.
Its original “age” was estimated at 40-92 million years, so it was estimated to have retained 2-5% of its argon. Other xenoliths may have potassium/argon ages of over 1 billion years.28 And a report from Olduvai Gorge indicates that individual biotite crystals in tuff could retain 400-800 million years’ worth of 40Ar.29 Apparently the clocks in granitic xe-[131]noliths can be only partially reset by heating that has usually been assumed to completely reset them.

Several examples of multiple minerals including hornblende and biotite which even evolutionists admit have excess argon can be found in Dalrymple and Lanphere.30 Another particularly obvious example is a biotite cited by Faure31 whose potassium/argon age exceeds the traditional age of the earth!

So we can’t be sure that the clock is fully reset for biotite or other granitic minerals either, and the evidence that does not depend on evolutionary presuppositions is in favor of it not being reset.

Is there a mineral that someone who does not start with evolutionary presuppositions might believe to be completely reset at the time of formation? Yes, there is. Potassium minerals in evaporite deposits should have equilibrated their argon with the atmosphere when they crystallized. One would expect that any argon incorporated into the mineral should have the same isotopic composition as that in air. Sylvite in particular is over 50% potassium by weight, which would make the potassium and argon easier than usual to measure, and can form crystals up to an inch across or larger, which would seem to make it a good candidate for argon retention.

But evolutionists do not use sylvite and similar evaporites, because of the “poor retentivity” of salt minerals, and because they recrystallize below 100° C.32
32Geyh and Schleicher, pp. 61-2. The difficulty with sylvite has been noted since the very first use of potassium/argon dating. See Aldrich LT, Nier AO: “Argon-40 in potassium minerals.” Phys Rev 1948;74:876-7.
How do we know this? Is it because someone has measured the diffusion of argon in sylvite? Or has someone tried to mildly heat or deform the crystals to see if the argon is released? Has someone irradiated sylvite with neutrons to see if 39Ar will diffuse out of the crystal? No, experimental evidence is not the basis for these assertions about retentivity and recrystallization. In fact, the experimental evidence is actually against these assertions.33
33The diffusion of argon from sylvite has actually been measured by some of these methods, and it has turned out to be negligible under geological conditions. See Fechtig and Kalbitzer, note 26.
The reason these assertions [131] are made is because sylvite crystals in particular, and evaporite salts in general, give potassium/argon dates much younger than their evolutionary ages, so they must have lost argon somehow. That is a logical deduction as long as one knows that the evolutionary time scale is largely correct. However, if one is not irreversibly wedded to that time scale, another explanation presents itself. Perhaps the minerals are not really that old. Perhaps there is something wrong with the evolutionary time scale.

And on second thought, the theory that argon diffuses out of sylvite crystals seems contrived. If argon does not diffuse out of biotite, with its loose cleavage planes, why should argon diffuse out of sylvite, which has a close-packed crystal structure? It is of interest that several other minerals “lose argon”, and yet we are told that in another mineral (this time igneous), sanidine, “diffusion of argon is several orders of magnitude faster at low temperatures than extrapolation from high temperature data would indicate.”34
34Geyh and Schleicher, p. 62, citing Marshall BD, Woodard HH, DePaolo DJ: “K-Ca-Ar systematics of authigenic sanidine from Waukau, Wisconsin, and the diffusivity of argon.” Geology 1986;14:936-8. The potassium/argon age in this paper was up to 75 million years less than the stratigraphic age (>454 million years).
And we read that

Initially, it was hoped that these experiments [determining argon diffusion characteristics of minerals] would lead to a classification of these minerals according to their ability to retain argon. In addition, it was thought that experimentally determined diffusion coefficients might provide a way to correct “apparent” ages for argon loss and to provide a basis for using argon loss to determine the exact geologic conditions (heating, burial, and so forth) that caused the loss. Unfortunately, these goals have not been reached. Although the relative ability of most common minerals to retain argon is known, this knowledge has come largely from geologic studies rather than from diffusion experiments.35
35Dalrymple and Lanphere, p.151.

In other words, the experimental evidence is against the diffusion which must have happened if the evolutionary time scale is correct, and so the standard approach has been to ignore the experimental evidence and try to create a scenario compatible with the evolutionary time scale. Now that is fine if you know [133] that the evolutionary time scale is correct. But if we are trying to make an unbiased effort to determine the validity of the evolutionary time scale, the evidence does not appear to support that scale.

I have seen no independent evidence to support the suggestion that mild heating accounts for low potassium/argon dates in sylvite. The best evolutionary theory to explain the evaporite data would seem to be that which has also been advanced to explain the “anomalously” young 87Rb/87Sr and 40K/40Ca ages of the same minerals. The sylvite periodically re-dissolves in water. This does not seem unreasonable. It might be interesting to re-examine these deposits to see whether there is other evidence for recrystallization that would support this repeated solution theory.

From a creationist standpoint the evaporite deposits still present a problem. For if they are truly simple evaporite deposits, the potassium/argon dates would be predicted to be zero, and yet Devonian sylvite deposits (conventionally dated ca. 350 million years old) have potassium/argon dates of around 200 million years. So the question remains, why don’t the deposits date at zero? I can think of two possible explanations which would allow a creationist time frame. First, it is possible that sylvite absorbs argon underground, either directly or as a result of recrystallization. It is of interest that if there is recrystallization, buried sylvite and carnallite can apparently incorporate argon, and radiogenic argon at that, on recrystallization.36
36This is because rubidium/strontium and potassium/calcium dates give a maximum age of 2-100 million years for (Permian) sylvite that dates 200 million years old by the potassium/argon method (Baadsgaard H: “Rb-Sr and K-Ca isotope systematics in minerals from potassium horizons in the Prairie Evaporite Formation, Saskatchewan, Canada.” Chem Geol (Isot Geosci Sect) 1987;66:1-15). Thus even by evolutionary criteria this sylvite occluded over 100 million years’ worth of argon on (re?)crystallization.

Incidentally, this discussion illustrates one of the problems in comparing two general scientific theories (or research programs, as Lakatos would call them). There is often no single piece of evidence that conclusively proves one theory superior to the other. The judgment more often has to be made on the basis of which theory fits the relevant facts best, and that depends on which facts are considered most important and well-established, and how hard we search for facts and theoretical predictions. In this case should we close our investigation on hearing that published potassium-argon dates often match the evolutionary time scale, or on hearing that sylvite dates too low for the evolutionary time scale, or on hearing that sylvite dates do not neatly fit a creationist model, or on hearing that the rubidium-strontium dating evidence suggests the possibility of sylvite occluding argon? In theory, of course, even this is not enough. We shall have to test that possibility experimentally, and the exploration goes on forever. In practice we have to stop at least for now with the observations that have been done and make the best tentative judgment we can on the basis of the available data.

However, I do not think that the process is totally subjective. Eventually we should reach the place where one theory continually runs into problems and the other continually points to new correct observations. When this happens, we can provisionally accept the latter theory, and insist that the former explain parsimoniously the difficulties presented to it before we reconsider our judgment.

This means that we may be required to do a good deal of gruntwork in order to establish whether multiple pieces of evidence fit obviously better into one theory, or whether the evidence is truly equivocal. We may not like it, but intellectual honesty requires a fair treatment of all the evidence.

37Hardie LA: “The roles of rifting and hydrothermal CaCl2 brines in the origin of potash evaporites: An hypothesis.” Am J Science 1990;290:43-106. Hardie argued for most of the evaporite deposits being the result of the evaporation of CaCl2 brines in rift and other extensional fault basins. He cited no modern examples of extensive KCl deposits.
38For example, see the recommendation to get a geochronologist to select pilot samples, which then guide the selection of final samples, in Geyh and Schleicher, p. 7.
It would be interesting to [134] crystallize evaporites under high argon pressures to see how much argon is actually incorporated.

Second, these deposits may not be evaporite deposits at all. It has been strongly (and persuasively) argued37 that whatever else they are, they are not seawater evaporites. The puzzle of their formation has not been solved, and it would seem premature to use them as proof of an old earth until their formation is better understood, although they may be given due weight.

One final point deserves consideration. It is still proper for an evolutionist to point out that potassium/argon dating as currently used matches the evolutionary time scale. A creationist explanation of potassium/argon dating must state not only why the current usage is incorrect but also why the dates at present line up with the evolutionary time scale as well as they do. A start can be made by noting that many dates do not fit, as noted above, and that there is some selectivity in the kinds of samples and specific samples that are dated.38 There is also selectivity in which samples are submitted for publication (no one likes to submit ambiguous or chaotic data), and which are published (review-[135]ers do not like to approve publication of these data, and editors do not like to publish them). Incidentally, these distortions of the data are for the most part done innocently. The same thing happens, for example, in medicine. Positive results are always easier to publish than negative results, and both are easier to publish than chaotic results. However, this selectivity may be somewhat offset by a reverse selectivity which can happen in the publishing of textbooks. That is, the textbooks I have cited may collect the problem dates more than the usual ones (and most of my data is obtained from literature cited by various textbooks).39
39 That is my feeling on reading the texts cited above. However, it is not the usual case in textbooks. Most texts simplify experiments and emphasize the positive, often glossing over problems. The possibility remains that the texts I have cited are also biased in favor of the approved theories, in which case the evolutionary interpretation of potassium/argon dating is in even worse trouble than here portrayed.
There is still a real order to most of the potassium/argon dates which needs explanation.

There are three explanations from a creationist perspective for a gradation of potassium/argon dates from older to younger for rocks without significant differences in real age. First, there may have been gradually decreasing argon concentrations and pressures as time during a Flood passed, perhaps because of gradual degassing of the mantle. Second, rocks later during the Flood may have formed under less (hydrostatic) pressure than those formed earlier. This would allow them to be more thoroughly degassed for the same temperature. Finally, the later rocks may have been more thoroughly melted, and for a longer time period, allowing more inherited argon to escape from the later rocks. Perhaps all three mechanisms were operative to some extent. All these explanations seem plausible, and there is evidence for the first one.40
40Beryl and cordierite contain essentially no potassium, yet “It may be stated that the helium and argon content of beryl and cordierite increases with the age of the mineral and there is no relationship between this phenomenon and the alpha emission, potassium content, chemical composition or mineralogical environment of the mineral.” (Damon PE, Kulp JL: “Excess helium and argon in beryl and other minerals.” Am Min 1958;43:433-59. The quote is from p. 445, italics theirs) This is true especially of 40Ar. This observation would be predicted by a creationist, but I have not seen a good explanation of this from an evolutionary standpoint.
So it would seem that there is a creationist model which is believable, and which has supporting evidence for its explanation of the general trend of potassium/ argon dates in the geological column. [136]

To summarize, modern basaltic lava potassium/argon dates indicate that the current use of potassium/argon dating is probably invalid, and the glass dates at least suggest a shorter chronology. Biotite and other plutonic minerals are consistent with this position, although the supporting data there is not as complete as we would like. The data on sylvite and other evaporite minerals are problematic for both a short chronology and a long chronology, although the problems for a long chronology appear to me to be as great or greater in spite of the greater research effort to solve them from that chronology’s point of view. With the data examined so far, a (special) creationist model for the age of life on the earth provides a more straightforward approach than an evolutionary one.

Home Page
Go to the next part of Chapter 5: The Pentateuch and Joshua