The Radiocarbon Clock Gets a Checkup
AMONG the scientific tools devised to help to satisfy man’s curiosity about his past, none is better known than the radiocarbon clock. This method of dating organic material in ancient artifacts is based on measurement of the radioactive carbon that is formed by cosmic rays in the atmosphere and taken up by plant life. It is most useful for dating things made of wood, charcoal and plant or animal fibers. Its workable range goes back more than 10,000 years.
Archaeologists are keenly interested in the results of such dating, because they study ancient men and their works. Bible students too have been interested in radiocarbon dating, because its range overlaps the 6,000-year history of man recorded in the Bible.
Perhaps you know that the radiocarbon clock was used to date the linen wrapping of the ancient manuscript of Isaiah discovered near the Dead Sea.1a The wrapping was found to be eighteen or twenty centuries old, thus confirming other proofs that the manuscript is genuine, not a clever recent forgery.
Symposium at Uppsala
Interest in radiocarbon dating has been stirred up anew by the recent publication (in 1971) of the proceedings of the Twelfth Nobel Symposium, held in Uppsala, Sweden, in 1969. There the radio-chemistry experts from many countries met together with geologists and archaeologists. They discussed their latest researches into the theory and the practical use of radiocarbon (carbon 14) for dating. The honorary president was Nobel Prize winner W. F. Libby, of the University of California at Los Angeles, who pioneered carbon-14 dating in 1949.
The report of the conference conveys an overall feeling of satisfaction with current successes of the method. Conflicting results, which sometimes came out of different laboratories, have largely been reconciled. An accuracy of within fifty to one hundred years in the date is now expected. It is true that divergences larger than this have been found between the “radiocarbon age,” as calculated from the radioactivity, and the real age of known samples, but this may be taken into account with a calibration curve measured in several laboratories.
This curve is based chiefly on wood taken from long-lived trees that have been dated by counting their annual rings. For example, a piece of wood 7,000 years old according to the ring count may give a radiocarbon age of only 6,000 years. So the 1,000 years is applied as a correction to be added to the radiocarbon age of any sample from that era.
The theory on which the radiocarbon method rests has been found to be much more complex than was expected twenty years ago, and many of the corrections to the theory have been studied to see how they would affect the measured ages. By taking all this into account, it would appear possible to get a fairly exact age of organic material that was formed at any time in the past 7,400 years.
Now there are some samples taken from ancient men’s houses and hearths that, according to the radiocarbon dates, are more than 6,000 years old. Such findings conflict with the Bible chronology, according to which the first man was created only 6,000 years ago. This raises some possibly disturbing questions. Has the increased refinement and apparent success of the radiocarbon clock made the Bible chronology obsolete? Can we still put our faith in the Biblical count of years, or has science shown it to be unreliable?
Before we jump to any conclusion, it would be prudent to look a little more closely into some of the details that were discussed at the Uppsala conference. When we do, we begin to wonder whether the detailed corrections in the theory of radiocarbon dates, which at first appear to make it more exact, do not actually open up more possible ways in which it can be wrong.
The relatively simple theory as it was seen twenty years ago was based on the following assumptions:
(1) That carbon 14, the radioactive component of natural carbon, decays with a half-life of 5,568 years.
(2) That the ratio of carbon-14 atoms to the stable carbon-12 atoms in “live” carbon has always been the same as it is today. This depends on two other assumptions (2a and 2b).
(2a) That the number of carbon-14 atoms has been constant; this means that the cosmic rays that form them must not have varied in the past 15,000 or 20,000 years.
(2b) Also, that the total amount of stable carbon in the “exchange reservoir” has been constant during the same time. This includes the carbon dioxide in the air, as well as the organic carbon in living things, because they are continually taking up carbon dioxide by photosynthesis and releasing it by respiration. Also, carbon dioxide dissolves in seawater, where it forms carbonic acid and carbonate, which becomes mixed with the dissolved carbonate in the ocean. This process also is reversible, although it may take fifty years. Mineral carbonate in the rocks is, of course, not considered to be part of the exchange reservoir.
(2c) Related to number two is the assumption that the production of carbon 14 has continued steady all this time, and this implies that its decay, on a worldwide basis, is in balance with its production.
(3) That any living thing, plant or animal, incorporates radiocarbon in its tissues while it is alive; then, after its death, the activity decreases mathematically according to the natural radioactive decay; it does not pick up radiocarbon through contact with younger materials, nor lose it by exchanging atoms with older carbon.
(4) That for practical use of radiocarbon dates, the sample must be contemporaneous with the event that it marks, and not something that grew a long time before.
Now let us keep in mind that, if the radiocarbon clock is to give correct dates, all of the above assumptions must be true. If even one of them is untrue, the method breaks down and will not give the correct age.
The first samples of wood from old trees and from the tombs of Egyptian kings, measured in Libby’s laboratory, showed a reasonably good correspondence with the accepted ages of these samples, back to about 4,000 years. So it was thought that perhaps the assumptions were correct, at least nearly so. But how does the picture look now, after twenty years of investigation into the machinery of the radiocarbon clock? Do the assumptions still look as well-founded as they did then?
Reading through the reports of the Uppsala conference, one comes to the conclusion that, in fact, not one of the assumptions listed above is now known to be correct! Some of them are perhaps just a little wrong, but others have turned out to be quite wrong. Let us look at each of them again, in the light of present knowledge—or, it may be, of continuing ignorance.
Validity of Sample
Among the more obvious possibilities of error in radiocarbon dating is the loss in integrity of the sample. (Assumption 3) If a sample is altered by contact with, or contaminated by inclusion of, material that contains older or younger radiocarbon, the analysis cannot give the right answer. But the practical archaeologist has learned what to do about it when a sample comes back from the laboratory with a date different from what he expected. As Dr. Evzen Neustupný, of the Archaeological Institute of the Czech Academy of Sciences, told the symposium: “Contamination of samples by either modern or ancient carbon can often be clearly discerned if the result of a measurement deviates considerably from the expected value.”2
To paraphrase his words, he does not recognize the contamination of the sample before he sends it in, but when he looks at it again, with the unpalatable answer attached, he can see clearly that it was contaminated.
The same expert also pointed out, relative to the importance of selecting contemporaneous samples (Assumption 4): “It should be clear, although many archaeologists seem to ignore it, that radiocarbon measurements date the age of the organic tissue of the sample, i.e., the time when it originated. The tissue of a sample dating some historical (or prehistoric) event might have been biologically dead for several decades or even centuries when it was used by ancient man. This applies to wood for building, charcoal from hearths, and most other kinds of materials.”2
This is a point that the reader would do well to keep in mind when he sees a news item that radiocarbon dating of a piece of charcoal dug up from a cave somewhere proves that the cavemen lived there so-and-so many thousand years ago. There are places today where a camper could pick up firewood that had grown hundreds, even thousands, of years ago.
Errors of these kinds have occurred often enough to hinder the general acceptance of radiocarbon dates by archaeologists. But they have to do only with the application of the method to particular samples, so that one sample might be dated wrongly, but another correctly.
Beyond these, harder questions are being put to the radiocarbon-dating people, questions that strike at the very core of the theory itself. These questions, if not answered satisfactorily, raise doubts as to whether it can give the correct age of any sample.
Half-Life of Radiocarbon
One of the questions concerns the very first assumption. How sure is it that the half-life of carbon 14 is correct? Note the following comment by two experts from the radiocarbon laboratory of the University of Pennsylvania:
“What causes the most worry about the veracity of these half-life determinations is the fact that they all depend upon the same basic methods—namely, the absolute calibration of a gas counter for determination of the specific disintegration rate, and the subsequent mass spectrographic measurement of the exact quantity of C-14 that was counted. In the first phase there is the difficulty of obtaining an absolute calibration of a gas counter, and in the latter there is the problem of precise dilution and introduction of the ‘hot’ C-14 into the mass spectrograph. An error caused by adsorption of C-14 on the walls of the containers may be prevalent and of roughly the same magnitude in all of the half-life determinations. Clearly, there is need for an entirely independent approach and technique before one can say with certainty what is the true value of the half-life of C-14.”3
Libby himself was aware of this limitation in the accuracy of half-life. In 1952, writing of the vital importance of measuring absolute disintegration rates, he said: “It is to be hoped that further measurements of the half-life of radiocarbon will be made, preferably by entirely different techniques.”4 As yet this hope has not been realized.
Production of Carbon 14
What about the constancy of cosmic rays? (Assumption 2a) Observations have shown that they are not at all constant. Several factors are now known that cause large fluctuations in the cosmic rays.
One of these is the strength of the earth’s magnetic field. This affects the cosmic rays, which are mostly protons (charged nuclei of hydrogen atoms), by deflecting the less energetic particles away from the atmosphere. When the earth’s magnetic field becomes stronger, fewer cosmic rays reach the earth and less radiocarbon is produced. When the earth’s magnetic field becomes weaker, more cosmic rays reach the earth and more radiocarbon is produced.
Studies indicate that the magnetic field doubled in strength from about 5,500 years ago to about 1,000 years ago, and is now decreasing again. This effect alone can account for the needed correction of almost 1,000 years in the older dates.
Solar phenomena also cause large changes. The sun’s magnetic field extends far out into space, even beyond the earth’s orbit. Its strength changes, although not very regularly, along with the sunspot cycle of about eleven years, and this also affects the number of cosmic rays reaching the earth.
Then there are the solar flares. These great streams of incandescent gas burst out of the sun’s surface sporadically and eject enormous numbers of protons. Those that reach the earth produce carbon 14. This makes for an unpredictable surplus in the supply. A table and a graph in the report show the production of carbon 14 from typical flares. On February 23, 1956, there was a flare that produced as much carbon 14 in a few hours as in a whole year of average cosmic radiation. It is obviously impossible to include this kind of effect in the corrections to the radiocarbon clock, for no one knows whether the flares in past millenniums were more or less active than they are now.
The intensity of cosmic rays entering the solar system from the galaxy is another little-known factor. Geochemical scientists have tried, by measuring the very faint radioactivities of various elements produced in meteorites by cosmic rays, to get some idea of average intensities in the past. However, the results do not help much in giving the desired assurance of constancy over the past 10,000 years.
The radiocarbon theory would be in a stronger position (though still not invulnerable) with respect to the above objections if it could be shown that the radiocarbon is today decaying as fast as it is being formed. (Assumption 2c) If this is found not to be true, then the assumption of a constant inventory of carbon 14 is also proved untrue, and the assumed constant activity of radiocarbon is put on a precarious tightrope between two mooring posts that may be rising independently of each other.
The production rate is very difficult to calculate. Libby attempted to do this with the best data available up to 1952. He found a production corresponding to about nineteen atoms of radiocarbon per second for every gram of carbon in the reservoir. This was somewhat higher than his measurement of sixteen disintegrations per second. But in view of the complexity of the problem and the rough estimate that had to be made of so many factors, he regarded this as agreeing well enough with his assumptions.
Seventeen years later, with better data and better understanding of the process, can this be calculated more precisely? The experts at the symposium could say nothing more definite than that the radiocarbon is being produced at a rate probably between 75 percent and 161 percent of the rate at which it is decaying. The lower figure would mean that the amount of radiocarbon is presently decreasing; the higher figure, that it is increasing. The measurement gives no assurance that it is constant, as the radiocarbon theory demands. Again, recourse is taken to the view that “the relative constancy of the C-14 activity in the past suggests that [this ratio] must be confined to a much narrower range of values.”5 So one assumption is used to justify another.
Reservoir of Carbon 12
Not only the inventory of carbon 14, but also the stable carbon 12 in the exchange reservoir, must be constant to keep the radiocarbon clock synchronized. (Assumption 2b) Have we good reason to believe that this assumption is valid?
Since there is about sixty times as much carbon in the ocean as in the atmosphere, we are concerned chiefly about that oceanic reservoir. This point came up for discussion at the Uppsala meeting, where the consensus was that what they call an “Ice Age” could cause major perturbations. Libby had pointed out this possibility in 1952:
“The possibility that the amount of carbon in the exchange reservoir has altered appreciably in the last 10,000 or 20,000 years turns almost entirely on the question as to whether the glacial epoch, which, as we will see later, appears to reach into this period, could have affected the volume and mean temperatures of the oceans appreciably.”6
Effects of the Deluge
Mention of the volume of the oceans immediately raises in the mind of the Bible student the possibility of major dislocations in the radiocarbon clock at the time of the global deluge of Noah’s day, 4,340 years ago. The oceans must certainly have been much greater in extent and depth after the Flood. This in itself would not increase the amount of carbonate in the ocean; it would merely dilute it. The amounts of carbon 14 and carbon 12, as well as their ratio, which determines the specific activity, would not have been changed merely by the fall of the water. However, the increased volume would give the ocean the capacity ultimately to carry a much larger load of dissolved carbonate.
And adjustments in the crust of the earth would be expected because of the greatly increased weight of water on the ocean basins. This pressure would be greater than that over the continents. It would push the underlying plastic mantle away from the ocean beds toward the continents, thus lifting them to new heights. This would expose rock surfaces to increased erosion, including the limestones in the beds of shallow seas that geologists show in low-lying continental areas in their maps of Pliocene times.
So, beginning shortly after the Flood, the oceanic reservoir of carbonate would steadily increase until it reached the concentration we have today. Then, rather than assume that the carbonate reservoir has been constant, we should consider the possibility that it has been gradually increasing over the past 4,300 years.
How would the Flood affect the carbon 14? Since the Bible indicates that the water that fell in the Deluge was previously suspended in some way above the earth’s atmosphere, it must have impeded the entrance of cosmic rays and hence the production of radiocarbon. If uniformly distributed in a spherical shell, it could have prevented completely the formation of radiocarbon. However, it is not necessary to assume this; the water canopy might have been thicker over the equatorial parts than over the poles, thus admitting cosmic rays at low intensities. In any case, the removal of this shield by its falling to the surface would increase the rate of producing carbon 14.
Thus, we should expect that, after the Flood, both the radioactive carbon 14 and the stable carbon 12 in the oceanic reservoir would begin to increase rapidly. Remember that it is the ratio of carbon 14 to carbon 12 that fixes the specific activity. So, depending on just how quickly the erosion of the land added carbonate to the seas, the activity might either increase or decrease. Indeed, it would be possible, though not probable, that the growth of one would just balance the growth of the other; in that case, the radiocarbon clock would have continued to run uniformly right through the Flood. Libby pointed out the possibility that such a fortuitous balancing could bring about the “agreement between the predicted and observed radiocarbon contents of organic materials of historically known age.”7 But he did not prefer this explanation.
Since the inventories of carbon 14 and carbon 12 are independent of each other, it is possible to postulate values that would account for the excessive ages reported on old samples. For example, if we assume that the specific activity before the Flood was about half its present value, all pre-Flood specimens would appear to be about 6,000 years older than they really are. This would also be true for a while afterward, but with a rapid erosion of carbonate in the centuries after the Flood, the error would be reduced. It appears that by about 1500 B.C.E. the activity had approached its present value, since radiocarbon ages seem to be nearly right since then.
The Simultaneity Principle
These are some of the recognized problems that beset the radiocarbon chronology. There are others that have hardly been considered, and possibly some yet unthought of. These are the reasons why the theory set forth twenty years ago is no longer tenable. It is just not possible, merely by measuring the radiocarbon in a sample and comparing it with the present-day activity, to tell with any assurance the age of the sample. However, one feature of the radiocarbon theory seems to have held up so far, the principle of simultaneity.
This principle states that at any time in the past, the radiocarbon level was the same all over the world, so that all samples that originated at the same time had the same activity. So, barring alteration and contamination, they will have decayed to the same measured activity today. So, even if all the other assumptions have to be abandoned, if enough samples of absolutely known dates can be measured to construct a correction curve, then radiocarbon measurements can be made to find the position of a sample on this curve, and so its age can be inferred.
One laboratory has collected a series of samples of wood from long-lived trees, and has assigned dates to them by counting the growth rings. They have supplied such samples to the radiocarbon laboratories, and these dates are now widely accepted as providing a solid foundation for the radiocarbon chronology. Indeed, without this emergency footing, the radiocarbon clock would by now be so battered that it could hardly be trusted to give more than a rough idea of the true ages of things.
Now, if we are to believe the corrected radiocarbon dates, we must be ready to transfer our faith to tree-ring dating as the fundamental standard. How reliable is this new method? Let us examine it in the following article.
a References are found on page 20.
[Chart on page 9]
(For fully formatted text, see publication)
CARBON-14 DATES—CORRECTION CURVE
The carbon-14 dating method has been “corrected” so much that it is difficult even for other scientists to understand. Do the “corrections” open up more ways in which it could be wrong?