WHEN Willard Libby and his col leagues introduced carbon-14 dating a number of years ago the technique was immediately recognized as a potentially valuable tool for objectively determining the age of carbonaceous archeological specimens. The basic validity of carbon-14 dating has now been established beyond reasonable doubt. What has not been established securely are the limitations of the method. All competent scholars agree that occasionally carbon-14 dates are at variance with other scientific data. This in no way discredits the basic technique as there are known sources of error that are expected to produce occasional anomalies.
Creationists have typically insisted that carbon-14 dates are not valid indicators of time for more than about four to five thou sand years before the present. Evolutionists do not recognize this limitation and accept the entire range of dates based on carbon-14 analysis. The relative merits of these positions have been extensively considered in creationist publications. Recently, the technique of tree-ring dating (dendro chronology) as applied to bristlecone pines has figured prominently in discussions of the validity and limitations of carbon-14 dating. This is because (1) tree-ring dating is considered to be an independent method of determining absolute dates— precise to the year; (2) tree-ring dates of up to 9,000 years have been reported; (3) tree-ring dates generally harmonize with carbon-14 dates. Such observations have been advanced to "prove" the credibility of carbon-14 dates as indicators of real time, on the one hand, and of tree-ring dates on the other.
Because of the fundamental question concerning the Biblical record that is involved, a thorough examination of tree-ring dating and its applications is in order.
How Are Dates Obtained From Tree Rings?
There are two basic principles of tree-ring dating. The first can be simply stated: One year; one ring. That is, each year a tree forms one growth ring that can be seen when the trunk is cut.
This principle is so well known that even school children deter mine the age of fallen trees by counting the rings of clean-cut stumps. However, botanists have long recognized certain pitfalls in this procedure. Occasionally climatic conditions are such that no identifiable ring is formed in a given year, or on the other hand, two or more rings may form. The consequences are designated as "missing rings" or "multiple rings," respectively. If rings are missing in the count, a falsely low age will appear. The apparent age will be incorrectly long if there are multiple rings.
The second basic principle is more elegant and relies on the fact that the width of a ring may depend to a large extent on the amount of rainfall in the year it was formed. Varying rainfall from year to year results in varying widths of rings. In certain locales the consequence is clearly defined ring width patterns reflecting distinct variations in climate. The principle used for dating is that: If the same distinctive pattern of ring widths is observed in two wood specimens, those two specimens were contemporary and their similar rings may be correlated one by one, year by year.
The value and application of this principle is illustrated on page 37. 1 The living tree when cut is found to have a distinctive pattern that overlaps (correlates) with part of the distinctive pattern found in a beam in a house (application of second principle). This allows the count of rings to be extended to include the entire older specimen (application of first principle). Similarly, the count can be successively extended back in time as long as successively older wood specimens are found with the requisite matching distinctive ring width patterns. The major limitation to application of this principle is caused by the requirement of a series of specimens, each with distinctive overlapping patterns.! As a matter of fact, most wood shows little variation in ring width from year to year. Such an indistinctive pattern is termed complacent. Complacent specimens are not suitable for use in tree-ring dating because there is no way to be sure which rings in one specimen match with specific rings in another.
Bristlecone Pine Chronology
For a number of years, C. W. Ferguson of the University of Arizona has been collecting and studying specimens of bristlecone pine wood from the White Mountains along the California-Nevada border. A variety of factors combine to make these trees long-lived. The trees are not large but some of them do exhibit thousands of growth rings. By application of both principles of tree-ring dating, Dr. Ferguson has derived a chronology extending to as much as 9,000 years before the present.
Careful study of this chronology has raised serious questions.
Bristlecone Pine Data Quality
Ferguson states, "In bristlecone pines, problems of cross dating are caused by so-called 'missing' rings associated with the extremely slow growth rate of this species on arid sites. One specimen, for example, contains more than 1,100 annual rings in 12.7 centimeters of radius. ... In some in stances, 5 per cent or more of the annual rings may be missing along a given radius that spans many centuries." 2 In fact, up to 10 per cent of the rings may be missing along a given radius.3 The significance of this is that the very narrowest rings are the ones most likely to be missing and the narrow rings are in general responsible for the distinctiveness of the pat tern. In fact, nearly half of the samples used for the original 7,104-year chronology have mean sensitivities of less than .30.* Such low sensitivities are indicative of complacent samples that are unsuitable for cross matching. How did Ferguson cross match them? Consider the following:
Cross matching of one pattern with another is largely a matter of visual inspection and judgment. Researchers have usually relied on "skeleton plots," showing only the distinctive features of a distribution as visual aids. This technique, though quite subjective is very useful; especially when dealing with short patterns. And one could reason that it is not really important how a crossmatch is found as long as it is verified statistically. However, matching a specimen with hundreds of rings against a chronology with thousands of rings is not easy. Therefore, in the case of the bristlecone pine chronology, the carbon-14 date of the wood was used to determine the approximate location for crossmatching. Says Ferguson, "I often am unable to date specimens with one or two thousand rings against a 7,500-year master chronology, even with the 'ball-park' placement provided by a radiocarbon date." 4
From the foregoing it is evident that a bristlecone pine tree-ring date is at least partially dependent on carbon-14 dating. How dependent it really is can be inferred from the complacent nature of the specimens. Complacent specimens (all rings about alike) will fit about as well in one place in the chronology as another. In fact, without a carbon-14 date one would never know where they should fit. This is a very important point since Ferguson claims that the bristlecone pine chronology "constitutes the first independent time control of such length for radiocarbon analysis." 5 (Italics supplied.)
In reality, one wonders how the bristlecone pine chronology can be used to verify the accuracy of carbon-14 dating when the researcher used carbon-14 dating to assist him in developing the tree-ring chronology. This might not be so important if each crossmatch was statistically verified to be in harmony with the second principle of tree-ring dating. But since the details of such verification have not been published and since the quality of the data is so poor (highly complacent ring sequences) the crossmatching has not been verified adequately.
Data Not Available
Usually the conclusions of scientists are open for investigation. Such investigation rightfully includes the methods and data used in arriving at the conclusion. There should be no "secrets" in the prosecution of truth. With respect to the bristlecone pine chronology, substantiating investigation has been hampered since the data upon which the conclusions have been based have not yet been made public. In view of the importance of the claims made for the bristlecone pine chronology, there are strong reasons why this data should be available for study.
Although tree-ring dating is a valid scientific technique, evidence indicates that the most widely known application of this technique is uncertain because of poor quality data and dependency on carbon-14 dating. This latter fact makes it questionable to cite the bristlecone pine chronology in support of carbon-14 dating, or vice versa.
Notes:
1 M. A. Stokes and T. L. Smiley, An Introduction to Tree King Dating (University of Chicago Press: Chicago, 1968).
2 C. W. Ferguson, Tree-Ring Bulletin 29?. (The Laboratory of Tree-Ring Research, University of Arizona, 1969.)
3 C. W. Ferguson, personal communication, March 3,1970.
4 Ibid.
5 C. W. Ferguson, prepublication abstract for footnote No. 2 above.
* Relative year-to-year variation less than 30 percent with respect to a normalized average value. Sensitivity of 0.25 (25 per cent) is considered "complacent."
