Last week, I briefly discussed historical approaches in science and how this applies to geologic dating methods – that is, how do geologists assign ages to a given rock? My goal was provide a basic understanding of scientific models in general, noting that the scientific method is used to falsify hypotheses and assumptions intrinsic to those models. Thus the scientific method can be used to discard models that don’t represent reality, while refining (and providing evidence for) models that do represent reality. At this point, I want to more specifically address the points made by Dr. John K. Reed in his Creation Research Science Quarterly article found here (downloads PDF file). I’ll divide my comments into three sections, dealing first with his comments on the geologic column, secondly with his comments on specific dating methods, and thirdly with my own thoughts on the strength of the geologic time scale.
Dr. Reed’s presentation of the geologic column
By way of preface, a bulk of Dr. Reed’s reference material is taken from the book A Geologic Time Scale by Gradstein, Ogg, and Smith (2004). If you are looking for a detailed explanation on how the geologic timescale is constructed, this is the authoritative work (and anyone with access to a university library can find it). However, I suspect that Dr. Reed has not spent much time with primary research in the fields of stratigraphy or geochronology. The reason is that Dr. Reed constructs a series of strawman arguments against the methods employed to construct the timescale (perhaps unintentionally?) and relies heavily on irrelevant citations in the text to give the impression that Gradstein et al. might even agree with his critique. My intention, however, is not simply to accuse Dr. Reed of dishonesty or incompetence. Rather, my goal is to exhort any serious reader to take advantage of the widely available reference, and decide whether he has accurately represented it.
In his introductory sentence, Dr. Reed asserts that scientists use the geologic timescale as a means to promote their philosophical disposition to naturalism. I would point out, however, that this accusation is no more meaningful that accusing Dr. Reed (or any YEC) of using the rock record to promote his/her predisposition to a so-called Biblical model of history. Scientists from a range of philosophical (and religious) backgrounds have constructed the geologic timescale through a variety of scientific methods. Whether you agree with the validity of these methods is not relevant; the point is that scientists have long worked together to reconstruct Earth history and subsequently interpret the philosophical implications of that history within their respective worldviews. Many of the earliest attempts to construct a geologic timescale were made by Christians, some of whom speculated ages of rock formations much older than had been previously assumed (e.g. Nicolas Steno; or Thomas Chalmers, who fully expected that young-Earth models would disappear within decades). Uniformitarian principles of geology were in place long before Darwin’s biological theories were articulated, let alone widely accepted. Most early biostratigraphers (scientists that correlate rocks based on fossils) rejected Darwin’s theories on the origin of species, despite their own predispositions to naturalism. Notwithstanding accusations by Dr. Reed and other YECs, most Christian geologists have been comfortable interpreting the rock record as a reliable proxy for Earth history (including the evolutionary development of life), recognizing that in itself, the reality of the geologic timescale cannot speak to philosophical commitments that underlie our investigation of nature. Granted, if the evidence pointed to a very young Earth, pure naturalists would face a greater challenge in accounting for life’s origin and development, but a young Earth in itself does not preclude naturalism. This accusation is a category error on the part of Dr. Reed.
Only a couple sentences later, Dr. Reed introduces a red herring to the discussion by citing Gradstein et al. (2004, p. 3), who notes that “the chronostratigraphic scale is an agreed convention, whereas its calibration to linear time is a matter for discovery or estimation.” Apparently, Dr. Reed understands this to mean that scientists no longer empirically investigate the chronostratigraphic scale (or never did?), but rather ‘fit’ the facts by means of ‘convention’ into their preconceived template of Earth history, and uses the citation to cast doubt on the methodology of stratigraphers. If you’re confused by the terminology, let’s take a quick detour. The chronostratigraphic scale refers to the relative ages assigned to rocks using the methods I discussed last week. For example, we assume that a rock layer is younger than underlying rock layers. Furthermore, the consistent order of fossils is used to group rocks into Periods, such as the Cambrian, Ordovician, Silurian, etc. One does not need a degree in geology to understand how Dr. Reed has abused the citation (but it helps to read the full paragraph preceding the citation). In saying that the chronostratigraphic scale is “an agreed convention”, Gradstein et al. (2004) have only described how scientists have assigned labels to each interval in the rock record, not how they determined the order. Through empirical investigation, scientists have documented succession of brachiopod fossils throughout the rock record, for example, but assigning a categorical cutoff (such as Cambrian brachiopods versus Ordovician brachiopods) is an agreed convention. In other words, scientists cannot, by definition, ‘discover’ that the Ordovician period actually preceded the Cambrian period any more than one could ‘discover’ that the Egyptian Middle Kingdom actually preceded the Old Kingdom! One could, however, propose new calendar dates for the range of each Kingdom through empirical investigation, just as geologists can propose new ‘calendar dates’ for the Cambrian-Ordovician boundary if the evidence demands it.
Within the introduction, Dr. Reed also asserts that the geological timescale lacks an absolute chronometer, which would mean that scientists have no way of assigning absolute ages to rocks. As I mentioned last week, no scientist believes that we can determine absolute ages of rocks, but rather that we have a working scientific model of estimating those ages. The difference is subtle, but important – can you pick it out? As with any scientific model, assumptions are made, but the progress of geochronology has only refined the respective assumptions and improved our confidence in the ages now assigned. Dr. Reed is correct in noting that dating methods “exhibit uncertainty, and…assume rather than prove deep time,” but it is unclear why this is relevant to the discussion. If a coroner examines a corpse and estimates the age of the person to be 85 years at death, his/her method not only exhibits uncertainty but also assumes the reality of the last 85 years. Likewise, when scientists estimate that Codex Sinaiticus (the oldest complete copy of the Christian scriptures) was compiled in ~325 A.D., their methods exhibit uncertainty and assume the reality of the past 1700 years. Neither case precludes the dating method from adding meaningful information to the discourse, however. Epistemological principles underlying historical scientific methods are by no means unimportant, but simply citing such principles as reason to dismiss the results constitutes yet another category error on the part of Dr. Reed. When a geologist obtains a radiometric date of 500 million years, it is understood that the method makes assumptions about the physical history of the rock and the reality of the past 500 million years. But why does Dr. Reed think the geologist has made an error in assuming deep time? Because of a particular understanding of Scripture – an understanding that is rooted, no less, in the principles of Hebrew grammar and syntax, textual criticism, and hermeneutics.
If you recall the analogy I made to reconstructing history from a set of tattered diaries, I attempted to show how a multiplicity of dating methods can provide internal checks, verify or falsify key assumptions, and improve the overall resolution of the model. In geology, the case is no different, yet Dr. Reed claims that “the need to bounce back and forth from one method to another reveals the fundamental lack of a consistent ‘clock’ against which the rocks can be calibrated.” It is unclear what Dr. Reed means by “the need to bounce back and forth” between methods — perhaps he is referring to the fact that not every method can be applied to every rock? — but it does not logically follow that no reliable ‘clock’ exists. While radiometric dating methods have been refined (or replaced) over the years, this hardly constitutes “repeated failures” that undermine the geologic timescale. Moreover, it is a caricature for Dr. Reed to imply scientists deemed these methods “infallible” or proclaimed them as “scientific gospel.” On the contrary, inconsistencies in radiometric dates have only improved our understanding of the respective methods. For example, when historic lava flows yielded anomalously old Potassium-Argon (K-Ar) dates (e.g. Dalrymple, 1969), the assumption that all argon should be excluded during crystallization was falsified. Advances in scanning electron microscopy (SEM) and electron microprobe analysis revealed compositional zonation within individual minerals that were previously assumed to be homogeneous. Such technological advances were seminal to the development of the more accurate and consistent 40Ar/39Ar method, upon which the modern geologic timescale now heavily relies. However, one should never forget that radiometric ages will always represent model ages, and thus are subject to change in the case that our improved understanding of geology falsifies the underlying assumptions.
Dr. Reed continues his assessment by characterizing the stratigraphic record as patchy and incomplete. What this means is that in any given location, only a fraction of Earth history has been recorded in the rock record. This comes as no surprise to any geologist, nor should it to anyone in the general public. Currently, sediments are accumulating (and thereby recording Earth history) in the San Joaquin Valley of California, but are not accumulating the adjacent mountain ranges (which are actually the source of those sediments). In order for sediments to accumulate where the Sierra Nevada range is currently located, the mountains must be weathered down and subside to form a sedimentary basin. How long do you suppose this would take? As you try to ‘guesstimate’ the answer, you can appreciate why considerable time gaps are expected to exist within the rock record. Although Dr. Reed presents this fact as an embarrassing challenge to the “pure empiricist” (which, by the way, no scientist is), the absence of rock record for a given time interval commonly provides valuable information to the geologist. First of all, it reveals that the area did not constitute a sedimentary basin, but rather a source of sediments to adjacent regions. For example, Cretaceous rocks can be found throughout much of eastern Utah. From east to west, the rocks transition from silty/limey sediments with marine fossils to sandy/silty sediments with terrestrial fossils, suggesting that a shoreline ran through the middle of the state, with the sea to the east and highlands to the west. Thus we can predict that a mountain range was present in western Utah and eastern Nevada during the Late Cretaceous and provided sediments to riverine deposits found to the east (for reference, Bryce Canyon National Park contains a record of these deposits). If our interpretation is correct, there should be no rock record for this interval to the west (an unconformity), but we should find evidence of those mountains in rocks from central to eastern Utah (i.e. fragments of previously formed sedimentary and igneous rocks). Since I took the time to describe the example, you may have guessed that is exactly what we find. Not only does eastern Nevada and western Utah contain a discontinuity in the stratigraphic record for this time interval, but sandstone layers from the Upper Cretaceous rocks in central-eastern Utah contain abundant fragments of older (Paleozoic) sedimentary rocks now exposed in Nevada and western Utah. Moreover, detrital zircons (pieces of zircon mineral grains from igneous rocks now found in sedimentary rocks) can be dated to track the source of sediments, and have been examined in Upper Cretaceous rocks from central Utah. A recent study documented clusters of detrital zircon ages in the range of 81-76 Ma (Jinnah et al., 2009), which is consistent with ages of volcanism in western Nevada and southern Arizona.
Before moving on, we should consider the implications of the previous example. The data imply that sediments comprising Upper Cretaceous sandstones of central Utah came from recycled sedimentary rocks to the west, with some source from volcanic rocks farther to the west. Thus Paleozoic rocks of Nevada/Utah needed time to accumulate, harden into rock, and undergo weathering and erosion before being carried more than 100 km to accumulate in a newly formed basin. This is in addition to large volcanic eruptions, which needed time to cool and crystallize (but not too much time, since rapid crystallization forms glass and not euhedral crystals). Is it any surprise that geologists have not quickly abandoned their assumption of deep time in spite of difficulties encountered while refining the geologic timescale?
In case you did not follow my example, consider another analogy to history. The record of human history is notably patchy and incomplete, just like the rock record. History has not been preserved for a majority of ancient peoples, due to an absence of written records or a subsequent destruction of evidence. While this provides a significant challenge to historians and archaeologists, they have been able to apply a multiplicity of scientific methods to piece together isolated records and reconstruct a meaningful history of mankind. Similar assumptions go into detailing human history as in geology, yet there is no outcry against historians for presenting an equally uncertain history with confidence.
Finally, Dr. Reed’s claim that Creationists can use biblical history as a template for understanding geologic history is simply misguided. Even assuming Dr. Reed’s interpretation of biblical history (i.e. a young Earth and a global flood), biblical history is by no means exhaustive. It is equally valid to propose that “little green men…influenced the course of evolution” after the Flood as it is to propose the same happened during a depositional hiatus in the Cretaceous. To respond that the preposterous story is contradictory to biblical history would be an argument from silence. I would encourage any readers to strongly consider the implications of Dr. Reed’s silly thought experiment. Are we to fear gaps in our understanding of nature, past and present, because they introduce uncertainty to our conclusions? I hope to address Christians in particular: do not advances in science improve our understanding of the world that was made and the one who made it? Yet scientific advances are not possible without treading boldly across those gaps in the hope that we can diminish uncertainty. Biblical theology lays the epistemological framework for empirical investigations of the natural world, but a reliance on all of scripture (including Genesis) as authoritative in matters of faith does not preclude our need of scientific investigation to understand science and history. Rather our use of science is warranted thereby. Once again, it is worth mentioning that even Dr. Reed’s interpretation of biblical history is dependent on historical and social sciences.
Dr. Reed’s understanding of geologic dating methods
I think that perhaps I should have begun with this section, in which I want to address the claims by Dr. Reed concerning specific dating methods. As much fun as it is to discuss the philosophy of historical sciences, don’t we just want to know whether such dating methods even work? Absolutely, and if you’ve read to this point, I appreciate your patience. So let’s take a closer look at each method mentioned in Dr. Reed’s criticism.
Radiometric dating methods have been constantly refined as our knowledge of the geology and physics behind the methods improves. The most commonly used methods for constructing the geologic timescale are the 40Ar/39Ar and U-Pb techniques, but other older methods have by no means been “thrown out”, as Dr. Reed asserts. I think the confusion lies in the fact that he primarily references Gradstein et al. (2004), who mainly considered dates for Period and Stage boundaries in the geologic timescale (i.e. they were dating only stratigraphic units of rock, such as volcanic ash layers). Dr. Reed does not offer an firsthand critique of the supposed shortcomings of each method, but is confident that Young-Earth critiques have sufficiently proven each to be “fatally flawed,” and that “the rock-solid chronology of radioisotopes has turned into quicksand.” I can only respond that the assessment is extremely optimistic, to say the least. Any review of published scientific literature employing radiometric dating techniques will show that by and large the results are consistent, and many underlying assumptions can be verified. Radiometric dating methods use a scientific model that does not always correspond to reality for a given sample, and hence discordant results do exist. I am aware that Dr. Andrew Snelling and others have compiled such outlier cases to promote uncertainty and doubt among the public acceptance of these methods, but I would warn against taking their claims too seriously. As a researcher in geology, my exhortation to you is to look more closely at the big picture, and realize that scientists have neither ignored discordant data or uncertainties in each method. On the contrary, such discordant data can give very useful information about a rock’s history. If you are still interested in the particular ‘problems’ raised by Dr. Snelling and others, I would be happy to add more detail to the discussion in future posts. For the time being, let’s consider the rest of Dr. Reed’s comments.
Although Dr. Reed does not believe geologists have any absolute chronometer, he is aware that radiometric dating “provides the only theoretical way to directly obtain absolute dates for virtually all of the rock record.” So if he were wrong about the reliability of radiometric dates, then the entire argument fails because geologists do have an absolute chronometer, or ‘reliable clock’, against which they can calibrate the chronostratigraphic scale. But Dr. Reed insists on adding confusion to the discussion with a rather nonsensical line of reasoning: “Fundamentally,” he says, “isotopic dates cannot confirm the stages of the timescale because uncertainty in these methods precludes a certain chronology.” By “isotopic dates”, I assume he means radiometric dates, and by “stages of the timescale” I assume he is referring to the intervals labeled over the years by geologists (such as the Cambrian, Ordovician, etc.). Is he trying to say that geologists can not confirm the time span of the Cambrian or Cretaceous, for example, because there are uncertainties in the dating methods? Does he believe that geologists have a preconceived age of each stage? (They don’t.) And what does it mean that uncertainty in the methods “precludes a certain chronology?” Which chronology? Or does he mean to say that the ± sign is too much uncertainty for scientists to handle? Dr. Reed continues:
are accurate reveals inaccuracies in other methods that once enjoyed equal confidence.”
Now things are getting ridiculous. I am not sure where Dr. Reed picked up his notions about how science is supposed to work, but it certainly wasn’t by contributing research to the field. It appears that he expects geologic dating methods to be proven infallible or considered useless, but where does this expectation come from? Isaac Newton used rather simple geometric methods and gravitational theory to estimate the distance to the moon. As technology improved so did estimates for this distance, and Newton’s original calculation was shown to be reasonably accurate (despite errors in some of his assumptions and variables). Technology will continue to improve, and estimated distances to all planetary objects will be updated correspondingly. But according to Dr. Reed’s line of reasoning, this means that nobody should tout with confidence that the moon, Sun, and stars are long distances away because there are uncertainties in our calculations! In fact, it’s probably just an illusion and no planetary object is further than the uppermost stratosphere. Yes, I know manned spacecraft have been there, but I could always propose that they fail to take into account changes in the physical laws of the universe as one ventures farther from the Earth’s surface. I would encourage Dr. Reed to spend more time arguing science as it is used by scientists, and less time redefining terms to play games with semantics. Yes, radiometric dates are wrong some of the time, and when they are wrong (discordant, at least) then geologists devote much more time exploring why they were wrong in that case. Then they formulate a hypothesis, test the hypothesis, and repeat the experiment in line with the scientific method. Dismissing scientific models because of uncertainties is not science; it is unwarranted skepticism (the same brand of skepticism employed by those who doubt the early authorship or textual transmission of the New Testament, for a distant but relevant analog).
Following the quote above, Dr. Reed cites Gradstein et al. (2004) to convince his audience that with each new radiometric dating method, older methods lose their once-held confidence. However, the citation was only discussing why certain methods (such as the Rb-Sr and Sm-Nd method) are not used as precision chronometers. This is a mis-citation on the part of Dr. Reed, who apparently does not understand the geologic reasons behind the preference. Methods such as Rb-Sr, Sm-Nd, and K-Ar still have application in geology and yield meaningful results, but there is more room for error in stratigraphic units where the interaction of hydrothermal fluids is more prominent, due to a much higher porosity and permeability of rocks (i.e water flows more freely through sedimentary rocks) and less isolated crystal systems. Although precise tuning of the geologic timescale is possible with these methods, it requires much more work in terms of quality control. So why waste the time and money?
Before moving on, I wanted to point out that Dr. Reed seems to think the point of radiometric dating methods is to substantiate a common belief in evolutionary theory by demonstrating the existence of deep time. However, any geologist (or geochronologist) would scoff at the association, recognizing that the age of rocks and the validity of evolutionary theory are two separate issues. Unfortunately, Dr. Reed’s association is very effective when it comes to the general public, which is more skeptical about (and spiteful of) evolution than the age of the Earth. Lastly, Dr. Reed claims that “while radiometric dating remains the mainstay of the timescale, it does so because the alternative is to admit…that the age of the earth has not been demonstrated to be measured in billions of years and that the historical record of the Bible is back on the table.” Once again, I think any geologist (Christians included) would scoff at the claim. Radiometric dating methods are not on the brink of extinction and geochronologists are by no means scrambling to counter the claims of AiG’s RATE team. But assuming they were, would a 6,000-year history be the only alternative? Deep time is not demonstrated by radiometric dating alone (or even primarily), but through a broad understanding of geologic processes responsible for rocks seen today: the accumulation of sediments; the emplacement of large magma bodies; crystallization and exhumation of igneous plutons; regional metamorphism of massive sedimentary rock bodies; spreading of the ocean floor; and biogenic structures, including the shear number of fossils and biomass contained within sedimentary rocks. Yet YECs like Dr. Reed create the illusion of a discipline in crisis by addressing these evidences in isolated cases rather than in the big picture.
I am going to take this one point by point and save a lengthier discussion for another time. Also, I encourage you to read Dr. Reed’s section on biostratigraphy in full before considering my comments.
them. As has been noted by many creationists, the argument is circular because the deep time of evolution is a presupposition of the method.”
This is false. Biostratigraphy is a method used to correlate rocks based on the fossils they contain, since it is assumed that fossils represent the flora and fauna living at time of deposition (an assumption verifiable by other geologic methods). Fossil assemblages were categorized early on, based on the location of the rocks containing them (e.g. Cambrian for Cambria, England). Further categorization allows biostratigraphic correlations to be more precise as new species are found and more sections of sedimentary rock are analyzed. Fossil species are considered index fossils if 1) the first and last appearance of that species in the rock record can be dated radiometrically, with repeatable results; 2) the fossil can be found in multiple localities around the world, and radiometric dates for those rocks are consistent with others; 3) the fossil is abundant in many rock types (i.e. dinosaurs need not apply). If these criteria are met, index fossils can be used to assign an age range (not an absolute age) to sedimentary rocks containing that fossil. The reasoning and process is quite simple, but how do we know it works? Well, I would first point you to the success of the oil industry, which relies heavily on biostratigraphy to pinpoint the location of oil reservoirs. Furthermore, I have worked in sedimentary sections myself, and have collected thousands of fossils. The order of fossil organisms is amazingly consistent, down to the subspecies level, and provides excellent evidence for the evolutionary history of life, as well as the long ages estimated radiometrically. But that is a discussion for another day.
That’s true. If you want to constrain the duration of geologic stages on the timescale, then a radiometric date taken from a stage boundary is much better than one from the middle of the stage. But I am thinking Dr. Reed has confused the use of the word ‘key’ here. It does not mean ‘dates that agree with our presuppositions.’
This is completely false. Geologic time periods were assigned long before evolutionary theory entered man’s conscience. The order is determined by the relative ages of rocks, which is determined by basic principles of stratigraphy. Here, Dr. Reed is citing another author, who was only making the point that I’ve been making all along. Biostratigraphy is used in tandem with sedimentary stratigraphy to assign relative ages and define stages for fossil-bearing rocks. This process does not require a knowledge of, or reference to, evolutionary history. Dr. Reed’s thinking is completely backwards on this topic.
Not at all. Again, Dr. Reed is playing semantic games with a citation from A Geologic Time Scale. On a side note, it seems most of his citations come from the first page of chapters in the book, leaving me to wonder whether he is familiar with the actual content. The original author referred to geological scaling techniques used in biostratigraphy. For example, the range of a certain fossil must be measured in multiple sections of sedimentary rock, but the thickness of each section will range from one to the next (sediments do not accumulate at the same rate in different water depths, climates, etc.). Scaling techniques allow geologists to estimate the age to thickness ratio for each section, so that an age can be assigned to each fossil or event once radiometric dates are available. Interestingly enough, radiometric ages invariably become younger throughout the rock section, as predicted by the interpreted relative ages of those rocks, and fit the geological scaling very well. Thus the “stretching” referred to by Dr. Reed has nothing to do with apparent lengths of time, but calibration of an unknown timeline to a timeline with known points of reference.
That is absolutely true. However, the catastrophic transport of anything, fossils included, leaves behind distinct sedimentary structures and characteristics. Thus the assumption can be verified by a simple field analysis of the rocks, as well as chemical analyses in the laboratory (that would be my field of study). A vast majority of index fossils are taken from fine-grained marine shales and carbonates, which show no evidence of transport (catastrophic or not).
Dates assigned to index fossils, once again, have nothing to do with evolutionary theory. I am stunned that Dr. Reed thinks it would be appropriate other than for the purpose of entertainment to compare “evolutionary dates from the nineteenth century” with radiometric dates now assigned to index fossils. On what were nineteenth century dates based? And for the record, many fossils do show evolutionary transitions.
Our knowledge of the fossil record is certainly incomplete, and nobody denies this. However, biostratigraphy relies on rock sections where the first and last appearances of a given fossil are documented in multiple sections around the world. Furthermore, correlations are never based on a single fossil type, but on dozens of fossil species that comprise a complex assemblage. Thus even if several fossil species disappeared from the rock record without actually going extinct, it would not affect biostratigraphic correlations by any meaningful degree. If new evidence suggests a better constraint on radiometric dates assigned to biostratigraphic intervals, then the range will change, but this has nothing to do with “ignorance of the complete fossil record”. Finally, although living fossils exist, these organisms are never used in biostratigraphy. Index fossils are typically microorganisms such as foraminifera, pollen, and radiolarians, or small shelled organisms such as brachiopods and trilobites. Has anyone demonstrated the existence of Cretaceous foraminifera in modern oceans?
This is by no means arbitrary. The reasons for using marine invertebrates are 1) their skeletal structure changes more frequently throughout the rock record, so that species can be distinguished more easily; 2) they occur in rocks formed in marine environments, where deposition is more constant and erosion is more rare. But I can’t figure out what Dr. Reed means by “sampling” here. Sampling of what?
This claim is both inaccurate and unfair to all. First of all, Darwin’s theories were not used in biostratigraphy until decades after he introduced them (and hundreds of years after the advent of biostratigraphy). Secondly, evolutionary history is preserved in the rocks, but this has nothing to do with Christianity, the tenets of which do not define our expectations for the rock record.
Dr. Reed devotes the rest of the section to commenting on a citation from Gradstein et al. (2004), who admit that some problems exist with “treating strata divisions largely as biostratigraphic units.” Of course, this admission seems very exciting to Dr. Reed, who perceives that “the biostratigraphic interpretation of the rock record is perhaps not so clear after all”, but I am certain he doesn’t understand the implications thereof. For one, Gradstein et al. are explicitly referring to cases where stage boundaries are defined only by biostratigraphic markers (fossils). Currently, this applies to about half of all stage boundaries, but that number is decreasing rapidly. Secondly, the uncertainty introduced by the problems that Gradstein et al. summarize do not affect the absolute ages of the timescale, but only where to place the age marker in a given sedimentary rock section. Imagine that an argument existed over where to define the beginning of the day: should it be at midnight, or should it vary based on sunset/sunrise? This is similar to the argument over which fossils should be used as boundary markers, but notice that neither option results in shorter or longer days. Now consider times in history before the invention of mechanic clocks. How do you define midnight then? More importantly, do uncertainties in rudimentary time-keepers give us reason to doubt the reliability of human history before the advent of Swiss watchmakers? Obviously not, and likewise there is no reason to dismiss the strength of biostratigraphy to correlate rocks. Uncertainties exist, but they don’t change the big picture by any stretch of the imagination.
If you’re not familiar with the concept of Milankovitch Cycles, don’t worry. The theory is rather straightforward: 1) Earth does not follow the same path every time it orbits the Sun; 2) variations in the axis of Earth’s rotation and the shape of its orbit occur over long periods of time; 3) variations in Earth’s orbit and axis affect the amount of energy received by the Sun, which effects the strength of seasonality and overall climate; 4) these variations are cyclic, like a sine wave, so the path of Earth’s orbit can be extrapolated over time. Combined, these four premises (and yes, I’m simplifying) are used to formulate a predictive theory about Earth history. We can predict, for example, that climate-dependent characteristics of sedimentary rocks should record astronomical cycles to some extent. If you’re confused, just think of it this way. Day and night are the result of an astronomical cycle — namely, the rotation of the Earth (sometimes you face the Sun, sometimes you don’t). Seasons are the result of Earth’s orbit around the Sun, combined with the fact that Earth rotates on an axis not perpendicular to that orbit. Milankovitch cycles are no different, qualitatively. Just imagine them as long-term seasons, which recur on the scale of 26,000, 41,000, and 100,000 years.
So what does Dr. Reed have to say about the use of astronomical cycles in stratigraphy? He states, “All such oscillations boil down to variations in solar radiation…reaching Earth…” As I also mentioned, changes in solar radiation are an important factor, but certainly not the only one. One must also consider the degree of seasonality (i.e. temperature and precipitation difference between winter and summer)and changes in sea level (and not just resulting from climate change, but directly from astronomical forcing). All of the above factors directly affect the water depth, temperature, and rate of primary production, which affect several characteristics of the sediments. At this point, Dr. Reed points out three major assumptions that he sees behind the use of astronomical cycles in stratigraphy:
this “signal” overrides terrestrial influences, (2) cyclicity and continuity in sedimentation driven predominantly by climate, and (3) uniformity of rates and preservation that enable the “signature” to be manifested.”
With regard to the first, I don’t see why Dr. Reed deems it necessary for the astronomical signal to “override” factors on Earth that affect sedimentation (say, tectonics?). In other words, geologists do not assume that astronomical cycles dominate the signal (such as chemical or lithological changes in sediments), but recognize that the astronomical signal will be superimposed on any terrestrial signal. As for the second ‘assumption’, geologists recognize that discontinuities in sedimentation occur, and such would pose a challenge to interpreting any astronomical signal. However, such discontinuities can be interpreted through a variety of geological methods (petrographic analysis and/or isotopic trends, for example). Furthermore, determining whether an astronomical signal is present requires thorough statistical criteria (as opposed to pure visual discernment: “Yeah, I think I see some cycles there?”). The use of non-parametric statistical analyses removes assumptions about perfect preservation. This applies to the third supposed ‘assumption’ as well.
Dr. Reed follows with a wonderful citation from A Geologic Time Scale, which describes briefly how the method is applied to the last 23 million years (where the model remains predictive). However, Dr. Reed jumps pass the brilliant success of the model in predicting sedimentary rock ages, which are later confirmed by radiometric dates, and proposes the existence of more supposed problems with the theory. For one, he notes that the method cannot be applied to rocks older than ~20 million years. This is true, but not in the sense that Dr. Reed assumes. Cycle stratigraphy can be applied to rocks of any age, just not when it comes to predicting the absolute age of those rocks from interpreted orbital cycles. In such cases, the method works more like using a ruler on a football field: we can use it to measure out fine-scale distances from a known marker (say, the 50-yard line). Thus if we have a single rock layer of known age (from a radiometric date), then we can use astronomical signals to estimate the age of the surrounding layers as we move away from the layer of known age. This has been used to estimate the exact duration of biostratigraphic zones, where the uncertainty in radiometric dates is larger than the duration itself (e.g. Locklair and Sageman, 2008).
Notice that at this point, we must ask the question: if the calibration of sediments to astronomical cycles can be verified for the past 20 million years (especially for the past 420,000 years), then why does Dr. Reed continue to write anything? The model has already been tested and tried for timescales much longer than the ~6,000 years he is defending, so what good is it to nit-pick about sources of uncertainty that are negligible to the big picture? Once again, it is unwarranted skepticism:
No, the assumption is not simplistic by any means. It is only accepted after being demonstrated by multiple independent methods. One could argue that a modern lake with 20,000 varves is not necessarily 20,000 years old, but when multiple radiometric dating methods obtain single-layer ages consistent with the predicted sedimentation rate, then the argument becomes a gratuitous assertion. Dr. Reed fails to realize that in some cases, the criteria he names are not present (slow and uniform sedimentation in response to climatic cycles) and geologists are familiar with such cases. The “whole theory” has not “crumbled”, however, because it still explains the big picture and there are physical reasons for such exceptions. Deeming such cases as exceptions requires application of the scientific method rather than wholesale, unwarranted dismissal of the facts. But Dr. Reed continues with rapid firing of more gratuitous assertions.
And can be ruled out easily by sedimentological, stratigraphic, and geochemical criteria. I’ve done this myself. It takes work; it takes time; but it’s not hard.
Turbidites are quite easy to pick out in the rock record. For one, they produce coarse-grained lithologies in deep-water settings — a good indication that you picked a bad spot to interpret “astronomical forcing”. And to answer Dr. Reed’s rhetorical question, the chemical ratios would be stochastic, and would fail statistical criteria. Of course, this can be tested quite easily, and I’d be happy to run the samples for Dr. Reed if he were to provide them.
On a final note, Dr. Reed offers that the Flood model would undermine all assumptions made by cyclostratigraphers. Of course that is true, but Flood geologists have yet to propose a working model that could predict sedimentary and geochemical trends observed in sediments, ice records, speleothems, and more. Until then, Dr. Reed’s comments only resound of skepticism based on personal preference. In the meantime, geologists have produced thousands of studies that use orbital cycling to correlate sedimentary rocks. Their success is witness to the viability of the method.
I don’t think I’ll spend any time here discussing the details of magnetostratigraphy; I would prefer to challenge you all to read Dr. Reed’s comments on the discipline and see whether his argument is consistent. In any case, I felt it was worth commenting on at least one misconception:
Here, Dr. Reed is referring to the dating of magnetic reversals using the ocean floor, but he obviously does not see the application to other rocks. Magnetic signatures can be taken from sedimentary rocks of all brands, and are more typically used to reconstruct the movement of continents over time (magnetic signatures also provide the latitude during deposition). Changes in the polarity of those signatures is used to correlate the sedimentary record (the result of sediments burying fossils) to the basalt record of the ocean floor (the result of volcanism at mid-ocean ridges), and with minor exceptions, they match up very well. So we must then ask Dr. Reed, how do you explain the correlation in a young-Earth model? I understand that Dr. Reed is confident that rapid magnetic reversals can be explained by Dr. Humphreys and others’ geophysical models, and we should expect to see reversals in both rock records, but why should they correlate at all? For example, why should sedimentary rock sections containing the Barremian-Aptian boundary (determined by the fossils present) also yield similar radiometric dates (~125 Ma) and show similar magnetic reversal patterns (e.g. He et al., 2008)? In the ‘uniformitarian model’, the answer is obvious. But it is yet unclear how in a Flood model all of these processes are related or should produce consistent data.
Dr. Reed devotes the remaining sections of the article to demonstrating his lack of familiarity with the construction of the geologic timescale, and particularly his inability to understand the application of geologic dating methods. Furthermore, he does not fully understand the assumptions that go into each method, and contradicts himself in trying to articulate them. For example, he repeatedly refers to an assumption of constant sedimentation rate, or constant spreading rates of mid-ocean ridges, while ignoring the fact that he has already cited authors who would never consider those assumptions as valid or necessary.
On an unrelated note, Dr. Reed’s writing style can be misleading in itself. For one, the use of “quotes” around words to encourage doubt is simply inappropriate for scholarly discussion, because it creates the illusion that a meaningful argument has been made by subtly adjusting the connotation for the reader. I highly doubt any of Dr. Reed’s audience would take me seriously if I constantly referred to the “magical instance” called “the Flood” that Dr. Reed has “verified” by “science.” Out of respect for the discussion and for the truth, I’d prefer to take the issue more seriously.
So while there is obviously more detail to be discussed, I wish to stop here and simply ask you, which model has thus far explained the big picture? Do you believe that the geologic timescale is in crisis? If so, to what degree and what is the alternative? I hope that I have been able to accurately summarize methods used by geologists to interpret Earth history. Further, I hope that you would not be afraid to ask a geologist if you have questions about how things work, and especially if you find Dr. Reed’s arguments to be convincing on any point. As you can see by the length of my discussion here, many geologists are more than happy for the opportunity to just…talk about rocks.
Gradstein, F.M., Ogg, J.G., Smith, A.G., 2004, A Geologic Time Scale: Cambridge University Press, 589 p.
Jinnah, Z.A., Roberts, E.M., Deino, A.L., Larsen, J.S., Link, P.K., Fanning, C.M., 2009, New 40Ar-39Ar and detrital zircon U-Pb ages for the Upper Cretaceous Wahweap and Kaiparowits formations on the Kaiparowits Plateau, Utah: implications for regional correlation, provenance, and biostratigraphy: Cretaceous Research, v. 30, p. 287-299.
Locklair, R.E., and Sageman, B.B., 2008, Cyclostratigraphy of the Upper Cretaceous Niobrara Formation, Western Interior, U.S.A.: A Coniacian–Santonian orbital timescale: Earth and Planetary Science Letters, v. 269, p. 540-553.