Still Looking for a Date

Uncertainties & Systematic Errors in Dating Human Origins

When did humans originate and why does it matter? Is the date consistent with humans arising from common descent or through special creation? Is the scientific date consistent with the biblical date? Crucial to answering these questions is recognizing the scientific errors and their significance.

Establishing a scientific date for the origin of humans is challenging and methods for doing so far from reliable. All the methods involve big statistical uncertainties and even larger systematic errors.

Statistical uncertainties, also known as random errors, refer to lack of precision in making measurements. Systematic errors refer to environmental and instrumental factors that could shift all the measurements either up or down in value.

Carbon-14 Dating

The most reliable dating method developed for human history is based on carbon-14, a carbon isotope with six protons and eight neutrons (carbon-12 has six of each). This method measures how much time has passed since a living organism or tissue has stopped respiring, drawing carbon molecules from the atmosphere.

Carbon-14's half-life (time for radioactivity to reduce the original isotope quantity to half its value) is 5,715 ± 30 years.¹ Dates derived from radioactive decay measurements cease to be reliable when a given sample's age is less than 1/7th or greater than 7 times the half-life. For carbon-14, dates for anything that lived less than 800 years ago or more than 40,000 years ago will be unreliable.

Carbon-14 dating must also account for systematic effects. Carbon-14 is produced in the atmosphere by cosmic rays striking nitrogen-14. During the past 44,000 years, four supernova events have occurred within 360–820 light-years of Earth,² close enough for their cosmic rays to have altered carbon-14 dates, though likely by no more than about 10 percent.

Nitrogen-14 exposed to uranium-235, uranium-238, and thorium-232 radioisotopes transforms tiny amounts of nitrogen-14 into carbon-14. These radioisotopes explain why some billion-year-old zircons and diamonds have registered carbon-14 ages of only 58,000 years.

An organism's location when alive may also alter the carbon-14 measurement. If it lived at high elevation, it would have been exposed to more cosmic radiation. If it lived underground or under a dense forest canopy, it would have been exposed to less.

Systematic errors in carbon-14 dating can almost always be detected, and their source identified and measured. Therefore, carbon-14 dating, where applicable, is the preferred method. It is the most reliable method, but useful only for remains and artifacts between 800 and 40,000 years old.

Thermal & Optical Luminescence

Most often, anthropological samples are too old, the carbon is not from once-living tissue, or the carbon comes from disparate sources for use of the carbon-14 method. The most common alternate dating methods measure luminescence, through heat—thermal luminescence—or light—optical luminescence.

Heat and light cause certain chemicals in samples to fluoresce. Thermal and optical luminescence indicate how long a sample has been cut off from heat or light exposure, usually sunlight. Typically, a sample's capacity for fluorescence measures how long it has been buried.

When a crystalline grain, such as quartz, is buried and cut off from sunlight, the radioactive decay of uranium and thorium in surrounding rocks and soil knocks electrons in the crystal out of position. Some of these electrons build up over time. Optical luminescence measures the buildup degree to determine how long ago the crystal was buried.

Because the heat or light intensity before a sample's burial can be high or low, depending on its environment, the systematic errors in thermal and optical luminescence can be large. Additionally, the burial process may be stretched out over time, rather than occurring immediately, or the burial may have been interrupted or disturbed on occasion.

When using thermal and optical luminescence, researchers date one or more mineral crystals in a proximate artifact, not the organism's remains. One must assume the artifact arrived at its location by its association with one or more humans (not by other means) at approximately the same time. An artifact buried in recent sediment may be moved by several geological processes or by animals—for example, by bears digging into an older sediment layer.

Examples of major systematic errors in thermal and optical luminescence dating occurred with the Jinmium Rock Shelter artifacts in northern Australia. Anthropologists cited thermoluminescence dates of aboriginal artifacts as evidence humans occupied Australia 60,000 years ago.³ A later carbon-14 analysis showed that the oldest artifacts were only 3,000 years old.⁴

This Jinmium artifacts' reassessment does not rule out human occupation in Australia prior to 3,000 years ago. However, it reduces the date estimate by a factor of two. Other Australian sites where both radiocarbon dating and atomic mass spectrometry have been applied yield dates of about 30,000 years ago for human artifacts.⁵

Electron spin resonance, another frequently used method, is a sophisticated version of thermal and optical luminescence methods. It measures the quantity of unpaired electrons in a sample that has been previously exposed to natural radiation (the infrared and visual portion of the electromagnetic spectrum). However, for an accurate age determination, a researcher must know the timing and degree of the sample's past exposure. Thus, electron spin resonance dating is subject to the same systematic errors hampering thermal and optical luminescence dating.

Chemical Precipitation

Another frequently used method examines the relative abundance of uranium and thorium in a sample. This method takes advantage of uranium's solubility and thorium's insolubility in water. The uranium-234 isotope decays into thorium-230 with a half-life of 245,000 years. (Uranium-234 exists despite Earth's 4.567-billion-year age because it is an indirect decay product of uranium-238, with a half-life of 4.468 billion years.) By determining the ratio of thorium-230 to uranium-234 in a sample, researchers can determine the time since its precipitation—its transition from a dissolved substance into an insoluble solid. But that's only if (1) the sample is known to have come entirely from a single, rapid precipitation event, and (2) the sample suffered no significant disturbances or contamination subsequent to the precipitation event. Herein lies the potential for errors.

DNA Dating

The two most familiarly cited tools are mitochondrial DNA and Y-chromosomal analysis. All humans' mitochondrial DNA comes exclusively from their mothers. All males' Y-chromosomes come exclusively from their fathers. Therefore, geneticists seek to determine how far back in time the first woman and first man lived, from whom we all descended. They

1. measure the genetic diversity in the present human population,
2. assume mutation rates for mitochondrial DNA and Y-chromosomal DNA, and
3. assume an average time between human birth and reproduction in a single generation.

Scientists realize, however, the mutation rate is not the same for all humans in all times and all regions. While certain environmental and social factors are known to alter human mutation rates substantially, a host of other factors have yet to be studied.

Even known environmental and social factors can involve huge uncertainties. Examples include mutations generated by cosmic rays from past supernova events and major solar flares. During the past 10,000 years, no supernova eruptions have occurred closer than 5,000 light-years from Earth.⁶ However, four occurred 22,000–44,000 years ago at distances of 320–820 light-years, and nine more occurred 35,000–115,000 years ago at distances of 350–700 light-years.⁷ Similarly, no major solar flares have blasted Earth during the past 10,000 years, but it seems likely several occurred 10,000–115,000 years ago. Failure to consider just these two types of events suggests published DNA dates for the origin of humans may be much too ancient.

Likewise, assumptions about the average time between birth and reproduction throughout human history likely are incorrect. Current values, based on widespread birth control and long career-launch times (before childbearing), likely skew the numbers.

From a biblical perspective, when God created Eve, he may have endowed her eggs with a diversity of mitochondrial DNA. In that case, the time calculated to have elapsed from the present back to when the first woman lived, based on mitochondrial DNA analysis, would be much greater than the actual elapsed time. Similarly, the biblically reported shortening of potential human lifespans after Noah's flood may have involved God's alteration of Y-chromosomal DNA.

Dating Prehuman Hominids & Their Artifacts

Reliable dating methods exist for pre-human artifacts and remains. For example, argon-argon dating and paleomagnetic dating can be employed reliably on samples ranging from 250,000 years to several million years in age.

These methods are not entirely free of systematic effects. However, possible errors measure much smaller than those associated with methods used in dating human artifacts. This difference explains why we can place greater confidence in scientific dates for pre-human hominids.

Four Cautions

How, then, should readers evaluate and interpret scientific dates for human origins and/or artifacts attributed to early humans? First, we must beware of edge-of-the-error-bar bias. For example, some journalists and bloggers claim humans date back 200,000 years, based on published calculations that mitochondrial Eve lived 157,000 ± 40,000 years ago. The 200,000-year date (outermost edge of the error bar) ignores the given range of 117,000–197,000 years ago.

Second, many published human origins dates acknowledge statistical errors only. Often such errors amount to less than 10 percent of the claimed age. We might look at a small statistical error and conclude the claimed age is trustworthy when, in fact, a systematic error could be ± 1,000 percent.

Third, with the exception of carbon-14 dating, researchers simply cannot yet determine values for all likely systematic errors. For this reason, anthropologists refrain from publishing both statistical and systematic error bars. They sometimes identify likely systematic effects, but without assigning values. Such researchers are to be commended.

Fourth, systematic effects in age measurements are almost always much larger on the minus side than on the plus side. That is, ages likely are closer to the more recent end of the range. Where systematic effects probably are large, stated ages are best viewed as approximate upper limits.

An appreciation of how statistical and systematic errors can impact human origins dates calls for tentativeness and humility. For now, determining a precise scientific date for humanity's origin remains elusive.

Notes
1. Norman E. Holden, "Total Half-Lives for Selected Nuclides," Pure and Applied Chemistry 62, no. 5 (1990), 941–958: doi:10.1351/pac199062050941.
2. R. B. Firestone, "Observation of 23 Supernovae That Exploded <300 pc from Earth During the Past 300 kyr," Astrophysical Journal 789, no. 1 (July 1, 2014): doi:10.1088/0004-637X/789/1/29.
3. R. L. K. Fullagar, D. M. Price, and L. M. Head, "Early Human Occupation of Northern Australia: Archaeology and Thermoluminescence Dating of Jinmium Rock-Shelter, Northern Territory," Antiquity 70, no. 270 (December 1996), 751–773: doi:10.1017/S0003598X00084040.
4. Richard Roberts et al., "Optical and Radiocarbon Dating at Jinmium Rock Shelter in Northern Australia," Nature 393 (May 28, 1998), 358–362: doi:10.1038/30718.
5. Bruno David et al., "New Optical and Radiocarbon Dates from Ngarrabullgan Cave, a Pleistocene Archaeological Site in Australia: Implications for the Comparability of Time Clocks and for the Human Colonization of Australia," Antiquity 71, no. 271 (March 1997), 183–88: doi:10.1017/S0003598X00084672.
6. Firestone, "Observation of 23 Supernovae."
7. Firestone, "Observation of 23 Supernovae."