Archaeological Dating: Stratigraphy and Seriation

Timing is Everything - A Short Course in Archaeological Dating

Gravestones in an old Massachusetts cemetery, with iconography studied by Deetz and Dethlefsen
Gravestones in an old Massachusetts cemetery, with iconography studied by Deetz and Dethlefsen. Markus Goeres / Getty Images

Archaeological Dating Table of Contents | Part 1: Stratigraphy and Seriation | Part 2: Chronological Markers and Dendrochronology

Archaeologists use many different techniques to determine the age of a particular artifact, site, or part of a site. Two broad categories of dating or chronometric techniques that archaeologists use are called relative and absolute dating.

  • Relative dating determines the age of artifacts or site, as older or younger or the same age as others, but does not produce precise dates.
  • Absolute dating, methods that produce specific chronological dates for objects and occupations, was not available to archaeology until well into the 20th century.

Stratigraphy and the Law of Superposition

Stratigraphy is the oldest of the relative dating methods that archaeologists use to date things. Stratigraphy is based on the law of superposition--like a layer cake, the lowest layers must have been formed first.

In other words, artifacts found in the upper layers of a site will have been deposited more recently than those found in the lower layers. Cross-dating of sites, comparing geologic strata at one site with another location and extrapolating the relative ages in that manner, is still an important dating strategy used today, primarily when sites are far too old for absolute dates to have much meaning.

The scholar most associated with the rules of stratigraphy (or law of superposition) is probably the geologist Charles Lyell.

The basis for stratigraphy seems quite intuitive today, but its applications were no less than earth-shattering to archaeological theory. For example, JJA Worsaae used this law to prove the Three Age System.

For more information on stratigraphy and how it is used in archaeology, see the Stratigraphy glossary entry.

Seriation

Seriation, on the other hand, was a stroke of genius. First used, and likely invented by archaeologist Sir William Flinders-Petrie in 1899, seriation (or sequence dating) is based on the idea that artifacts change over time. Like tail fins on a Cadillac, artifact styles and characteristics change over time, coming into fashion, then fading in popularity.

Generally, seriation is manipulated graphically. The standard graphical result of seriation is a series of "battleship curves," which are horizontal bars representing percentages plotted on a vertical axis. Plotting several curves can allow the archaeologist to develop a relative chronology for an entire site or group of sites. 

For detailed information about how seriation works, see Seriation: A Step by Step Description. Seriation is thought to be the first application of statistics in archaeology. It certainly wasn't the last.

The most famous seriation study was probably Deetz and Dethlefsen's study Death's Head, Cherub, Urn and Willow, on changing styles on gravestones in New England cemeteries. The method is still a standard for cemetery studies.

Archaeological Dating Table of Contents | Part 1: Stratigraphy and Seriation | Part 2: Chronological Markers and Dendrochronology | Part 3: The Radiocarbon Revolution
 

Absolute dating, the ability to attach a specific chronological date to an object or collection of objects, was a breakthrough for archaeologists. Until the 20th century, with its multiple developments, only relative dates could be determined with any confidence.

Since the turn of the century, several methods to measure elapsed time have been discovered.

Chronological Markers

The first and simplest method of absolute dating is using objects with dates inscribed on them, such as coins, or objects associated with historical events or documents. For example, since each Roman emperor had his own face stamped on coins during his realm, and dates for emperor's realms are known from historical records, the date a coin was minted may be discerned by identifying the emperor depicted. Many of the first efforts of archaeology grew out of historical documents--for example, Schliemann looked for Homer's Troy, and Layard went after the Biblical Ninevah--and within the context of a particular site, an object clearly associated with the site and stamped with a date or other identifying clue was perfectly useful.

But there are certainly drawbacks. Outside of the context of a single site or society, a coin's date is useless.

And, outside of certain periods in our past, there simply were no chronologically dated objects, or the necessary depth and detail of history that would assist in chronologically dating civilizations. Without those, the archaeologists were in the dark as to the age of various societies. Until the invention of dendrochronology.

Tree Rings and Dendrochronology

The use of tree ring data to determine chronological dates, dendrochronology, was first developed in the American southwest by astronomer Andrew Ellicott Douglass. In 1901, Douglass began investigating tree ring growth as an indicator of solar cycles. Douglass believed that solar flares affected climate, and hence the amount of growth a tree might gain in a given year. His research culminated in proving that tree ring width varies with annual rainfall. Not only that, it varies regionally, such that all trees within a specific species and region will show the same relative growth during wet years and dry years. Each tree then, contains a record of rainfall for the length of its life, expressed in density, trace element content, stable isotope composition, and intra-annual growth ring width.

Using local pine trees, Douglass built a 450 year record of the tree ring variability. Clark Wissler, an anthropologist researching Native American groups in the Southwest, recognized the potential for such dating, and brought Douglass subfossil wood from puebloan ruins.

Unfortunately, the wood from the pueblos did not fit into Douglass's record, and over the next 12 years, they searched in vain for a connecting ring pattern, building a second prehistoric sequence of 585 years.

In 1929, they found a charred log near Show Low, Arizona, that connected the two patterns. It was now possible to assign a calendar date to archaeological sites in the American southwest for over 1000 years.

Determining calendar rates using dendrochronology is a matter of matching known patterns of light and dark rings to those recorded by Douglass and his successors. Dendrochronology has been extended in the American southwest to 322 BC, by adding increasingly older archaeological samples to the record. There are dendrochronological records for Europe and the Aegean, and the International Tree Ring Database has contributions from 21 different countries.

The main drawback to dendrochronology is its reliance on the existence of relatively long-lived vegetation with annual growth rings. Secondly, annual rainfall is a regional climatic event, and so tree ring dates for the southwest are of no use in other regions of the world.

See the glossary entry for Dendrochronology for more information and a bibliography.

Archaeological Dating Table of Contents | Part 2: Chronological Markers and Dendrochronology | Part 3: The Radiocarbon Revolution | Part 4: New Fangled Methods

It is certainly no exaggeration to call the invention of radiocarbon dating a revolution. It finally provided the first common chronometric scale which could be applied across the world. Invented in the latter years of the 1940s by Willard Libby and his students and colleagues James R.

Arnold and Ernest C. Anderson, radiocarbon dating was an outgrowth of the Manhattan Project, and was developed at the University of Chicago Metallurgical Laboratory.

Essentially, radiocarbon dating uses the amount of carbon 14 available in living creatures as a measuring stick. All living things maintain a content of carbon 14 in equilibrium with that available in the atmosphere, right up to the moment of death. When an organism dies, the amount of C14 available within it begins to decay at a half life rate of 5730 years; i.e., it takes 5730 years for 1/2 of the C14 available in the organism to decay. Comparing the amount of C14 in a dead organism to available levels in the atmosphere, produces an estimate of when that organism died. So, for example, if a tree was used as a support for a structure, the date that tree stopped living (i.e., when it was cut down) can be used to date the building's construction date.

The organisms which can be used in radiocarbon dating include charcoal, wood, marine shell, human or animal bone, antler, peat; in fact, most of what contains carbon during its life cycle can be used, assuming it's preserved in the archaeological record. The farthest back C14 can be used is about 10 half lives, or 57,000 years; the most recent, relatively reliable dates end at the Industrial Revolution, when humankind busied itself messing up the natural quantities of carbon in the atmosphere.

Further limitations, such as the prevalence of modern environmental contamination, require that several dates (called a suite) be taken on different associated samples to permit a range of estimated dates.  See the main article on Radiocarbon Dating for additional information. 

Calibration: Adjusting for the Wiggles

Over the decades since Libby and his associates created the radiocarbon dating technique, refinements and calibrations have both improved the technique and revealed its weaknesses. Calibration of the dates may be completed by looking through tree ring data for a ring exhibiting the same amount of C14 as in a particular sample--thus providing a known date for the sample. Such investigations have identified wiggles in the data curve, such as at the end of the Archaic period in the United States, when atmospheric C14 fluctuated, adding further complexity to calibration. Important researchers in calibration curves include Paula Reimer and Gerry McCormac at the CHRONO Centre, Queen's University Belfast.

One of the first modifications to C14 dating came about in the first decade after the Libby-Arnold-Anderson work at Chicago. One limitation of the original C14 dating method is that it measures the current radioactive emissions; Accelerator Mass Spectrometry dating counts the atoms themselves, allowing for sample sizes up to 1000 times smaller than conventional C14 samples.

While neither the first nor the last absolute dating methodology, C14 dating practices were clearly the most revolutionary, and some say helped to usher in a new scientific period to the field of archaeology.

Archaeological Dating Table of Contents | Part 3: The Radiocarbon Revolution | Part 4: New Fangled Methods | Part 5: A Few Cautionary Notes

Since the discovery of radiocarbon dating in 1949, science has leapt onto the concept of using atomic behavior to date objects, and a plethora of new methods was created. Here are brief descriptions of a few of the many new methods: click on the links for more.

Potassium-Argon: The potassium-argon dating method, like radiocarbon dating, relies on measuring radioactive emissions. The Potassium-Argon method dates volcanic materials and is useful for sites dated between 50,000 and 2 billion years ago. It was first used at Olduvai Gorge. A recent modification is Argon-Argon dating, used recently at Pompeii.

Fission Track Dating: Fission track dating was developed in the mid 1960s by three American physicists, who noticed that micrometer-sized damage tracks are created in minerals and glasses that have minimal amounts of uranium. These tracks accumulate at a fixed rate, and are good for dates between 20,000 and a couple of billion years ago. (This description is from the Geochronology unit at Rice University.) Fission-track dating was used at Zhoukoudian. A more sensitive type of fission track dating is called alpha-recoil.

Obsidian Hydration: Obsidian hydration uses the rate of rind growth on volcanic glass to determine dates; after a new fracture, a rind covering the new break grows at a constant rate.

Dating limitations are physical ones; it takes several centuries for a detectable rind to be created, and rinds over 50 microns tend to crumble. The Obsidian Hydration Laboratory at the University of Auckland, New Zealand describes the method in some detail. Obsidian hydration is regularly used in Mesoamerican sites, such as Copan.

Thermoluminescence dating:Thermoluminescence (called TL) dating was invented around 1960 by physicists, and is based on the fact that electrons in all minerals emit light (luminesce) after being heated. It is good for between about 300 to about 100,000 years ago, and is a natural for dating ceramic vessels. TL dates have recently been the center of the controversy over dating the first human colonization of Australia. There are several other forms of luminescence dating< as well, but they are not as frequently used to date as TL; see the luminescence dating page for additional information.

Archaeo- and Paleo-magnetism: Archaeomagnetic and paleomagnetic dating techniques rely on the fact that the earth's magnetic field varies over time. The original databanks were created by geologists interested in the movement of the planetary poles, and they were first used by archaeologists during the 1960s. Jeffrey Eighmy's Archaeometrics Laboratory at Colorado State provides details of the method and its specific use in the American southwest.

Oxidized Carbon Ratios: This method is a chemical procedure that uses a dynamical systems formula to establish the effects of the environmental context (systems theory), and was developed by Douglas Frink and the Archaeological Consulting Team.

OCR has been used recently to date the construction of Watson Brake.

Racemization Dating: Racemization dating is a process which uses the measurement of the decay rate of carbon protein amino acids to date once-living organic tissue. All living organisms have protein; protein is made up of amino acids. All but one of these amino acids (glycine) has two different chiral forms (mirror images of each other). While an organism lives, their proteins are composed of only 'left-handed' (laevo, or L) amino acids, but once the organism dies the left-handed amino acids slowly turn into right-handed (dextro or D) amino acids. Once formed, the D amino acids themselves slowly turn back to L forms at the same rate. In brief, racemization dating uses the pace of this chemical reaction to estimate the length of time that has elapsed since an organism's death.

For more details, see racemization dating

Racemization can be used to date objects between 5,000 and 1,000,000 years old, and was used recently to date the age of sediments at Pakefield, the earliest record of human occupation in northwest Europe.

Archaeological Dating Table of Contents | Part 4: New Fangled Methods | Part 5: A Few Cautionary Notes | Part 6: For More Information

In this series, we've talked about the various methods archaeologists use to determine the dates of occupation of their sites. As you've read, there are several different methods of determining site chronology, and they each have their uses. One thing they all have in common, though, is they cannot stand alone.



Each method that we've discussed, and each of the methods we haven't discussed, may provide a faulty date for one reason or another.

  • Radiocarbon samples are easily contaminated by rodent burrowing or during collection.
  • Thermoluminescence dates may be thrown off by incidental heating long after the occupation has ended.
  • Site stratigraphies may be disturbed by earthquakes, or when human or animal excavation unrelated to the occupation disturbs the sediment.
  • Seriation, too, may be skewed for one reason or another. For example, in our sample we used the preponderance of 78 rpm records as an indicator of relative age of a junkyard. Say a Californian lost her entire 1930s jazz collection in the 1993 earthquake, and the broken pieces ended up in a landfill which opened in 1985. Heartbreak, yes; accurate dating of the landfill, no.
  • Dates derived from dendrochronology may be misleading if the occupants used relict wood to burn in their fires or construct their houses.
  • Obsidian hydration counts begin after a fresh break; the obtained dates may be incorrect if the artifact was broken after the occupation.
  • Even chronological markers may be deceptive. Collecting is a human trait; and finding a Roman coin a ranch style house which burned to the ground in Peoria, Illinois probably doesn't indicate the house was built during the rule of Caesar Augustus.

    Resolving the Conflict with Context

    So how do archaeologists resolve these issues? There are four ways: Context, context, context, and cross-dating. Since Michael Schiffer's work in the early 1970s, archaeologists have come to realize the critical significance of understanding site context. The study of site formation processes, understanding the processes that created the site as you see it today, has taught us some amazing things. As you can tell from the above chart, it is an extremely crucial aspect to our studies. But that's another feature.

    Secondly, never rely on one dating methodology. If at all possible, the archaeologist will have several dates taken, and cross check them by using another form of dating. This may be simply comparing a suite of radiocarbon dates to the dates derived from collected artifacts, or using TL dates to confirm Potassium Argon readings.

    I believe it is safe to say that the advent of absolute dating methods completely changed our profession, directing it away from the romantic contemplation of the classical past, and toward the scientific study of human behaviors

    Thanks go to reader Roger Hall, for suggesting the series, and to Douglas Frink, for lending a hand when I needed one.