Radiocarbon dating was discovered in the 1940s. There are a number of laboratories around the world which now specialize in this form of analysis.  We are fortunate that one of the most important of these is located in Ireland - at Queen’s University Belfast.

The method (although somewhat more complicated in reality than is explained here) is based on the fact that there are three isotopes (atomic variations) of one of the most common elements in our environment - carbon.  One of these isotopes is radioactive - C14.  In the compounded form of carbon dioxide, C14 is passing through all living organisms naturally, either by respiration or ingestion and, effectively, at a constant rate. However that process ceases to occur at the death of the organism in question, be they human, animal or plant. The radioactivity of the carbon present in the organism at the time of death begins to decay at a rate that is well-known.  The difference between the amount of radioactivity remaining in a dead sample (e.g. from an archaeological find such as a bone, a piece of wood, some cereal grains etc) and that present generally in the atmosphere now can indicate the time that has elapsed since carbon dioxide was passing through the object in question - i.e. since it was alive - and hence its date.

There are a number of caveats to the basic assumptions that underlie this method, so certain corrections and compensations have to be made before one can arrive at an accurate usable date. For example, industrial activity over the last 200 years has altered the amount of C14 in the atmosphere and past natural events and processes such as solar activity and alterations in the Earth’s magnetic field have caused similar fluctuations.

The necessary process of adjusting ‘radiocarbon dates’ so as to give us corrected ‘calendar dates’ is known as calibration. Radiocarbon dates are always given as a range with plus and minus variations (e.g. 2159 + 20 BP i.e ‘before present’ which is estimated as AD 1950), while calibrated dates are also given as a range (e.g. 1264-1093 cal. BC) with a calculated indication (‘sigma’) of the probability of accuracy (e.g. 2 sigma indicates 95% accuracy).

Specialists always distinguish between ‘raw’ dates and ‘calibrated’ dates and, although a calibrated date may be preferred on account of its greater precision and clarity, the raw dates or the specific information as to how they were arrived at will always be required also.

Examples of C14 dating from the Medieval Rural Settlement Project

The Medieval Rural Settlement Project (see pages in Research section) excavations at Tulsk Fort, Co. Roscommon have presented a large number of contexts and material suitable for dating using radiocarbon dating techniques. Bone, charcoal, and seed remains survive in abundance, and the strategy used to obtain dates has focused on sampling the lower levels of the ditch deposits as well as the various occupation/activity horizons. The samples are being sent to the 14Chrono Centre at the Queen’s University Belfast for analysis. Four determinations have been made so far, and the results confirm the stratigraphic observations:

• UBA-15010 is from a charcoal sample recovered from the base of the early medieval external ditch, yielding a result of 1 sigma cal AD 660-758, 2 sigma cal AD 656-767.
• UBA-15011 is from a charred seed or nutshell recovered from the base of a fire-pit associated with the grain-drying kiln exposed within the ringfort’s interior, yielding a result 1 sigma cal AD 900-1011, 2 sigma cal AD 897-1016.
• UBA-15012 is from a charred seed or nutshell recovered from a rake-out deposit associated with the grain-drying kiln, yielding a result 1 sigma cal AD 783-936, 2 sigma cal AD 778-949.
• UBA-15013 is from a charred seed or nutshell recovered from the lowest level of the medieval tower’s garderobe, yielding a result of 1 sigma cal AD 1431-1445, 2 sigma cal AD 1419-1451.

From the perspective of the Medieval Rural Settlement Project, the mid-fifteenth century date from the garderobe is very exciting. The garderobe deposit lay beneath a layer of catastrophic collapse, and therefore the radiocarbon determination presents a direct fix on when the tower was last standing. It is interesting that contemporary accounts record the taking of Tulsk Castle in 1427 by rival factions within the O’Conor lordship. The coincidence of dates is intriguing.

Further radiocarbon determinations are awaited that will hopefully provide a fuller chronology of the site, including a date for the infant burial discovered (see page in this section on Osteoarchaeology), as well as the early modern deposits that seal the medieval strata.