Accelerator Mass Spectrometry (AMS) — counting 14C atoms by accelerating carbon ions in a sample to very high speeds and then separating the 14C using powerful electric charges and magnets.

Age-at-death offset — difference in age between a sample and the contemporary atmosphere, arising from the time when the carbon in the dated organism was laid down.

Agreement indices — a statistical measure employed in the OxCal software to assess the compatibility of standardised likelihoods with the prior beliefs in a model.

Apatite — a mineral form of calcium phosphate. Hydroxyapatite forms the main mineral component of bone.

Bayesian statistics — branch of statistics in which evidence about the true state of the world is expressed in terms of degrees of belief.

Bayes’ theorem — express the relationship between prior and current beliefs (see Fig. 6).

β-particle — an electron emitted during radioactive decay.

Calcined — burnt grey/white. White fragments are preferred for radiocarbon dating.

Calibration — process of converting a radiocarbon measurement to an estimate of calendar date.

Carbon reservoirs — different stores of carbon on the Earth (e.g. the atmosphere, peat bogs).

Collagen — fibrous protein, one of the key skeletal substances.

Convergence — a diagnostic statistic that measures the stability of the solutions of a Bayesian model.

Conventional Radiocarbon Age (BP) — radiocarbon age calculated using the Libby half-life (5568±30 BP) and corrected for isotopic fractionation (Stuiver and Polach 1977).

Dated event — the event dated by a radiocarbon sample (e.g. the shedding of an antler).

Dendrochronology — tree-ring dating.

Dietary offsets — offset between the radiocarbon age of an organism and the contemporary atmosphere arising from diet.

Fraction Modern (F14C) 14C content of a post-bomb sample in relation to the 14C content of the atmosphere or ocean reservoir.

Fractionation — change in the ratio of two isotopes of a chemical element caused by the preferential loss or retention of one of them.

Freshwater reservoir effect — offset between the radiocarbon age of a sample and the contemporary atmosphere due to depleted carbon ingested from freshwater sources.

Fulvic acid — the fraction of bulk organic sediment that is soluble in acid.

Gas Proportional Counting (GPC) — counting the decay of 14C atoms in a gas sample using the current induced in a high voltage chamber by the electron discharged by a decay event.

Half-life — the time required for half the atoms in a sample of radioactive material to decay.

Hard-water errorsee freshwater reservoir effect.

Heartwood — the inner part of a tree that provides structural stability, but does not transport water or food reserves.

Highest Posterior Density interval — a range in which a certain proportion (usually 95% or 68%) of the true values of a distribution will lie.

Humic acid — the fraction of bulk sediment that is acid insoluble, and alkali soluble.

Humin — the fraction of bulk sediment that is insoluble in both acid and alkali.

Isotope — one of two or more forms of an element differing from each other in the number of neutrons present.

Liquid Scintillation Spectrometry (LSS) or Liquid Scintillation Counting (LSC) — counting the decay of 14C atoms in a liquid sample using the flash of light produced by a scintillant chemical on each decay event.

Markov Chain Monte Carlo (MCMC) methods — a class of algorithms for sampling from a probability distribution.

Marine reservoir effect — offset between the radiocarbon age of a sample and the contemporary atmosphere due to carbon ingested from marine sources.

Misfits — radiocarbon dates that do not accurately reflect the age of the target event (arising from either laboratory or archaeological error). These are termed systematic offsets in the statistical literature.

Minerotrophic — plant or substrate receiving most of its nutrients and water from streams or springs.

Mixed-source calibration — calibration of a radiocarbon measurement from material that obtained its carbon from more than one reservoir, where calibration curves are mixed proportionately according to estimates of the amount of carbon in the sample deriving from each reservoir.

Offsets — systematic difference between two sets of radiocarbon measurements (e.g. contemporary samples from the terrestrial and marine biospheres).

Old-wood effect — see age-at-death offset.

Ombrotrophic — plant or substrate receiving all its nutrients and water from rain (rain fed).

Outlier analysis — a formal statistical method for identifying and dealing with outliers; a form of model averaging.

Outliers — the 1 in 20 radiocarbon ages whose true value lies outside the 95% range given by a measurement’s quoted uncertainty.

Perfect pairs — a pair of samples of short-lived, single-entity materials from different carbon reservoirs that were freshly deposited in a context at the same time.

Posterior beliefs — our state of understanding a problem after considering new data.

Posterior density estimate — a function that describes the probability of a date occurring at a particular point in time.

Pretreatment — physical and chemical processing of a sample to remove exogenous carbon.

Prior beliefs — our state of understanding a problem before considering new data.

Probability — the chance of something happening.

Radiocarbon calibration — the process of converting a radiocarbon measurement into a distribution, or range, of possible calendrical dates, expressed as cal AD or cal BC.

Radioactive decay — the spontaneous disintegration of atoms by emission of matter and energy.

Range-finder date — single calibrated radiocarbon date used to identify the time when the activity occurred to within several centuries.

Reservoir effects — offsets between two sets of measurements on contemporary samples arising from differences in the age of different carbon reservoirs.

Sapwood — the outer part of a tree that contains living cells that transport water and store food reserves.

Sensitivity analysis — a series of alternative models that assess the changes in model outputs when the components are varied.

Single-entity sample — a sample composed of material derived from a single living organism.

Stable isotope — an isotope that does not undergo radioactive decay.

Standardised likelihoods — the data input into a Bayesian model (often calibrated radiocarbon dates).

Taphonomy — study of the routes and processes whereby material becomes part of the archaeological record.

Target event — the archaeological event a sample is intended to date (e.g. an antler pick is sampled to date the digging of the ditch in which it was found).

Total organic fraction — the chemical fraction of a bulk organic sediment that remains after the acid-soluble fraction has been removed.

Weighted mean — an average of two measurements, weighted to account for the errors on those measurements (see Ward and Wilson 1978).

Wiggle-matching — comparison of a series of radiocarbon dates separated by a known number of years against the calibration curve.