Membership is completely free, giving members unlimited access to all available resources, services, features, and benefits without any restrictions or fees.
Welcome back. Sign in to continue enjoying unlimited access to all available resources, services, features, and member benefits at no cost.
Ireland was not yet the familiar island of woodland, fields, towns and coastlines. Sea level was much lower than today because vast quantities of water were held in northern ice sheets. The present Irish Sea and Celtic Sea basins were correspondingly altered, and Ireland was part of a changing north-west European land-and-sea borderland. This physical connection made animal movement possible at some times, but it does not itself demonstrate a human road, a settled population or regular travel across a particular route.
Radiocarbon ages are not calendar years waiting to be read directly from a bone or piece of peat. Atmospheric carbon-14 varied through time, so measured ages must be calibrated against independently dated records. At roughly 34,000 years ago the calibration curve is necessarily less fine-grained than in recent millennia, and individual determinations can translate into wide, sometimes discontinuous calendar ranges. A nominal BCE year is useful for arranging a chronological archive, but it cannot identify a particular season, migration or human visit.
Marine Isotope Stage 3, broadly dated to about 57,000–29,000 years ago, was neither uniformly frozen nor climatically stable. Greenland ice cores and North Atlantic records show repeated abrupt changes over centuries and millennia. Ireland lay on the western edge of that North Atlantic world, where temperature, precipitation, vegetation, sea level and the position of glaciers could alter substantially. Around the notional date used here, the great maximum expansion of the last British–Irish Ice Sheet still lay in the future: evidence from land and sea places the principal Late Midlandian maximum broadly after about 27,000 years ago, with many Irish offshore ice margins reaching their greatest extent later still.
Radiocarbon laboratories measure ages in radiocarbon years, not BCE calendar years. Those measurements must then be calibrated against changing atmospheric carbon levels, producing ranges rather than an exact historical date. At ages beyond thirty millennia, the uncertainties are especially significant. Moreover, the climatic stages used by Quaternary scientists are broad intervals reconstructed from marine sediments, ice cores and terrestrial evidence; their boundaries do not behave like the edges of political eras. It is better to say that 31,822 BCE lies approximately within late MIS 3 than to pretend that it marks a precise turning point.
This was an age of pronounced climatic instability, not a single, unchanging “Ice Age”. Ireland lay on the Atlantic fringe of north-west Europe, exposed to shifts in temperature, precipitation, sea level and the growth or retreat of ice. The great expansion of the last British–Irish Ice Sheet, conventionally associated with the Last Glacial Maximum, lay ahead. Current reconstructions place its most extensive later phase mainly between about 27,000 and 23,000 cal BP, though individual sectors developed differently and the history of ice growth remains actively refined. Around 33,800 cal BP, Ireland was therefore approaching a much harsher glacial world rather than already standing at its fullest ice-sheet extent.
Ireland was not yet at the fullest extent of the last British–Irish Ice Sheet. Earlier syntheses place the broad build-up of that ice sheet after about 35,000–32,000 years ago, with its maximum extent later, broadly around 27,000–21,000 years ago. More recent work has stressed that growth was regionally uneven and that evidence for pre-maximum ice is difficult to reconcile into one simple island-wide map. Thus it would be misleading to imagine either a wholly ice-covered Ireland or a uniformly open land in this nominal year. Ice margins, frozen ground, snow cover and drainage conditions changed across relatively short geological timescales.
The principal difficulty is chronological as well as archaeological. Radiocarbon measurements are not calendar dates: they must be calibrated against changing levels of atmospheric carbon-14, and at this antiquity their calibrated results can span centuries or millennia. IntCal20, the current internationally used Northern Hemisphere calibration curve, reaches back to 55,000 calibrated years before present but makes no claim that every sample can be dated to a precise year. A date such as 31,827 BCE is consequently useful for arranging a modern archive, but it must not be mistaken for the resolution of the evidence.
Radiocarbon dating does not produce a simple ancient calendar. A conventional radiocarbon age measures remaining carbon-14 relative to AD 1950; it must then be calibrated against a curve built from independently dated natural archives. Beyond about 14,000 years ago, those calibration curves rest on combinations of floating tree-ring sequences, lake and marine sediments, corals and cave deposits rather than an unbroken annually counted chronology. At around 34,000 calendar years ago, calibrated ranges are consequently far more meaningful than a single reconstructed year.
MIS 3 broadly covers the interval before the most extensive phase of the last British–Irish Ice Sheet. It was not a single, unchanging “Ice Age”. Greenland ice-core records and Irish cave mineral deposits both indicate abrupt alternations between colder stadials and relatively milder interstadials. At Crag Cave in County Kerry, uranium-thorium dates show that stalagmites formed episodically during the last glacial period. Because liquid water was needed for their growth, these phases provide strong evidence that south-west Ireland sometimes experienced conditions above freezing, although they do not by themselves reveal the precise appearance of the landscape outside the cave at 31,831 BCE.
Dates this remote are produced by scientific measurements with margins of uncertainty, not by calendars. Radiocarbon ages must be calibrated against changing atmospheric carbon levels; even then, the results are ranges, sometimes broad ones. Sediments can be displaced by ice, bones can be redeposited within caves, and material from different episodes can become mixed. A fossil collection labelled “Ice Age” is not automatically a single living community or a single moment in time.
Radiocarbon dating does not deliver a direct calendar anniversary. A measurement estimates the remaining radioactive carbon in once-living material; it must then be calibrated against independently built records of changing atmospheric carbon. The resulting calendar range can be broad, particularly this far back in time. Moreover, a date on a bone establishes, within statistical limits, when the animal died. It does not automatically date a human action on that bone, identify where the animal was killed, or prove that people lived nearby.
The label 31,836 BCE gives a useful position on a modern timeline, but it cannot identify an event in Ireland during a particular year, season or generation. There are no written records, settlements, graves or securely dated occupation surfaces from which such precision could be claimed. In conventional “before present” notation, where present means 1950, the calendar arithmetic places this notional year at about 33,785 years BP. That is not itself a laboratory age. Radiocarbon measurements carry statistical uncertainty and must be calibrated against changing atmospheric carbon-14 levels; at this depth of time, their calendar ranges can span centuries or millennia.
MIS 3, broadly between about 57,000 and 29,000 years ago, contained sharp shifts between comparatively milder interstadial episodes and colder stadials. These changes matter greatly for Ireland. Whether a valley, cave entrance or coastal lowland could support plants and grazing animals depended on temperatures, snowfall, permafrost, sea level and the proximity of glaciers. A calendar label such as 31,839 BCE therefore conveys a false exactness. Radiocarbon determinations require calibration against internationally maintained curves, and their probability ranges are not interchangeable with an invented BCE anniversary. Geological and ice-core chronologies provide broad correlation, not a diary of Ireland.
Radiocarbon dates are normally expressed in years before AD 1950 and must be calibrated against independently dated records of past atmospheric carbon. Calibration produces ranges rather than a single exact historical year, and the uncertainties become especially important this far back. The modern IntCal20 calibration curve reaches to 55,000 calibrated years before present, but it does not convert an Ice Age bone into evidence for activity in one numbered BCE year. The numerical date in this article should thus be read as an approximate pointer within a much longer span.
The position of the Derryvree beds is itself important. They lie between tills: sediments laid down by glaciers or ice sheets. They show that, before a later advance overrode the district, freshwater conditions existed in at least this part of Fermanagh. The discovery overturns any simple picture of Ireland as continuously entombed beneath ice throughout the whole last glacial period. Instead, the island experienced shifting arrangements of ice, bare ground, periglacial terrain, lakes, marshes and drainage channels. Precisely how much of Ireland was ice-free at any particular moment around 31,844 BCE remains uncertain, because surviving deposits are rare and later ice repeatedly eroded or buried earlier landscapes.
In archaeological convention, “before present” means before 1950, while radiocarbon measurements require calibration before they can be expressed as calendar ages. Those procedures produce probability ranges rather than annual anniversaries, especially this far back. Around 31,845 BCE—roughly 33,795 years before 1950—Ireland was approaching the major expansion of the last British–Irish Ice Sheet, but was not necessarily uniformly ice-covered. The island’s conditions differed greatly between regions and changed repeatedly over long human timescales.
Ireland was approaching, but had not yet reached, the maximum expansion of the last British–Irish Ice Sheet. That later maximum is now generally placed around 26,000 calendar years ago, with the greatest ice volume following several millennia later. Around the notional year considered here, cold conditions certainly prevailed, but neither temperature nor ice cover was uniform across the island. MIS 3 was punctuated by abrupt shifts between relatively milder interstadials and severe stadials. The island’s terrain was consequently altered by frost, seasonal thaw, wind, snow and moving water as well as by the fluctuating margins of ice.
About 33,800 calibrated years ago, Ireland lay within the broad North Atlantic world affected by abrupt climatic shifts recorded especially clearly in Greenland ice cores. In that record, this approximate horizon falls close to the beginning of Greenland Interstadial 7, one of several comparatively milder interruptions within an overall cold glacial period. Such labels are valuable for comparison, but they must not be treated as a weather report for a given Irish valley or a single year. Climate signals took time to travel through different environments, while Irish terrestrial sequences are few and often difficult to date precisely.
The wider North Atlantic climate was unstable. Greenland ice-core records show repeated swings between colder stadials and comparatively milder interstadials, sometimes over periods much shorter than conventional archaeological eras. Around the notional date used here, conditions in and around Ireland were cold, but “Ice Age” should not suggest a single unchanging white wilderness. Temperature, snowfall, sea level, vegetation and the reach of glaciers varied substantially through time and from one region to another. The landscape that future ice would scour was already being shaped by frost, wind, meltwater and changing ice margins.
Radiocarbon measurements do not translate directly into calendar years. They must be calibrated against internationally maintained curves, and uncertainty becomes substantial at this depth of time. A useful local comparison is Derryvree, near Maguiresbridge in County Fermanagh. There, organic-rich freshwater silts and fine sands lay between two tills: deposits left by separate phases of glaciation. The material produced a conventional radiocarbon determination of about 30,500 years before present, with a large error margin; published calibration places it approximately between 36,900 and 31,700 calendar years before present. That broad interval encompasses the modern label 31,852 BCE.
At about 33,800 cal BP, Ireland had not yet reached the Last Glacial Maximum, when the British–Irish Ice Sheet would cover all or almost all low ground on the island. That maximum expansion came substantially later, broadly between c. 27,000 and 21,000 years ago, with major ice-sheet growth and shifting margins rather than one simple, uniform advance. Around the date represented by 31,853 BCE, parts of Ireland were demonstrably ice-free, although the geography of ice, exposed land and sea changed repeatedly over centuries and millennia.
About 33,800 years ago, Ireland was in the Midlandian glacial stage, the Irish expression of the last glacial cycle. It was not yet necessarily the uniformly ice-covered island of the Last Glacial Maximum. Recent ice-sheet syntheses show that the growth, maximum extent and retreat of the British–Irish Ice Sheet were asynchronous: different sectors advanced and later withdrew at different times. Broad reconstructions place substantial build-up after roughly 35,000–32,000 years ago, while maximum ice extent in many sectors came appreciably later, around 27,000–23,000 years ago.
This was long before Irish written history, agriculture, monuments or identifiable peoples. It also preceded the climatic low point conventionally called the Last Glacial Maximum, dated globally to about 26,500–19,000 years ago. Ireland was not yet uniformly buried beneath the ice sheet that later reshaped much of the island. Evidence from sediments, fossil remains, cave deposits and ice-sheet reconstructions instead indicates a varied and unstable world: some districts were ice-free, conditions were cold and strongly seasonal, and the growth of the last Irish Ice Sheet still lay ahead.
The most important Irish archive for this broad interval lies at Derryvree near Maguiresbridge in County Fermanagh. Roadworks in 1969 cut through a drumlin and exposed organic silts and fine sands trapped between two separate beds of glacial till. Such a sequence matters because it records a time when water, plants and animals occupied ground that was later overridden or enclosed by ice-derived sediment. The organic layer produced a radiocarbon determination of 30,500 years before present, with a large original uncertainty of roughly a millennium either side. The deposit is not a diary entry for a single season, still less for a single calendar year; it accumulated over time and its date must be calibrated against changing atmospheric radiocarbon levels.
Radiocarbon dating does not normally yield a single prehistoric calendar year. It measures the decay of radioactive carbon in once-living material, produces a result with an error range, and must then be calibrated against changing atmospheric carbon levels. A date expressed as 31,858 BCE should consequently be read as an approximate place on a deep-time timeline, not as a day or season fixed in the Irish archaeological record. The best chronological language is “around 34,000 calendar years ago”, while recognising that individual bones and sediments may be older or younger than that broad reference point.
Radiocarbon results are not calendar dates in themselves. They must be calibrated against independently constructed curves that allow for past variation in atmospheric carbon-14. The modern IntCal20 curve extends to 55,000 calibrated years before present, with “present” conventionally fixed at 1950. Even then, a measurement is expressed as a probability range, often hundreds or more years wide. The further back one goes, the less defensible it is to translate a numbered BCE year into a precisely observed moment. Around 31,860 BCE, therefore, the proper historical unit is a dated environmental interval.
Radiocarbon chronology provides the essential caution. “Before present” conventionally means before 1950, while calibrated ages are estimates of calendar age derived from changing atmospheric carbon levels. A modern BCE year can be converted only approximately at this antiquity. The closest Irish environmental sequence is at Derryvree in County Fermanagh, where organic-rich freshwater silts lay beneath later glacial deposits. A published radiocarbon determination of about 30,500 radiocarbon years BP calibrates broadly to approximately 36,900–31,700 calendar years BP. That range encompasses the notional year considered here, but it does not prove that every feature of the Derryvree deposit formed in one particular season, generation or calendar year.
At roughly 31,863 BCE, Ireland lay in a climatically unstable portion of the Late Pleistocene. The island’s landscapes did not resemble the green, wooded country familiar today. Deposits preserved at Derryvree near Maguiresbridge, County Fermanagh, are especially important. Organic silts and sands trapped between two layers of glacial till have produced a radiocarbon chronology calibrated broadly to about 36,900–31,700 years before present. That range overlaps the date in question, though it emphatically does not identify conditions in every part of Ireland in a single year.
This was Ireland before the Last Glacial Maximum, not Ireland at the maximum. The great British–Irish Ice Sheet would later expand over almost the entire island and its surrounding shelf, but reconstructions indicate a rapidly changing situation during the preceding millennia. A major synthesis of ice-sheet growth places the combined British–Irish Ice Sheet’s pronounced expansion at around 30,000 years ago, with further Irish growth by 29,000 years ago. That chronology makes it unsafe to imagine a single blanket of ice covering every part of Ireland at c. 33.8 thousand years ago. Equally, it would be wrong to picture a mild, wooded island: cold conditions, advancing glaciers and highly variable regional environments were already defining the future landscape.
Ireland was not yet at the Last Glacial Maximum, when the British–Irish Ice Sheet expanded dramatically and covered much of the island. That later maximum is generally placed around 27,000–23,000 years ago in the Irish Sea and Celtic Sea sectors, with substantial regional variation in the timing and behaviour of ice. Around the present horizon, conditions were cold, unstable and markedly different from those of modern Ireland, but it is unsafe to picture the whole island as either continuously ice-covered or uniformly open. Glaciers, permafrost, snow cover, exposed ground and vegetated refuges changed across both space and time.
