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.
Modern calendar-year precision is unavailable. A point corresponding approximately to 33.8 ka ago falls close to the transition between Greenland climatic events conventionally called Greenland Stadial 6 and Greenland Interstadial 6 on the extended GICC05 ice-core timescale. The onset of GI-6 is placed at about 33,740 years before AD 2000, but correlations between Greenland ice, Irish sediments and cave finds are neither exact nor automatic. An interstadial in a Greenland isotope record identifies a major North Atlantic climatic shift; it does not prove that the same temperature, vegetation or ice limit occurred simultaneously in every Irish valley.
Ireland was not then the familiar green, wooded island of later prehistory. Its climate was cold, but not simply and permanently frozen. MIS 3 was marked by sharp climatic oscillations: relatively milder interstadials alternated with severe stadials. In western Europe these changes could occur on timescales much shorter than conventional historical periods. A date around 31,868 BCE therefore belongs to an unstable environmental threshold, as the British–Irish Ice Sheet was expanding towards the far greater extent it would achieve later in the glacial cycle.
Ireland at this point lay in the Pleistocene, long before farming, monuments, metalworking, written language or the political communities familiar from later history. Archaeologists and Quaternary scientists commonly use Midlandian for the last cold stage in Ireland, broadly corresponding to the Devensian of Britain and parts of Marine Isotope Stages 4–2. Around 33.8 thousand calibrated years BP, the climate was cold, but the final and most extensive phase of the last Irish Ice Sheet had not yet overwhelmed all parts of the island.
Radiocarbon ages are not calendar ages. They measure the remaining carbon-14 in organic material and must be calibrated against records of changing atmospheric carbon. For dates this old, calibration commonly produces wide ranges rather than a neat conversion. The designation 31,870 BCE should consequently be read as an approximate point within a larger climatic and archaeological interval, not as an anniversary. It is safer to say that Ireland around this time lay in MIS 3, during the long and uneven approach towards the Last Glacial Maximum.
Radiocarbon dates are not calendar dates. A laboratory measurement records the remaining radioactive carbon in a sample and is conventionally expressed in radiocarbon years BP; it must then be calibrated against independently dated records of changing atmospheric carbon. At the depths of time involved here, calibration can produce broad and sometimes discontinuous probability ranges. A date such as 31,874 BCE is consequently useful only as a modern reference marker: it cannot identify a season, a generation, or even a discrete archaeological horizon. The relevant evidence comes instead from deposits whose ranges overlap that point in time.
Ireland lay on the north-western edge of Europe during a long cold age. The British–Irish Ice Sheet was not a fixed white blanket that appeared everywhere at once: it grew, shifted and responded to regional conditions. Major syntheses place its broad build-up after roughly 35,000–32,000 years ago, with maximum extent reached at different times in different sectors, broadly between c. 27,000 and 21,000 years ago. Around the notional date represented by 31,875 BCE, Ireland was thus moving towards, rather than sitting at the peak of, the Last Glacial Maximum.
For Ireland, this was the Middle Midlandian phase of the last glacial cycle: a long, unevenly cold era in which climate, ice cover, sea level and animal ranges repeatedly changed. The island’s later historical map is of little use here. There were no counties, settlements, peoples or political territories in the documentary sense. Even whether Ireland was continuously inhabited at this point cannot be demonstrated. What survives instead are deposits beneath later glacial sediments, cave-bone collections, pollen and plant remains, landforms, and laboratory dates. Each is a fragment with its own strengths and limitations.
For this period, Ireland’s archive is exceptionally uneven. Later prehistory is rich in houses, monuments and artefacts; the evidence for the earlier Upper Palaeolithic is sparse, frequently disturbed and difficult to date. Caves preserve bones but can also be used by carnivores, reworked by water, disturbed by digging or mixed by later processes. Sediments beneath glacial deposits preserve environmental snapshots, but only at rare sites exposed by quarrying, road works or research excavation. The resulting history is necessarily regional and provisional.
Ireland was approaching, but had not yet reached, the fullest expansion of the last British–Irish Ice Sheet. The Last Glacial Maximum is generally placed later, broadly after c. 26,500 calendar years ago, although ice-sheet growth was neither simple nor simultaneous in every region. Reconstructions show that the British–Irish Ice Sheet expanded in phases, with different sectors reaching their greatest extent at different times. Around the chronological neighbourhood of 31,888 BCE, Ireland stood within a changing glacial world: substantial ice existed in the wider north-west European region, but parts of Ireland had at earlier points in MIS 3 remained free of overriding ice long enough for lakes, wetlands, plants and animals to leave traces.
Derryvree preserves an unusually precious glimpse of an Irish landscape that later glaciation mostly destroyed or concealed. Mosses, sedges, herb pollen and cold-adapted beetles indicate open vegetation and a periglacial climate: ground subject to intense freezing and thawing, rather than a wooded country like modern Ireland. Sparse cover of grasses, sedges and low herbs would have occupied damp margins and more sheltered ground; exposed surfaces could have been stony, wind-scoured and seasonally frozen. This was tundra-like country, but it should not be pictured as a uniform white wilderness. Lakes, channels, wet hollows and patches of vegetation created local variety in a harsh landscape.
Radiocarbon dating does not translate automatically into a calendar year. A laboratory measures remaining carbon-14 in organic material and produces an age in radiocarbon years before present, where “present” conventionally means 1950. That result must then be calibrated against records of past atmospheric carbon-14. At this antiquity calibration produces ranges, sometimes broad or irregular ones, rather than an annual date. The internationally used IntCal20 curve extends to 55,000 calibrated years before present, but its older section necessarily combines several kinds of environmental archive and retains uncertainty.
The Derryvree sediments show that this part of Fermanagh was, for a time, free of overriding glacier ice. They also show that the interval lay between glacial episodes: older till beneath the organic deposit records an earlier advance, while younger till above it records ice that later crossed and incorporated the wetland. This is an important corrective to any simple image of an Ireland either wholly frozen or wholly open. The island’s glacial history was uneven, with local ice, ice-free ground, changing margins and environments that could be buried or destroyed by subsequent ice movement.
In global archaeological terminology, Ireland lay in the later part of Marine Isotope Stage 3 (MIS 3), an unstable interval of the Last Glacial Period. It preceded the coldest and most extensive phase of the Last Glacial Maximum, but was not a uniformly mild respite before the ice. North Atlantic climates swung sharply between colder stadials and relatively warmer interstadials. Ice masses, sea level, vegetation and animal routes could alter over periods far shorter than the immense span suggested by a date in BCE.
Ireland at roughly 33.8 thousand calibrated years ago is known through a thin and difficult archive: fossil bones from caves, plant and insect remains preserved in sediments, layers of glacial debris, and mineral growth in caves. These are not historical records. They cannot reveal a ruler, community name, battle, treaty, shrine, market or spoken language. Indeed, they do not securely demonstrate a settled human population. They instead allow a more fundamental question: what kinds of landscapes and animal communities could exist in Ireland as the last great ice-sheet expansion approached?
“31,898 BCE” is a modern conversion, not a date anyone living then could have recognised. No written records, settlements, graves, hearths, tool assemblages or works of art securely belong to this particular point in Irish prehistory. Even the broad chronology needs care. Radiocarbon ages are measured in radiocarbon years before present and then calibrated against independently dated records, including tree rings and other archives, to estimate calendar ages. At around 34,000 calendar years ago, the calibration curve is less precise than it is for recent millennia. A specimen may be assigned a calibrated probability range, while the event that deposited or altered it may remain less securely dated still.
31,899 BCE is a modern calendar label, not a year for which Ireland preserves an event, a ruler, a settlement diary or even an annually resolved archaeological sequence. At this depth of time, the responsible language is approximate. Counted from the conventional radiocarbon “present” of AD 1950, the label falls broadly around 33,800–33,900 calendar years before present (cal BP), within the Late Pleistocene and Marine Isotope Stage 3 (MIS 3), conventionally spanning roughly 57,000–29,000 cal BP. It belongs to the Middle Upper Palaeolithic world of north-west Europe, but there is no basis for assigning a particular Irish human action to this single numbered year.
Ireland was then part of the wider north-west European world of the last glacial cycle. It was not the uniformly ice-smothered island sometimes imagined from later phases of the Ice Age. The great British-Irish Ice Sheet would eventually expand dramatically, reaching its maximum extent on the Atlantic shelf roughly 26,000–24,000 cal BP and then retreating in stages. Around the equivalent of 31,901 BCE, however, evidence from beneath later glacial sediments indicates that at least some Irish regions had remained ice-free during earlier parts of the cycle. This does not mean that conditions were mild. It means that cold, seasonal and periglacial environments could exist beyond, between or before major ice advances.
Radiocarbon laboratories report ages in years before present, where “present” conventionally means 1950, not the present day. Those measurements must then be calibrated against independently dated records of past atmospheric carbon, using curves such as IntCal20. At great antiquity, a radiocarbon measurement can correspond to a wide calendar range rather than a single year. “31,902 BCE” is thus best read as a notional place within the later Middle Midlandian, the Irish term commonly used for this part of the Last Glacial Period, rather than as an anniversary that archaeology can observe.
The closest important environmental archive is at Derryvree near Maguiresbridge in County Fermanagh. There, freshwater silts and organic layers were preserved between two glacial tills: deposits left by ice advances before and after the quieter interval in which the sediments accumulated. A radiocarbon determination from the organic series, once calibrated, has been published as approximately 36.9–31.7 thousand cal BP. That is a broad range, but it encompasses the notional date of 31,904 BCE. Derryvree consequently provides a more responsible guide to conditions around this point than an attempt to assign a precise seasonal scene to the whole island.
“Before present” conventionally means before AD 1950, whereas BCE dates are a later conversion. Around 34,000 years ago, even well-preserved organic material does not yield calendar precision of the kind implied by a numbered year. Atmospheric radiocarbon varied through time, and the calibration curve becomes especially consequential beyond 30,000 calendar years ago. A bone’s laboratory result, its calibrated range, the layer from which it came and the possibility of later disturbance all need separate consideration. Consequently, this article uses “around”, “about” and calibrated ranges where the evidence permits; it does not claim that any animal, group or occurrence can be assigned to 31,907 BCE itself.
Derryvree offers an unusually valuable glimpse because its fine sands and organic silts were sealed between two tills, deposits left by separate incursions of glacier ice. During the period represented by the intervening freshwater sediments, south Fermanagh was not buried beneath an ice sheet. It had shallow water, wet ground and plant-rich hollows in a cold, open landscape. Pollen, moss remains and insects indicate tundra or muskeg rather than woodland: grasses, sedges, herbs and juniper were present, while dry ground was probably patchy and sparsely vegetated. Pools and saturated mossy ground could persist beside exposed mineral soils, shaped by freezing, thawing and seasonal meltwater.
At this point in the last glacial cycle, Ireland was moving towards the much more extensive ice cover that developed later. The most recent British–Irish Ice Sheet reconstructions model substantial growth after about 31,000 years ago and an Ireland almost wholly ice-covered by around 28,000 years ago. That does not allow a simple picture for c. 33,900 calibrated years before present. Conditions varied between regions, seasons and short climatic phases, while the exact timing of ice advance remains an active research question. The island was not yet the uniformly ice-dominated terrain of the later glacial maximum, but neither was it a hospitable temperate landscape.
Prehistoric chronology needs careful translation. “BP” means before AD 1950, the reference year used by radiocarbon science; a calibrated age converts a radiocarbon measurement into a probability range in calendar years. Neither system licences the apparent precision of a named BCE year. The figure 31,911 BCE is therefore useful as a position on a long timeline, but it should not be mistaken for a securely dated Irish episode.
Ireland was then in the last Ice Age, but not yet necessarily in the uniform, ice-covered condition popularly imagined for the whole period. The major expansion of the British–Irish Ice Sheet towards its later maximum was approaching and did not occur everywhere at precisely the same time. Reconstructions of the ice sheet show asynchronous growth and advance across Ireland, Britain and the continental shelf. Around this date, conditions could alter sharply over comparatively short geological intervals as North Atlantic climate switched between colder stadials and less severe interstadials. Any picture of a fixed, timeless frozen Ireland is therefore misleading.
At this date Ireland was not the wooded, temperate island familiar today. It belonged to a northern Atlantic world repeatedly reshaped by cold, changing sea levels, snowfields, frozen ground and advancing ice. The great expansion of the British–Irish Ice Sheet lay ahead: the best large-scale reconstruction places its maximum volume at about 23,000 years ago, while Irish evidence shows that ice cover and its limits varied substantially from region to region. Consequently, 31,916 BCE should not be pictured as a moment when all Ireland lay sealed under one uniform ice sheet. It was instead part of a long pre-maximum cold phase in which habitats could alter dramatically over centuries and millennia.
Modern BCE notation is far more precise than the evidence permits. Radiocarbon determinations measure the remaining carbon-14 in once-living material, then must be calibrated against changing atmospheric carbon levels; their results are normally reported as probability ranges, not as a single historical date. Moreover, the relation between a date expressed in “years before present” and a BCE year is not a simple conversion. “Present” in radiocarbon convention is 1950, and calibration can widen or shift the apparent age substantially. For this reason, 31,918 BCE should be read as approximately 34,000 years ago, not as the anniversary of a known occurrence. The relevant archaeological and environmental record spans thousands of years.
Radiocarbon ages and calendar ages are not interchangeable. A measured radiocarbon result has to be calibrated against curves that account for past fluctuations in atmospheric carbon-14, and the uncertainties become especially important in the deep past. IntCal20 provides a Northern Hemisphere calibration framework back to 55,000 calendar years before present, but it does not turn a cave find into evidence for a particular modern-style year. Accordingly, 31,920 BCE is best used here as a point within a broad chronological neighbourhood, around 34,000 calendar years ago, rather than as a moment at which anyone can be shown to have acted in Ireland.
Ireland was entering a decisive climatic transformation. The Last Glacial Maximum, when the British–Irish Ice Sheet covered much of the island, lay ahead rather than behind: broadly, major ice expansion belongs to the period after about 34,000 years ago, with full glacial conditions developing later. Geological evidence from beneath tills in Fermanagh and from the west coast indicates that substantial areas of Ireland had been ice-free during the preceding millennia. That does not mean a mild or stable country. Temperatures fluctuated sharply, seasons were severe, and the growing ice sheet was beginning to reshape the wider region.
Ireland was then part of the long last glacial cycle, but it should not be pictured simply as an island sealed beneath permanent ice. Marine Isotope Stage 3, broadly c. 57,000–29,000 years ago, was climatically restless. Greenland ice-core records and North Atlantic sediments reveal abrupt warmer and colder swings. The British–Irish Ice Sheet was already an important presence, but its precise extent and behaviour around 34,000 calibrated years ago remain debated. Recent synthesis points to earlier growth, retreat and renewed advance before the Last Glacial Maximum, rather than one uncomplicated march towards a frozen peak.
Archaeologists use radiocarbon determinations, stratigraphy, faunal study and geological dating rather than historical chronologies for this period. Radiocarbon years are not identical to calendar years and must be calibrated against changing atmospheric carbon levels; even then, dates are expressed as ranges with statistical uncertainty. The numerical year 31,923 BCE therefore should not be mistaken for an anniversary of a hunt, a migration or a climatic event. It is best understood as an approximate point within a long Upper Palaeolithic world, perhaps roughly 34,000 years before the conventional radiocarbon “present” of 1950.
