Crag Cave’s Thaw Windows – Emerald Archives

Dateline

The label 31,652 BCE is a modern chronological convenience, not a year that anyone in Ireland could have recognised or recorded. There were no calendars, writing systems, named polities or dated events from which to construct a year-by-year narrative. Arithmetically, it falls about 33,602 years before AD 1950, the conventional ‘present’ used in radiocarbon reporting. That calculation should not be mistaken for archaeological precision: at this depth of time, radiocarbon measurements require calibration and are normally expressed as broad probability ranges. The responsible question is therefore what evidence can illuminate Ireland around this horizon.

The answer belongs to the Late Pleistocene, within Marine Isotope Stage 3 (MIS 3), a long and climatically unstable part of the last Ice Age. Ireland’s record is exceptionally fragmentary. Subsequent glaciers eroded old ground surfaces, moved sediments and buried others beneath till; early cave excavations also removed much material before modern standards of recording existed. No securely dated Irish hearth, dwelling, burial, tool-making floor or settlement can be assigned to 31,652 BCE. There is consequently no evidential basis for describing rulers, diplomacy, warfare, formal religion, agriculture or an established trading system. What survives instead is a difficult archive of sediment, cave bone and cave calcite.

An Interval Rather Than a Date

The nearest useful terrestrial evidence comes from Derryvree near Maguiresbridge in County Fermanagh. Roadworks exposed a freshwater deposit of organic silts and sands trapped between two bodies of glacial till. Such a sequence matters because it indicates an interval in which water, vegetation and sediment accumulated before later ice overran the area. A radiocarbon determination from the organic material, once calibrated, has been presented as a broad range of roughly 36,900–31,700 calibrated years before present. The nominal year considered here lies within that window, but Derryvree does not document a particular year, season or human generation. It is a time-depth record spanning many centuries or millennia, with uncertainties inherent in both dating and correlation.

Its plants and insects nevertheless offer a restrained picture of the landscape. The evidence indicates open, cold-country vegetation: mosses, sedges, grasses, herbs and low shrubs rather than a closed woodland. Wet hollows and shallow freshwater could support localised plant communities even where the wider terrain was exposed and treeless. In modern terms, the nearest analogy is not a single modern habitat but a mosaic of tundra-like ground, sedgy wetland, stream margins, bare mineral soil and periglacial surfaces. Snow cover, wind exposure, soil moisture and summer warmth would have made sharp differences over short distances.

This was not necessarily a uniformly frozen Ireland. MIS 3 was marked around the North Atlantic by abrupt shifts between relatively milder interstadials and harsher stadials. Glaciers advanced and withdrew, river regimes changed and suitable habitats expanded or contracted. The Irish deposits are too sparse to map those shifts year by year across the island. They do, however, warn against two equally misleading pictures: an Ireland continuously buried beneath ice, or an Ireland steadily open and hospitable. Conditions varied by region and through time. At roughly 33,600 calibrated years before present, Ireland was still before the later local maximum of the British–Irish Ice Sheet, generally placed around 27,000 years ago, but was moving through an era in which ice margins and ice-free refuges could change dramatically.

Liquid Water at Crag Cave

Crag Cave in County Kerry supplies a second, very different environmental archive. Uranium-series dating of broken and reworked speleothems—stalagmites, stalactites and flowstone—shows episodes of calcite growth through much of the last glacial period, including the latter part of MIS 3. Calcite formation in this shallow cave required liquid water to percolate through soil and limestone and is therefore evidence against permanently frozen ground at those particular times. It does not prove warm or forested conditions, and the cave dates cannot furnish a weather report for 31,652 BCE. Yet they show that south-west Ireland could experience milder intervals in which water moved through the karst and biological carbon dioxide again entered the cave system.

The contrast between Derryvree’s northern freshwater sediments and Crag Cave’s southern calcite is important. Neither site stands for all Ireland, but together they demonstrate environmental diversity and change. Cave growth ceased in colder phases, while freshwater basins could be overwhelmed, frozen, drained or later covered by glacial deposits. The landscape that survives in the record is therefore one of interruptions: brief windows for sedimentation or calcite growth between colder episodes that destroyed, displaced or concealed evidence.

Animals Crossing a Changing Land

Irish cave collections provide the strongest evidence that sizeable mammals occupied parts of the island during the broader pre-Last Glacial Maximum period. Castlepook Cave, near Doneraile in north County Cork, is particularly significant. Material collected there includes reindeer, brown bear, spotted hyaena, wolf or fox, hare, lemmings, giant deer and woolly mammoth remains. Not every species need have lived there at the same moment: a cave is a trap, shelter, den and bone-collecting place, not a photograph of one living community. Many of the original deposits were disturbed during antiquarian excavation, so associations once assumed between bones must be treated cautiously.

Radiocarbon dating undertaken by the Irish Quaternary Fauna Project and later reassessment nevertheless place much of Ireland’s cave fauna within the wide span from at least about 45,000 to 20,000 years ago. This establishes repeated animal movement into Ireland during favourable phases of the Late Pleistocene. Reindeer and cold-adapted small mammals suit the open environments suggested by Derryvree; hyaena and large herbivores show that a more complex food web could operate when habitat and access allowed. Carnivores may have accumulated bones in caves, while natural deaths, water transport and later disturbance added further material. A faunal list is therefore evidence for presence, not proof of a stable ecosystem at one fixed date.

How these animals reached Ireland remains an important question. Lower global sea levels altered the Irish Sea basin and reduced marine barriers, but changing shorelines, estuaries, ice margins and tidal channels mean that a simple, permanent ‘land bridge’ should not be imagined. Some movements may have been possible over exposed terrain or ice-associated routes; others may have depended on temporary connections or narrow crossings. The cave fauna shows that access occurred often enough for animals to arrive. It cannot, by itself, specify the route, season or duration of every migration.

The Castlepook Human Question

Castlepook also contains the most provocative possible clue to people in Ireland at this remote horizon: a reindeer femur from an old collection bearing a deep surface mark interpreted by Ruth Carden as consistent with cutting by a sharp stone tool. The bone has been widely reported as approximately 33,000 years old. If the mark is confirmed as anthropogenic and securely related to the animal’s age, it would push evidence for human activity in Ireland far earlier than the well-established post-glacial record.

That conclusion remains deliberately provisional. The specimen was not recovered from a newly excavated, undisturbed camp floor alongside a diagnostic stone-tool assemblage, hearth or butchered animal group. Old cave collections have incomplete contextual information, and bone marks can require detailed taphonomic study to distinguish human cutting from carnivore damage, sediment abrasion, excavation damage or other processes. The responsible interpretation is not that Ireland possessed a settled Upper Palaeolithic population in 31,652 BCE. It is that a single object raises the possibility of a brief human visit or butchery episode in the wider period, a hypothesis that continuing work at Castlepook may test.

If people did reach the island, their presence would almost certainly have been small-scale and mobile. Across north-west Europe, human groups at this time were hunter-gatherers who relied on close knowledge of animals, seasonality, stone resources, shelter and routes through dangerous terrain. That wider comparison can suggest possible ways of life, but it cannot supply missing Irish facts. No Irish evidence presently identifies a camp, family group, language, ritual practice, exchange partnership or political relationship. Absence of material proof does not demonstrate an absence of thought, social bonds or belief; it defines what history can responsibly claim.

No Polities, But a Living Landscape

In political terms, then, there were no demonstrable Irish states, chiefs, borders, alliances or conflicts. In economic terms, there is no evidence for farming, herding, metalworking, markets or regular long-distance trade. ‘Settlement’ must be used with particular care: the archaeological record does not establish permanent habitation. The important history is environmental and biogeographical. Freshwater plants at Derryvree, episodic drip-water calcite at Crag Cave and the animal remains of Castlepook show that parts of Ireland repeatedly became habitable for plants and animals during an otherwise severe glacial world.

The most important development around this nominal date was therefore neither a recorded event nor a human institution. It was instability itself: the opening and closing of ecological opportunities as climate altered water supply, vegetation, ice cover and routes into the island. Ireland was a shifting western edge of the European Pleistocene world. Its surviving evidence is thin, but it is enough to replace the old image of a featureless frozen void with a more credible one—cold, exposed and often hostile, yet intermittently watered, inhabited by animals and perhaps, still uncertainly, visited by people.

Primary Sources

Tags: Castlepook Cave, County Cork, Crag Cave, Ireland Year by Year, Last Glacial Maximum, Late Pleistocene