The Castlepook Femur – Emerald Archives

The numerical label 31,924 BCE cannot denote an Irish historical year in the ordinary sense. No people on the island used such a calendar; there are no texts, rulers, settlements or events that archaeology can assign to it. Measured against the conventional radiocarbon reference date of AD 1950, it falls roughly 33,900 calendar years before present, or 33.9 ka cal BP. At that depth of time, even good scientific dates are ranges, not anniversary dates. The responsible subject is therefore Ireland during a broad interval around the later part of Marine Isotope Stage 3, within the Late Pleistocene and Upper Palaeolithic of north-west Europe.

Evidence Rather Than Annals

The evidence is exceptionally uneven. Ireland has no securely identified house, hearth, grave, tool assemblage or settlement surface of this age. Its record comes principally from animal bones, cave deposits, glacial sediments, landforms and changing shorelines. These are powerful sources for reconstructing environments, but they do not provide an annal of a particular hunt, migration or season. Much of the island’s earlier surface archaeology was also vulnerable to erosion, burial and destruction as the last ice sheet expanded. Absence of a camp is not conclusive proof that no person ever entered Ireland; equally, it cannot be converted into proof of a settled population.

The central place in this inquiry is Castlepook Cave, near Doneraile in north County Cork. Excavations there between 1904 and 1918 recovered a large, historically accumulated collection of bones from more than one deposit and from a cave system later affected by water and disturbance. Geological Survey Ireland describes Castlepook as Ireland’s richest vertebrate-fauna site for the period leading towards the Last Glacial Maximum. Direct dates on individual bones span a very long interval, approximately 45,700 to 19,950 years BP. That range itself is a warning: the cave was a repository used or entered by animals over millennia, not a single frozen moment in time.

A Cold, Changing Island

Around 33.9 ka cal BP, Ireland lay within an unstable glacial world, but the great maximum of the last British–Irish Ice Sheet still lay ahead. Reconstructions differ in detail because ice advanced and retreated differently across regions, and because dates from land, seabed and cave sequences do not all resolve the same intervals. The broad picture is secure: ice-sheet growth was under way during the last glaciation, while full coverage of Ireland and its shelf belongs chiefly to the later maximum, generally placed around 27,000–23,000 cal BP. It would be misleading to depict all Ireland in 31,924 BCE as a uniform, permanent white desert, or to draw a precise ice margin for that nominal year.

What can be said is that conditions were colder, more seasonal and generally more open than those of modern Ireland. Treeless or sparsely wooded communities of grasses, herbs, sedges and low shrubs are more plausible than the familiar oak-and-hazel woodland of the Holocene. Wind, frost, snow cover and frozen or waterlogged ground would have affected where animals could graze and where people, if present, could travel. Yet conditions changed repeatedly within Marine Isotope Stage 3. A cave bone dated to a broad interval cannot by itself supply the weather of one Cork valley in one notional year.

Sea level was substantially lower than it is today because large volumes of water were stored in global ice sheets. The continental shelves around Ireland were consequently more exposed in places, altering coasts and possible routes for mammals. But low sea level did not automatically make an easy, continuous dry corridor from Ireland to Britain or mainland Europe. The Irish Sea was shaped by complex topography, local ice and relative sea-level change. Later modelling demonstrates how disputed and short-lived any connections could be. For this earlier period, an exposed coastal plain, narrow crossings, seasonal ice, or several changing routes are possibilities rather than demonstrated pathways.

Castlepook’s Animal World

Castlepook nonetheless proves that sizeable mammals reached parts of Ireland before the Last Glacial Maximum. Its Ice Age fauna includes woolly mammoth, giant deer, reindeer, brown bear and spotted hyaena; later deposits also contain Arctic fox, wolf and lemmings. These names should not be treated as a complete list of animals alive beside one another in 31,924 BCE. Some remains are separated by many thousands of years, and revised laboratory techniques have altered earlier chronologies. The strength of the collection is cumulative: it demonstrates repeated occupation of Irish landscapes by cold-stage fauna, not a single perfectly dated ecosystem.

The most evocative interpretation of the cave is as a carnivore-associated bone assemblage. Hyaenas could transport and gnaw bones, while water movement and earlier excavation mixed materials further. A mammoth bone in a cave does not mean that a mammoth died in that chamber; it may record scavenging, transport or redeposition. Nor should the modern label Irish elk obscure the point that giant deer was one element of a changing Pleistocene fauna, not an emblem of an already Irish national landscape. The animals reveal ecological opportunity as well as hardship: grazing and browsing resources supported herbivores, whose carcasses in turn supported predators and scavengers.

Reindeer are especially relevant. Their presence suggests open country capable of supporting highly mobile herd animals, and such herds could have provided a compelling resource for Upper Palaeolithic hunters elsewhere in Europe. But mobility matters. A herd passing through a region does not establish a permanent human community there, and the sparse Irish fauna indicates a more restricted record than that available from many British or continental sites. Ireland was on the north-western edge of a wider Ice Age world, rather than a documented centre of sustained population.

The Mark on a Reindeer Bone

One Castlepook object brings the human question unusually close to this article’s date: a reindeer femur reported by Ruth Carden and County Cork heritage publications as bearing a mark made by a sharp stone tool during butchery. The publication places it at about 31,000 BC, broadly close to but not precisely identical with 31,924 BCE. If the mark is correctly identified as anthropogenic, it is important evidence that a person or people processed a reindeer in Ireland during the Upper Palaeolithic, vastly earlier than the long-established Mesolithic settlement record.

Its implications must remain proportionate. A cut mark is physical evidence on one bone; a visiting hunting party, seasonal pursuit of reindeer, or a group travelling from Britain are interpretations built around it. The bone does not demonstrate that hunters killed the animal at Castlepook, occupied the cave, lived nearby or returned repeatedly. Indeed, the bone may have reached the cave after carcass processing elsewhere, carried by a carnivore or moved through natural processes. No securely associated stone tools, hearth, dwelling, human remains or intact occupation floor of comparable age has yet been recovered. The National Museum of Ireland continues to describe clear Palaeolithic human presence as unproven, while the 2022 excavation at Castlepook found no cultural artefacts.

This is not merely excessive caution. Castlepook’s early excavations did not meet modern standards of stratigraphic recording, and subsequent research has shown why old cave-bone dates need reassessment. Improved pretreatment and radiocarbon methods produced substantially older dates for several Castlepook specimens than earlier results, altering the timing of animals including giant deer and spotted hyaena. Future work on surviving deposits, microscopic analysis of modified bone, ancient biomolecules and new direct dating may either strengthen or qualify the human interpretation. The cave is valuable precisely because it preserves a question that can still be tested.

Society Without a Society?

There is no evidence for political authority, chiefs, territories, diplomacy, organised warfare, trade networks, religious rites or formal learning in Ireland at this date. Such categories require identifiable communities and social institutions; none can responsibly be reconstructed from the present record. There were no Irish kingdoms, no named peoples and no historical borders. Even language is beyond recovery. To assign later Celtic, Gaelic or national identities to these remote possible visitors would be an anachronism.

If the reindeer mark represents human butchery, the most that follows securely is the skilled use of a sharp cutting edge and knowledge sufficient to process animal tissue. Comparable Upper Palaeolithic groups in Europe were mobile hunter-gatherers who organised work, food-sharing, tool maintenance and knowledge of seasonal landscapes. That broader comparison can suggest the kind of practical social life possible in Ireland, but it cannot tell us the number, kinship, beliefs, speech or destination of the people involved. It is safer to speak of an episode of activity than of an Irish society.

Nor can an economy be described in the later sense of farming, livestock, markets or long-distance commerce. There is no evidence for cultivation, domesticated animals, pottery, metalworking or built settlement. Any human subsistence would have depended on wild resources, especially if hunters followed reindeer or exploited a carcass. The important economic relationship visible in the evidence is ecological: animal movement linked grazing grounds, predators, scavengers and perhaps occasional human foragers across landscapes whose coastlines and ice barriers changed through time.

Approaching the Ice Maximum

The future of this landscape was severe. In the millennia after the broad interval represented by 31,924 BCE, the British–Irish Ice Sheet expanded to cover all or almost all of Ireland and reached far onto the surrounding Atlantic and Celtic Sea shelves. Glacial erosion and deposition remade terrain, helping explain why early traces of people are so scarce and difficult to interpret. The later disappearance or local extirpation of particular animals also shows that Irish Pleistocene ecology was repeatedly broken and reassembled rather than steadily developing towards the modern fauna.

Thus the most historically meaningful account of 31,924 BCE is neither an invented first settlement nor an empty blank. It is a threshold within a long Late Pleistocene sequence. Cave bones from north Cork show that Ireland could support remarkable cold-stage animals; one marked reindeer femur raises the prospect of a fleeting human encounter with that world. The evidence does not yet reveal a community, a culture or an event in a calendar year. It reveals an island of shifting opportunities at the edge of the Ice Age, whose earliest possible human history survives in fragments and must be allowed to remain fragmentary.

Primary Sources

Tags: County Cork, Early Holocene, Ireland Year by Year, Irish Ice Shee, Last Glacial Maximum, The Castlepook Femu