The Doneraile Bone Puzzle – Emerald Archives

Dateline

The label “31,851 BCE” gives a useful point of reference for a modern chronology, but it must not be mistaken for a date that people then knew or recorded. No calendar-year event can be assigned to Ireland at this depth of time. Expressed in the archaeological convention of calibrated years before AD 1950, the nominal date falls at roughly 33.8 thousand calendar years BP, within Marine Isotope Stage 3 (MIS 3), the long and climatically unstable middle part of the last glacial cycle. Evidence for Ireland at this point comes not from chronicles, settlements or named peoples, but from cave bones, buried lake deposits, pollen, mosses, insects, sediments and the changing reach of ice.

A Date Within a Range

Radiocarbon dating does not yield a single, self-evident ancient calendar year. A laboratory measurement must be calibrated against independently constructed curves, and dates this old commonly produce ranges rather than a neat historical timestamp. In addition, the surviving Irish evidence was rarely laid down in one undisturbed sequence. A bone may date an animal’s death, not its arrival at a cave; a cave floor can combine material accumulated over many centuries or millennia; and early excavators often had neither the methods nor the records now expected.

The responsible question, therefore, is not “what happened in Ireland in 31,851 BCE?” but what conditions and evidence illuminate Ireland around the later part of MIS 3. The answer is necessarily regional and probabilistic. It reveals a cold, open and changing island-edge landscape, intermittently accessible to large mammals and perhaps, though far less securely, to small groups of people. It does not reveal an Irish community, a dynasty or a continuous national population.

Before the Ice-Sheet Maximum

This was not yet the Last Glacial Maximum, the period when the British–Irish Ice Sheet attained its greatest Late Pleistocene extent, broadly between about 27,000 and 21,000 years ago. Ice growth was nevertheless becoming an increasingly important force in the lands and seas around Ireland. Ice-sheet reconstructions show that expansion was neither simple nor simultaneous around every coast: glaciers, ice streams and upland ice masses advanced and retreated in response to an erratic North Atlantic climate. Research on the wider ice sheet places important growth phases after about 35,000–32,000 years ago, while the precise configuration over Ireland at any one moment remains difficult to recover.

At roughly 33.8 thousand calendar years BP, Ireland should not be imagined as uniformly buried beneath an ice cap. Some regions were ice-free or seasonally exposed; elsewhere, advancing ice and periglacial conditions remodelled the ground. The later ice sheet scoured, transported and buried earlier deposits, helping to explain why the island’s record of this period is so fragmentary. What survives is often preserved in sheltered limestone caves or trapped beneath later glacial sediments.

Sea level was substantially lower than today, so the outlines of coast and shelf differed greatly from the modern map. Broad areas now beneath the Celtic Sea, Irish Sea and Atlantic shelf were dry land or a shifting coastal plain at various stages of the glacial cycle. Such change mattered to animal dispersal. Yet it would be misleading to draw a single permanent land bridge on a map of this date: the nature, duration and usability of routes between Ireland, Britain and continental Europe varied with sea level, ice and climate. The presence of mainland species in Ireland demonstrates episodes of connection or crossing, not a continuously open highway.

Castlepook and Its Animals

The most important evidence close to this nominal date comes from Castlepook Cave, also called Mammoth Cave or Doneraile Cave, in north County Cork. The limestone cave system has produced Ireland’s richest known assemblage of vertebrate remains from the period before the Last Glacial Maximum. Its Ice Age fauna includes woolly mammoth, reindeer, giant deer, brown bear, wolf, Arctic fox and spotted hyaena. These finds make a powerful correction to any picture of Pleistocene Ireland as a biologically empty, uniformly frozen place.

They should not, however, be treated as a photograph of one afternoon on a Cork hillside. Carnivores could bring prey or scavenged carcass parts into a cave; animals could die in or near openings; water could move bones; and roof collapse, stalagmite growth and later disturbance could rearrange deposits. Castlepook’s hyaenas are especially significant because a den or habitual carnivore refuge can accumulate a mixed assemblage. Teeth marks and broken bones may record feeding and transport rather than a human hunt. The cave is an archive, but an archive assembled by natural processes as well as by living creatures.

Modern re-dating has sharpened this caution while strengthening the broad case for a pre-maximum Irish fauna. A 2020 reassessment found that several Castlepook bone dates were substantially older than previously believed, with some results shifted by thousands of radiocarbon years after improved sampling and pretreatment. The study suggested that giant deer and spotted hyaena may have disappeared locally before about 40,000 calibrated years BP. Thus some species listed from the cave need not have overlapped precisely with the nominal year. Reindeer and other cold-adapted mammals nonetheless show that suitable habitats and routes existed during MIS 3.

The ecological setting was probably a mosaic of herb-rich tundra, grassland, sedge-rich wet ground, bare mineral surfaces and low scrub, rather than woodland. Reindeer were well suited to open country and seasonal movement; mammoths, where present, belonged to a wider cold-steppe fauna. Predators and scavengers followed the opportunities supplied by large herbivores. A landscape of this sort could support animals in considerable numbers at favourable moments, but its resources would have been seasonal, patchy and vulnerable to abrupt climatic shifts.

Derryvree’s Buried Landscape

Evidence from Derryvree near Maguiresbridge in County Fermanagh offers a different kind of window. There, roadworks exposed freshwater silts and fine sands between two distinct till sheets within a drumlin. The organic deposits produced a radiocarbon result of 30,500 years BP, with a sizeable laboratory uncertainty, and contained plant and animal remains described as having a full-glacial character. The date is not an exact match for 31,851 BCE after calibration, but it belongs to the same broad Middle Midlandian environmental world and demonstrates why individual Irish calendar years cannot be reconstructed.

The Derryvree flora included poorly preserved herb pollen and moss macrofossils. Together with the sedimentary setting, these support the interpretation of open tundra and a periglacial regime: ground subject to severe frost, short growing seasons and sparse tree cover. Freshwater pools and channels nonetheless existed. In a cold country, such wetlands could be locally productive, supporting mosses, sedges and insects, while also preserving traces that open slopes later lost to erosion or ice. Derryvree is not evidence for a settlement; it is evidence for habitat.

The Human Question

The most arresting issue attached to this period is a reindeer bone from Castlepook reported as bearing marks consistent with cutting by a stone tool. It has been discussed as possible evidence for human activity in Ireland roughly 33,000 years ago. If that interpretation is ultimately confirmed in secure context, it would move the island’s known human story far back into the Upper Palaeolithic and place a brief human presence in an era when modern humans were established elsewhere in western Europe.

At present, that conclusion remains provisional. The bone derives from old cave investigations rather than a sealed, modern excavation context. It is not accompanied by a securely associated stone-tool assemblage, hearth, dwelling floor, human remains, art or a dedicated animal-processing area. A 2022 excavation at Castlepook sought such corroborating evidence and reported no cultural artefacts. This does not prove that people never reached Ireland in MIS 3; it means that the case cannot responsibly be expanded into a settled population, a repeated occupation or a named archaeological culture.

Comparison is helpful. The best-supported Palaeolithic human evidence presently known from Ireland is much later: a cut-marked brown-bear kneecap from Alice and Gwendoline Cave, County Clare, dating to the Late Upper Palaeolithic, around 12,800–12,600 calibrated years BP. That find shows that brief human visits can leave extremely slight traces. It also emphasises the gulf of more than twenty millennia between the late-glacial evidence and the disputed implication of the Castlepook bone.

No Kingdoms, No Recorded Beliefs

There is no evidence from around 31,851 BCE for political affairs, diplomacy, warfare between human groups, leaders, territories or institutions in Ireland. Nor can religion, language, teaching, art, kinship rules or social rank be recovered. In the absence of a securely established human occupation, even ordinary questions about houses, clothing, food-sharing and technology have no specifically Irish answer. It would be wrong to import the better documented cultures of France, Iberia or Britain and present them as Irish fact.

If people did visit, the most restrained inference is that they were mobile foragers capable of moving through demanding northern environments and exploiting animal resources. A cut-marked bone, if humanly modified, might indicate a short episode of carcass processing rather than a camp. It cannot show whether the visitors arrived by land, sea ice, boat or another route; nor whether they returned. Such possibilities belong to interpretation, not established evidence.

An Environmental History, Not Yet a Human One

For this remote point in Ireland’s past, the central development is environmental. North Cork caves preserve the remains of mammals that reached a cold, open Ireland during the unsettled climates of MIS 3. Fermanagh sediments preserve tundra plants and freshwater conditions later entombed beneath glacial deposits. The expanding British–Irish Ice Sheet was beginning to transform the terrain and would subsequently erase or conceal much of what had gone before.

The Doneraile bone puzzle belongs within that larger history. It is important precisely because it may hint at an unexpected human episode, but its importance should not be confused with certainty. Around 31,851 BCE, Ireland can be described as a changing Pleistocene landscape inhabited at intervals by cold-climate fauna. Whether people briefly shared that landscape remains an open archaeological question, one that future excavations, refined dating and careful study of old museum collections may yet answer.

Primary Sources

Tags: County Cork, Doneraile Bone Puzzle, Ireland Year by Year, Irish Ice Shee, Last Glacial Maximum, The Doneraile Bone Puzzle