Derryvree’s Arctic Lake – Emerald Archives

Dateline

The label 31,895 BCE is useful only as a modern point of reference. It cannot denote a recorded Irish year, season, event or generation. Around this notional date—roughly 33.8 thousand calendar years before AD 1950—there are no written sources, named people, settlements or calendar systems for Ireland. Archaeology instead works with ranges produced by radiocarbon dating, stratigraphy, pollen, insects, animal bone, cave mineral deposits and reconstructions of ancient ice. Those ranges are often broad, and they are revised as techniques improve. The responsible history of Ireland at this depth of time is therefore a history of evidence, not a narrative of happenings on a particular day.

A Landscape Within MIS 3

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.

Recent reconstructions make an important correction to older, simpler pictures of an Ireland either wholly frozen or wholly open. The British–Irish Ice Sheet was dynamic and geographically uneven. Research on raised beaches, offshore deposits, glacial landforms and dating evidence indicates that ice advanced and retreated in different sectors at different times during MIS 3. Some areas may have supported ice or lain close to an ice margin while other districts retained, or regained, ice-free ground. That does not permit a precise map for 31,895 BCE. It does mean that Ireland should be envisaged as a changing northern landscape, not as a single, static white expanse.

At approximately 33.8 ka cal BP, conditions were generally cold, exposed and highly seasonal. Much of the island’s modern surface did not yet exist in its present form: later glaciers would scour, redeposit and conceal older deposits. Sea level was lower globally, exposing more of the continental shelf, but the exact configuration of coasts, sea passages, grounded ice and marine waters around Ireland depended on local ice loading as well as global sea level. It would be unsafe to claim a simple dry-land corridor either to Britain or to continental Europe. Movement of animals, and any people, must be considered against a shifting and poorly resolved physical geography.

Derryvree’s Buried Record

The most directly relevant environmental archive is at Derryvree near Maguiresbridge, County Fermanagh. There, organic-rich freshwater sediments were found sandwiched between glacial tills: material laid down during an ice-free interval and later buried beneath renewed glacial deposits. Published radiocarbon evidence from the sequence has been calibrated to a broad range of about 36.9–31.7 thousand calendar years before present. The modern conversion of 31,895 BCE falls within that range, but the range must not be mistaken for exact dating of the deposit to that numbered year.

Derryvree offers unusually tangible evidence of a freshwater environment in northern Ireland during part of MIS 3. Plant remains, pollen and insect evidence have been interpreted as indicating an open, cold landscape with a sparse tundra or tundra-steppe vegetation cover rather than closed woodland. Grasses, sedges, herbs, mosses and low shrubs could exploit sheltered ground, lake margins and seasonally moist soils. Such vegetation supported a food web very different from the later wooded island, and it also helped stabilise only parts of a landscape exposed to frost action, wind and sediment movement.

The site’s importance lies as much in its limitations as in its discoveries. Derryvree records a particular waterbody and its catchment, not the entire island. Pollen can travel; insects and plants respond to local habitats; and sediment may have been disturbed before burial beneath ice. Nevertheless, the sequence demonstrates that at least one northern Irish locality held freshwater, vegetation and biological activity during a cold-stage interval broadly overlapping this chronological horizon. It rules out any easy assumption that every part of Ireland was continuously sealed beneath glacier ice.

Animals of the Open Country

Caves in southern Ireland provide the principal evidence for the island’s larger Late Pleistocene animals. Castlepook Cave, near Doneraile in north County Cork, produced a remarkable collection during excavations conducted mainly in the early twentieth century. Its recorded fauna includes reindeer, giant deer, woolly mammoth, spotted hyaena, brown bear, wolf, fox, mountain hare and lemmings. These are not a photograph of one moment. A cave assemblage can accumulate over long periods as animals die nearby, predators drag in carcass parts, scavengers disturb bones and water shifts sediments.

Modern redating has made Castlepook more significant and more complicated. Improved pretreatment of bone collagen produced dates substantially older than some earlier results, placing elements of the fauna in MIS 3. The revised work suggests that giant deer and spotted hyaena may have disappeared locally before about 40,000 cal BP, while other material represents later episodes. Thus, the animal list should not be treated as a guaranteed community living together in 31,895 BCE. It does show, however, that Ireland could support, at different points in this broad interval, a cold-adapted assemblage of grazers, browsers, carnivores and small mammals.

Reindeer are particularly evocative because they indicate landscapes capable of sustaining mobile herbivores across extensive open ground. Mammoth remains point to grass- and herb-rich habitats somewhere within the wider region, while hyaena bones and gnawing illuminate predation and scavenging. A hyaena den is not a hunting camp. Its contents may represent prey transported to shelter, naturally dead animals and material mixed by later processes. This distinction matters: cave bone deposits reveal animal behaviour and palaeoecology more securely than they reveal human activity.

Regional contrast was likely pronounced. Fermanagh’s lake sediments preserve a northern freshwater setting. Cork’s limestone caves collected bones in a southern landscape whose local topography offered crevices, chambers and shelter. Elsewhere, ice, periglacial ground, bare sediment, coastal margins and patches of vegetation created different opportunities and constraints. Rather than a familiar green island, Ireland was a mosaic of cold environments, constantly affected by climate shifts and ice-sheet behaviour.

The Unsettled Human Question

No securely excavated settlement, hearth, stone-tool assemblage, house floor, burial or artwork can presently be assigned to Ireland around 31,895 BCE. Accordingly, there is no sound basis for describing an Irish population, its language, kinship, dress, beliefs or social rank. There were no states, kingdoms, diplomacy, laws or political institutions detectable in the evidence. There is equally no evidence for farming, domestic animals, fields, trade networks or permanent villages. The appropriate entries for political affairs and economy at this date are not “unknown details”, but an absence of evidence sufficient to establish that such institutions existed in Ireland.

There is, however, a serious and closely watched possibility. A reindeer femur from Castlepook, recovered from older excavations and re-examined by Ruth Carden, bears a mark interpreted as having been made by a sharp stone tool during butchery. Dating associated with this find has placed it broadly around 33,000 years ago. If the mark is confirmed as humanly made and securely related to its date, it would push evidence for human activity in Ireland far back into the Upper Palaeolithic, long before the formerly accepted post-glacial horizon.

That conditional language is essential. The bone was not recovered in a modern, intact archaeological context alongside the tool, a hearth or a living surface. Natural damage, trampling, carnivore activity and excavation history must all be excluded through microscopic and contextual study. Recent work at Castlepook has found surviving sediments and renewed the search, but the 2022 excavation reported no cultural artefacts. The claim is therefore promising evidence under investigation, not proof of a settled human presence or a licence to populate Ireland with imagined Ice Age bands.

If people did reach the island during this phase, the most cautious interpretation is of brief, highly mobile visits connected to the pursuit or processing of animals, rather than a demonstrable resident society. This is an inference from the nature of the proposed bone modification and from wider Upper Palaeolithic lifeways in north-west Europe; it is not a fact established for Ireland. The absence of securely associated tools makes even that restrained possibility provisional.

Culture Beyond Recovery

Religion, art and learning leave no identifiable Irish trace at this horizon. European Upper Palaeolithic communities elsewhere created ornaments, images, specialised tools and ritualised deposits, but those achievements cannot responsibly be transferred across the sea to Ireland without Irish evidence. A cave need not be a shrine; a mammoth bone need not signal hunting by people; an old mark on bone need not constitute a cultural tradition. The same caution applies to conflict. Carnivores competed for prey and shelter, but no evidence records warfare or organised violence between human groups in Ireland.

What can be said is that survival in this landscape, for any animal or occasional human visitor, depended upon intimate environmental knowledge: sheltered routes, freshwater, seasonal forage, animal movements, weather and the risks of ice, exposure and predators. Such knowledge is a plausible human capacity, but its Irish expression remains archaeologically invisible. The material record preserves the stage more clearly than the actors.

How This Distant Past Is Dated

The difficulty of writing this article is itself historically important. Radiocarbon measurements are not calendar dates and must be calibrated against changing atmospheric carbon levels. Calibration curves can shift the calendar placement of old results, while improvements in collagen purification can transform the apparent age of cave bone. Stratigraphy adds another relative sequence—one deposit above or below another—but glaciers can erode and rearrange that sequence. U-series dates from cave calcite, optically stimulated luminescence dating of sediments, and palaeoenvironmental proxies each add evidence with their own uncertainties.

For 31,895 BCE, Ireland’s history is consequently an account of climate variability, freshwater deposits, animal remains and unresolved archaeological potential. Derryvree shows that cold-stage life endured around a northern lake during a broad interval encompassing this modern date. Castlepook shows that a distinctive Ice Age fauna reached southern Ireland, and possibly that people encountered or processed reindeer there. Neither site yields a named event in a named year. Their real value is greater: they preserve rare traces from an Ireland on the threshold of deeper glaciation, before the later ice erased so much of the island’s earliest past.

Primary Sources

Tags: County Cork, Derryvree’s Arctic Lake, Early Holocene, Ireland Year by Year, Irish Ice Shee, Last Glacial Maximum