The Lower Paleolithic is the earliest documented phase of human occupation in the territory of modern Georgia. Its archaeological record begins about 1.8 million years ago at the Dmanisi site, where stone artifacts and fossils of early Homo provide one of the earliest securely dated records of hominins outside Africa.
After the Early Pleistocene occupations of the Dmanisi region, securely dated evidence remains sparse until the Middle Pleistocene, when Acheulean assemblages are documented at high-altitude caves in the central Greater Caucasus. At Kudaro III, layer 8a yielded a small assemblage assigned to the Acheulean and a single thermoluminescence date of thousand years ago; the limited material and single date leave its chronology uncertain. Acheulean occupation is more securely documented at Kudaro I between approximately 360,000 and 245,000 years ago. Tsona has no direct numerical date and is placed around 350,000 years ago through comparison with Kudaro I. The open-air sites of Ziari preserve flake-and-core assemblages without bifacial handaxes, differing technologically from the handaxe-bearing cave assemblages.
Early and Middle Pleistocene environments
Environmental evidence associated with the Georgian Lower Paleolithic is geographically and chronologically uneven. The Early Pleistocene is best documented around Dmanisi and Orozmani in southern Georgia, whereas evidence associated with the Middle Pleistocene Acheulean comes principally from the high-altitude caves of Kudaro and Tsona.
Early Pleistocene Plant remains from the lower deposits at Dmanisi and Orozmani indicate a temperate landscape combining grassland and woodland under relatively warm and humid conditions. Pollen from the lower Orozmani sequence represents oak–hornbeam woodland, riverside forest containing wingnut, and higher forests containing chestnut and beech. Phytoliths—microscopic silica bodies formed in plants—show that both temperate and warmth-adapted grasses were present.
Near the end of the Olduvai subchron, about 1.78 million years ago, the plant evidence records reduced rainfall and expanding open vegetation. Warmth-adapted grasses declined, while indicators of water stress increased. During the documented hominin occupation, grasses and other plants adapted to dry conditions dominated locally, although woodland remained in the surrounding region and on higher ground. Messager and colleagues reconstructed a warm, relatively dry steppe-forest. Although its open structure resembled African savanna environments, the identified plants were predominantly temperate Eurasian species.
Small mammals, amphibians and reptiles support this reconstruction. Dry open ground and rocky or bushy terrain occurred alongside wetlands, streams and other permanent bodies of water, with limited woodland and gallery forest nearby. The amphibian and reptile assemblage indicates a warm, seasonally dry climate broadly comparable to that of the modern Mediterranean. Estimates based on the climatic ranges of related living species produced a mean annual temperature of about and annual precipitation of approximately .
Middle Pleistocene Environmental evidence associated with the later Lower Paleolithic comes principally from the Acheulean deposits of Kudaro and Tsona. These caves lie on the southern slopes of the Greater Caucasus at elevations between approximately . Their pollen and animal remains record several changes in local conditions rather than a single Middle Pleistocene environment.
Pollen from Kudaro I layer 5 records alternating climatic phases. Cold conditions at the top of sublevel 5v were followed by a climatic optimum. The base of sublevel 5b was humid and predominantly forested, resembling modern Colchic forest, whereas its upper part was warm and dry and contained plants associated with steppe and savanna environments. The base of sublevel 5a was cold, followed by temperate and then cooler conditions. At Tsona, rodents indicate warm and dry conditions, while pollen places the cave within an upper forest belt, with conifers and deciduous trees growing on the surrounding slopes.
Mgeladze and Moncel concluded that the Acheulean occupations of Kudaro and Tsona occurred during temperate intervals, when high mountain valleys were accessible from the lower plateaus and a larger area of the Greater Caucasus was available to hominin groups. They contrasted this with glacial periods, when the available highland territory contracted. These environmental reconstructions remain broadly dated: Tsona's Acheulean levels have no numerical age, while the dating estimates for Kudaro I layer 5 have wide uncertainty ranges.
Early Pleistocene occupation Dmanisi
Dmanisi lies on a promontory above the confluence of the Mashavera and Pinezauri rivers in southern Georgia. Its Paleolithic deposits accumulated on a basalt platform beneath the later medieval town and are divided into Strata A and B. Stone artifacts occur at several levels in Stratum A, recording repeated visits rather than a single occupation. In 2011 these levels were dated to between 1.85 and 1.78 million years ago; direct dating of the uppermost basalt beneath them later refined the interval to approximately 1.80–1.78 million years ago. Stratum B accumulated after a reversal of Earth's magnetic field. Its lowest division, B1, contains most of the animal fossils and all hominin remains recovered from the principal excavation blocks, deposited around 1.77 million years ago.
Nearly 80 hominin fossils representing at least five individuals have been reported. They include five crania, mandibles, isolated teeth and bones from the rest of the skeleton. The fossils combine small cranial capacities and anatomically primitive traits with features associated with early Homo erectus. Their classification remains disputed. Vekua and colleagues assigned a cranium and mandible described in 2002 to early Homo, noting affinities with both H. habilis and H. erectus. A later study of the complete fifth skull argued that the variation among the Dmanisi individuals could fall within a single evolving lineage of early Homo, which the authors placed within H. erectus. A 2025 analysis of tooth-crown dimensions instead supported the presence of two taxa, previously named H. georgicus and H. caucasi, while acknowledging that a single-species interpretation could not be excluded if differences between the sexes were comparable to those of living great apes.
The Dmanisi stone industry belongs to Mode 1 and is technologically comparable to the African Oldowan. An analysis of material excavated through 2006 included more than 8,000 objects. The assemblage consists mainly of flakes, cores and core-choppers; deliberately retouched pieces are rare, and bifacial handaxes are absent. Basalt, fine-grained volcanic tuff and andesite were the principal raw materials. Cobbles and angular stone blocks were collected from nearby riverbeds and rock exposures, and flakes and manufacturing debris show that stone was worked at the site.
The dense concentrations of animal bones in B1 resulted from several processes rather than a single hunting or occupation event. Hyenas and large cats accumulated or modified many remains, while other animals died naturally. Stone-tool marks occur on less than 0.5 percent of the examined medium- and large-mammal assemblage but demonstrate that hominins processed several kinds of animals. Some marks indicate access to limbs while substantial flesh remained, whereas others are consistent with scavenging.
Orozmani Kvemo Orozmani lies in the Mashavera drainage west of Dmanisi. Its approximately 15-meter sedimentary sequence is enclosed between two lava flows dated to approximately 1.825 and 1.765 million years ago. Correlation of this sequence with Dmanisi provided an independent chronological and environmental framework for the latter site.
A preliminary report on excavations begun in 2020 described Oldowan-like artifacts, fossils of carnivores and hoofed mammals, and a hominin tooth at Kvemo Orozmani. The authors reported the archaeological deposits as roughly contemporary with Dmanisi and presented Orozmani as a second Early Pleistocene hominin-bearing locality in the South Caucasus.
Middle Pleistocene industries A long chronological gap separates Dmanisi and Orozmani from the later Lower Paleolithic record characterized by Acheulean stone tools. Paleontological localities at Tsalka and Akhalkalaki, dated by their animal fossils to around one million years ago, fall within the intervening period but have not supplied archaeological evidence of hominin occupation.
Acheulean cave sites
The principal stratified evidence of Acheulean technology in Georgia comes from the caves of Kudaro I, Kudaro III and Tsona, situated on the southern slopes of the central Greater Caucasus at elevations between approximately . Handaxes, cleavers and other large bifacial tools are well represented at Kudaro I and Tsona, contrasting with the Mode 1 technology documented at Dmanisi; the evidence from Kudaro III is more limited. Most other Georgian localities attributed to the Acheulean are open-air sites with limited stratigraphic or chronological control.
Kudaro III layer 8a yielded a sparse assemblage attributed to the Acheulean and a single thermoluminescence date of thousand years ago. Tushabramishvili regarded this age as problematic because it rests on one date and limited archaeological material. Acheulean occupation is more securely documented in Kudaro I layer 5, whose sublevels produced dates ranging from approximately 360,000 to 245,000 years ago. Tsona has no direct numerical date and is tentatively placed around 350,000 years ago through comparison with Kudaro I.
Kudaro I preserves a comparatively extensive record of stone working. Layer 5 contains flakes, cores, pebble tools, handaxes, a cleaver and locally distinctive large tools known as tsaldi. Schistose rocks, sandstone and other materials available near the cave dominate the assemblage, although some flint was obtained from farther away. Abundant cores and flakes indicate that much of the flake production occurred at the site. Large bifacial tools were made from pebbles, flakes and slabs, primarily using local stone; they include triangular and heart-shaped forms with varying degrees of shaping.
The Tsona assemblage consists mainly of bifacial implements, cleavers, other heavy-duty tools and selected flake tools. Very few flakes were recovered, and the authors found no clear evidence of core reduction inside the cave, although four indeterminate core fragments were reported. Mgeladze and Moncel therefore concluded that most tools were produced elsewhere and carried into the cave. Broken implements were not generally reshaped, and missing fragments indicate that some tools may subsequently have been removed. These features are consistent with brief, specialized visits, whereas the more complete production sequence at Kudaro I indicates longer or more varied activity.
Mgeladze and Moncel interpreted raw-material procurement as evidence of movement between lower plateaus and high mountain valleys. Most stone was collected near the caves; some flint—including red and pink varieties absent around Kudaro I—may have come from the Kvirila valley approximately to the south. The researchers interpreted the cave occupations as evidence that Acheulean groups extended their territories into high mountain valleys during temperate intervals. They considered the Levant a possible source of the bifacial tradition, followed by the development of regional forms, while emphasizing the scarcity of securely dated comparative sites.
Ziari and non-biface industries