Modern Homo sapiens engage in substantial ecosystem modification, but it is difficult to detect the origins or early consequences of these behaviors. Archaeological, geochronological, geomorphological, and paleoenvironmental data from northern Malawi document a changing relationship between forager presence, ecosystem organization, and alluvial fan formation in the Late Pleistocene. Dense concentrations of Middle Stone Age artifacts and alluvial fan systems formed after ca. 92 thousand years ago, within a paleoecological context with no analog in the preceding half-million-year record. Archaeological data and principal coordinates analysis indicate that early anthropogenic fire relaxed seasonal constraints on ignitions, influencing vegetation composition and erosion. This operated in tandem with climate-driven changes in precipitation to culminate in an ecological transition to an early, pre-agricultural anthropogenic landscape.

a b s t r a c tThe ability of Homo sapiens to kill prey at a distance is arguably one of the catalysts for our current ecological dominance. Despite the importance of projectile technology in human hunting strategies, there is still no consensus on when it first emerged. Most evidence has stemmed from analysis of the lithic projectiles themselves, not the trauma left on the bones of hunted prey. There is a growing body of research focused on zooarchaeological projectile impact marks in European assemblages; however, comparable investigations are rare in the African Middle Stone Age (MSA), where it has been suggested that simple hafted projectile technology first arose. There are no standardised criteria for identifying projectile impact marks on bone and no large experimental studies exist that examine marks left by MSA points specifically. This paper defines the various forms of stone-tipped projectile impact marks on bone using a large and variable experimentally-produced sample, and then applies this system to description of marks left by replica MSA Levallois and Howieson's Poort points. The differences between projectile impact marks and slicing cut marks, marks created by different projectile modes (spear and arrow), lithic typologies (Levallois and Howieson's Poort), and distances (long versus short range) are examined. It is shown that although most projectile marks do not resemble slicing cut marks, the projectile mode, point type, and distance cannot be differentiated based on mark morphology.

The ability of Homo sapiens to kill prey at a distance is arguably one of the catalysts for our current ecological dominance. Many researchers have suggested its origins lie in the African Middle Stone Age or the European Middle Palaeolithic (∼300-30 thousand years ago), but the perishable components of armatures rarely preserve. Most research on this subject therefore emphasises analysis of armature tip size, shape, and diagnostic impacts or residues. Other lines of evidence have included human skeletal anatomy or analyses of the species composition of faunal assemblages. Projectile Impact Marks (PIMs) on archaeofaunal remains offer an ideal complement to this work, but their potential has been restricted mainly to the later Eurasian zooarchaeological record. A review of current evidence and approaches shows that systematic PIM research could add much to our understanding of early projectile technology, especially in Africa.

VR003 has been excavated over five seasons between 2009 and 2016. The initial two seasons had the primary objective of assessing the potential of the deposit, and only selected finds were plotted. The final three seasons 2014-2016 involved plotting of all cultural items using different size cut-offs for different classes of material (≥20 mm for lithics, ≥25 mm for bone). Some classes of material, notably OES, were only plotted if modified. All plotted artefacts were assigned unique provenience ID's and individually bagged. Sedimentary aggregates were plotted with a single indicative point location, and volume was quantified as proportional bucket volume (FULL, ¾, ½, ¼). All aggregates were sieved on site through nested 3 mm and 1.5 mm mesh, and the residues from each bagged separately. All plotting was undertaken using a local grid, managed by control points emplaced around the site.We currently recognise nine archaeological horizons and eight geological horizons in the Main Area sequence (1); for the sake of simplicity in this paper we refer to the lower two archaeological horizons (I-08 and I-09) in aggregate as the Lower Deposits. All horizons in the Main Area are associated with the MSA. Inside the small extant shelter (known as Sector III) we identify 23 geological horizons (III-01 through III-23), with Late Holocene LSA comprising the upper 14 strata (0.6-0.9 m depth). In the Main Area, I-04 was assigned to the Howiesons Poort and Still Bay (1), though initially dated by OSL in the Main Area to 45.7 ±2.8 ka. Inside the shelter in Sector III, Howiesons Poort artefacts occur in strata III-18 through III-23, with initial OSL ages of 42.3 ±2.7 ka and 41.7 ±2.9 (1) (Still Bay has yet to be encountered in this part of the site). In both areas these ages are unusually young for the Howiesons Poort. We provide a redating here (SI Geochronology below) using OSL on quartz grains from III-18 to III-20 that places the Howiesons Poort between 71.6±6.2 and 60.8±5.2, consistent with ages elsewhere, with ages for the immediately overlying post-Howiesons Poort stratum III-17 of 55.7 ±4.4 and 66.1 ±5.3 ka (Figure S1).The fossiles directeur for the Howiesons Poort and Still Bay (backed artefacts and bifacial points respectively) occur as a discrete band across the entire site and include a single piece of engraved OES in the Link Trench (Figure S1). In the Main Area there is no recurrence of such artefacts in the underlying deposits, which we assign to the pre-Still Bay MSA.

No abstract