In Africa composite-tools that involve the hafting of stone artifacts in handles with gums, cords, sinews, and/or other binding materials in first appeared at the transition from the Acheulean to the MSA. In 2012, J. Wilkins and colleagues from the University of Toronto presented multiple lines of evidence implying that stone points from the site of Kathu Pan 1 in South Africa were hafted to form spears around 500 k.a. ago. The points’ damaged edges and marks at their base are consistent with the idea that these points were hafted spear tips. Allthough the arguments, that these points were hafted projectiles, seem to reasonable, I am not so convinved by this very early age. Anyhow, new data from Wonderwork, LHA/GnJh-03 in the Kapthurin Formation (Kenya) and Kanth Pan 1 indicate an ESA/MSA boundary at 500 k.a rather than 300 k.a. in South/West-Africa.
Traces of bitumen adhesive on tools from Hummal and Umm El Tlel in Syria provide direct evidence that middle Paleolithic flake- tools were hafted. Additional evidence for hafting using a variety of adhesives during the South African MSA comes from Sibudu and Blombos Caves.
Recently in the Campitello Quarry (Italy), two stone flakes partly covered in birch-bark-tar and a third without tar were discovered in association with a late middle Pleistocene Fauna. Traces of hafting have been repeatedly observed on convergent tools during the European Middle Palaeolithic (Biache-Saint-Vaast, OIS 7; La Cotte-Saint-Brelade-layer 5, OIS 7; Staroselje, OIS 4; 9, Buran-Kaya III-Level B1,OIS3, Königsaue OIS 5d or 3). In Königsaue (Germany) a birch-bark pitch displays imprints of a bifacial tool and a wooden haft.
There is general agreement that composite-tool manufacture in the Middle Palaeolithic and the MSA marks a substantial increase in technological complexity compared with the single-component tools of the Lower Palaeolithic. Producing such tools requires planning depth and complex task coordination, abstract thought, and the use of recursion and concepts of past and future (Haidle 2010).
It is suggested that most of the backed tools in Prehistory were hafted. Backing can be traced back in Africa until the Howiesons Poort MSA and other MSA-industries of the continent, described in my post about “crescents”. The Later Stone Age (LSA) at Enkapune Ya Muto rock shelter (Rift Valley of Kenya) began at 50 k.a BP with very large backed crescents and backed microliths. Backed tools are ubiquitary present in the sub-Saharan LSA, and during the Epipaleolithic technocomplexes in North Africa.
In the Levant backing occasionally occurred during the EUP (Ksar Akil) and was constitutive for the Epipaleolithic. In Europe backing was continuously present after the Gravettian over a large territory.
Diffusion of innovations between Palaeolithic people was mainly based on social networks and personal knowledge. During the Middle Palaeolithic, social groups were much more isolated than during later phases of the Stone Age with relatively low informational exchange. Therefore the rate of adoption, defined by Rogers as the relative speed with which an innovation is adopted by the members of a social system, was also relatively low. Many potential innovations during the early and middle Palaeolithic would possibly never reach a “critical mass” of adopters.
Richter (1995) described the “social memory” of a group as “the ability of a group of humans to maintain a specific set of information by means of tradition over many generations”. He described social memory as a function of group-size and intra- and inter-group interactions. In small, dispersed social groups with low interactions with other groups, an innovation will be lost over time. This may explain that the discontinuous record of hafting during the Lower and early Middle Palaeolithic / MSA before OIS3. Hafting and backing, had presumably to be invented for several times, before they were adopted by a sufficient large pool of individuals and became part of the “social memory” during the upper Palaeolithic.