The Levallois Method was practiced by hominins in Africa, Europe, and West Asia.  ‘Levallois’ refers to the Levallois-Perret suburb of Paris, where the method was first described.  The Levallois Method is associated with ‘archaic’ species of Homo, including Homo neanderthalensis, and a new hominin that appeared on the scene—Homo sapiens.  The Levallois Method persisted to about 40,000 BP, followed by a greater emphasis on ‘blade’ technologies—flakes that are more than twice as long as they are wide.  The formal Levallois Method is absent from Asia and Australia.

The Levallois Method involves striking a large, invasive flake from a bifacial core.  To successfully remove a flake, the flintknapper must strike behind a zone of high mass.  The crack undercuts the mass, which falls away as the flake.  In simpler approaches to flake-making, zones of high mass are removed opportunistically as they are identified on the core.  In contrast, the Levallois Method involves carefully-considered strategic flaking to create a zone of high mass of a specific shape.  The carefully-shaped high mass is then removed as the objective flake, which is used as the tool.  This involves a two-step process:  1) remove the mass-shaping flakes, and 2) remove the objective flake.  This is cognitively more complex than removing flakes opportunistically, because removing each of the shaping flakes also involves creating and removing zones of high mass.  The structure of the process is ‘hierarchical’ because a successful objective flake depends on successful core shaping.  

A key feature of the Levallois Method is that only one face of the core was invasively flaked—first to set up the high mass for the objective flake, and then the removal of that high mass.  In contrast, the opposite face was usually flaked non-invasively, just enough to provide platform surfaces to strike the flakes from the ‘objective’ face (the face from which the objective flake would be removed).  The stone on either face can be conceptualised as a ‘volume’, and, in this case, one volume is used differently from the other—the volumes are also hierarchically related.  

In one variety of the Levallois Method, the objective face was shaped in such a way that the surface was domed, and this dome created the high mass subsequently undercut by the objective flake.  Archaeologists refer to this shaping process as manipulating the ‘convexities’ on the core, by the process of removing stone as small flakes.  The configuration of the flake scars from this process define the zone of high mass.  In another Levallois Method variant, the high mass is made triangular in shape, and the objective flake removing this mass is thus triangular in shape.  

Striking that final objective flake required the platform to be just right in terms of platform angle and depth.  To achieve this, much platform adjustment was sometimes necessary, resulting in multifacetted platforms.  A core might be reworked several times, producing multiple objective flakes.  The Levallois Method was accomplished by direct percussion using hard hammerstones.

The Levallois objective flakes were retouched and used as cutting and scraping tools, but some may have been used as spearpoints (although this is disputed by some archaeologists).  The latter examples include Levallois flakes ‘predetermined’ to be triangular in shape as struck from the core, but others were retouched into this triangular shape.  Levallois flake variants from North Africa were retouched at the proximal end to form a projecting stem, creating artefacts called ‘Aterian Points’.  The stem was presumably inserted into a spearshaft or handle.

The way that the Levallois Method is defined, applied, and interpreted is not without controversy.  The Levallois Method was recognised by French researchers from the middle 19th Century, but was more formally defined by the European lithic expert François Bordes in 1950.  But by the 1980s European archaeologists seemed to apply it uncritically to all sorts of bifacial assemblages—and particularly flakes—from all over the world.  Partly as a consequence of this, the French archaeologist Éric Böeda redefined Levallois in geometric and conceptual terms (as described above), thus restricting identification of the Levallois Method to reduction sequences that demonstrate these very specific criteria.  This definition is in general use today, although many analysts are far less strict that Böeda, and many technologies described as ‘Levallois’ do not fit Böeda’s formal definition.