|Clockwise from lower left: tortoise core, ventral flake, Levallois point, proximal points.|
The Levallois technique, as it's modelled by the majority of lithic analysts [of which, as you know, I'm not one], is a way to reduce a large block of usable flint to obtain one or at most two unifacial flakes that have the desired outline (viewed dorsally or ventrally). The illustration below is my 'paste-up' of François Bordes' nine 'types' of Levallois cores and flakes, which he envisaged in a 1980 article.
Bordes, François. 1980. Le débitage Levallois et ses variantes. Bulletin de la Société préhistorique française 77:45-49.
Do me a favour and scrutinize the objects displayed below, which are some of the Levallois cores recovered at Douara Cave, in Syria. The presumed desired outcome--the final, 'Levallois,' flake removals--are outlined in red. See how many you can assign to one of the nine Bordesian types shown above. Difficult? That's an understatement! 'Impossible' gets my vote.
|Levallois cores from Douara Cave (Credit Akazawa in Suzuki and Takai 1974).|
Just to show you that I'm not cherry-picking my evidence, have a look at the following illustration from a paper by Vallin et al. published in Paleo dealing with Levallois flake variability. This is an assemblage-level comparison of flake outlines that have been identified as Levallois flakes. You don't have to be a statistician to see that there isn't one dominant outline, much less evidence of any of the nine that Bordes 'identified.' It's clear to me, at least, that you'd have to be thinking like a Neanderthal to see the variability in products of the Levallois technique as anything other than continuous.
|Outline of unbroken Levallois flakes from Hermies-Le Tio Marché ; hinged flakes in red, plunging flakes in green (CAD : L. Vallin).|
Some would call that job security. I can't see any up-side in that. I'd call my project Sisyphean in proportion, and in its likely outcome.