"Nichols caused a stir in the 1990s by holding out the exciting possibility of a grand theory of language spread and distribution based on a few pieces of linguistic typology. On examining her evidence, however, this began to look increasingly like casuistry. In brief, Nichols views Eurasia as a giant geolinguistic pinball machine, in which any language which happens to wander into what she terms the ‘locus’ hits the jackpot and automatically spreads over a huge area. According to her, this has happened several times but the new most favoured language tends to obliterate all traces of the previous most favoured language, unless the latter has managed to expand into a ‘refuge’. As such, according to her, Kartvelian was a previous ‘most favoured language’ which spread into the Caucasus from Central Asia (ibidem), with the complete absence of Kartvelian speakers in Central Asia showing how successfully her spread zone theory works. Conversely, the Pontic Steppes is too far from the locus to be fit for purpose, so that she actually offers no support for Anthony’s model, but as seen, he is not concerned by such trivial details. Nichols’ hypothesis per se is an intriguing one, but it has to be backed up with hard evidence and hence a “Russell’s teapot” theory which assumes as a central postulate that the modus operandi of a model wipes out all of the empirical evidence for itself is deeply suspect."