Take the Michael ~ mer-oiiiii? :)

As to the argument of the paper, if I understand the outline, his argument is "that the integrated factors of phonetic motivation, psycholinguistic processes and socio-cultural motivation are responsible for the Proteus-like resilience and power of the 'pejorative' sl- schema and the historical localisation of the pejorative du- schema. That is to say, rather than any one of these factors being responsible for schema development, this is a combinatory process."

Or in plainer English, it's a combination of factors based on submorphemic characterisations ~ on which we subsequently accrete a complex association of cultural values. So if two or three words group around a common soundscape and have a similar tonality of meaning or social denotation, it's very likely that other words based on a similar 'scale' will be modulated in their structure to end up according with the same sounds. For example, if you notice (consciously or unconciously register) that words like crack and smack are part of a common family, it's not surprising to find yourself coining a neologism in a comic such as "thwack". [Yeah, I know Michael, it's at this point you tell me the OED cites thwack from about 932AD but :) ]

At least I think this simplification is at the root of the processes the author discusses, which are obviously more subtly examined in all their complexity.