The problem of the Shosu must be kept distinct from that of the ‘Apiru. The latter are known from documents from all over the Fertile Crescent from 2000 to 1150 B.C.E. They lived along the Tigris, in Anatolia, northern Syria, Canaan and Egypt. They had no common language and came from no common ethnic or national group. They were never associated with a pastoral way of life. We once read about ‘Apiru mercenaries from Yabiluma at Mari in the 18th century B.C.E. Yabiluma was a tribe but also a town. The troopers could have come from either. Otherwise, the ‘Apiru people have no resemblance to the Israelites. Linguistically, ‘Apiru has no connection with biblical ‘Ibri (Hebrew), which is an ethnicon like Amorite (Genesis 14:13). All attempts to find a linguistic connection between ‘apiru and ‘ibri, including Mendenhall’s, are pure wishful thinking. The second vowel in ‘apiru is surely long (technical discussion would take us beyond the scope of BAR). R. Borger, many years ago, proposed to equate ‘apiru with Syriac ‘afir, “dusty.” It is a pejorative term for outcasts, renegades, stateless people. The ‘Apiru people were runaways from the stable Canaanite society, usually from the upper classes. They always strove to hire out as mercenaries or to support some leader who was strugging to found a new Canaanite-style feudal state.
There is no way the ‘Apiru people could be equated with either the “revolting peasants” of Gottwald’s theory or with the Hebrews of the Bible.
For further details, see my article, “Khapiru,” in the Harper’s Bible Dictionary (San Francisco: Harper & Row, 1985). Be sure to use a later printing, not the first edition (where a stupid editorial addition was made to my article; it has mercifully been removed in later printings after my protest!).