In the half-century that has elapsed since C. H. Haskins published his Norman Institutions a number of documents and studies relating to the subject of early Norinan military service have appeared. Few scholars who have looked closely at the subject would now accept the clear and persuasive hypothesis Haskins then put forward; indeed he himself, with his keen critical sense and respect for sources, would probably have been one of the first to wish to modify it, had he been writing at the present day. But because his views have crept in an over-simplified form into general histories, to be repeated and exaggerated, the whole question of possible changes after the Normans came to England has tended to become distorted by wrong assumptions about their earlier customs in Normandy.

Haslcins looked at the feudal obligations of Norman abbeys and bishoprics in the 1172 returns and, starting from the fact that only the older foundations (though not all the older foundations) owed such service, argued plausibly enough that fixed quotas must have been imposed in ducal Normandy, probably (in spite of the slight anomaly of Saint-Evroult, founded in 1050) at least as early as the reign of Robert the Magnificent.’ He did not, in fact, explicitly argue that all the later accepted incidents of knight service - forty days a year at the vassal's own expense, castleguard, reliefs, aids, wardships - existed in their final form at so early a date. But others have certainly assumed that this was so. Henry Navel, for example, saw the obligations of feudal scrvicc as fixed and unchangeable over a long period; in discussing the vavassors of Le Mont-Saint-Michel he asserted that the military service owed to the abbey after it acquired certain properties c.1024 must have been imposed in its entirety by the duke at an earlier date, because ‘we know how difficult it was for a lord to impose fresh services on a vassal at this period'; yet his proof consists of a case occurring in 1157, over a hundred years later, Even Powicke set the seal of approval on the general hypothesis when he wrote: ‘Before 1066 the Norman dukes were able to regard their country as divided for the most part into a number of knights’ fees… The grouping of warriors was symmetrical and was evidently imposed from above'.