Residuum? You had it right in the first place - residual soils. The trouble with spec writers is that they do "what has always been done before" - and he might have worked, say on a highway project where, sadly, I see a desire to compact the foundation to a percentage of some standard - which might actually be lower than in situ for the sake of "what"?? In embankments, the "depth of compaction" of any natural subgrade soil will be small and not really influence settlements or strengths to any important degree. If, though, the soil is within, say, 1 m of the final pavement grade - I can see doing it - but as Ron indicated, it is more for the idea of finding soft zones or pockets of topsoil/peat that might be present - not in actually 'compacting'. For foundations, as you indicated in your few few posts, you develop a bearing capacity based on the materials' in situ characteristics. Would one compact a firm clay when putting on a footing (as GeoPaveTraffic indicated?) - No. Would I compact to a standard a sandy/gravelly soil? - No unless the footing is the minimum width of 600mm or so - and then only compact to build out any disturbance - but not to a "compaction level". Obviously the spec writer likes numbers . . . In the end - if you are satisfied that the soil is good for what you say, tell him that you prefer not to compact - as it might have an adverse affect - and you will confirm conformance with your design parameters by other means if they want - - dynamic cone test, etc.