thanx for the responses (& would be glad to hear more)...
just a bit of explanation: i've been going thru everyone from steno, william smith, wheeler, kenyon, harris, pitt-rivers, even darwin's book on worms, etc., and looking at what they've written about the basics of stratigraphy; a lot of the earlier stuff is unclear about how disturbance might be recorded in part because archaeologists didn't excavate, that was all done by various "workers," "labourers"; i think i even have P-R calling them "rogues" and some kind of "ne'er-do-wells" cuz if they were any good they'd have decent jobs and not willing to work for him...
and then there's petrie...
but mixed in with all of this there are several examples where (usually) a coin might be found in the wrong place, which seems to reflect early antiquaries' ties with numismatics...
then there is a lot of comparable material from introductory textbooks, which basically all state some variant on steno, smith, uniformitarianism, evolution, 3 age system for the genealogy of the discipline; that stratigraphy is based on the idea that a layer is "sealed" (in the most extreme examples) by the layer above...
so i'm wondering how different real-world practice is after the introductory textbooks, and outside academia (where i assume their may be more time & money for reflexivity, C&N transformations, post-processualism, etc.), especially since the textbooks don't really seem to explain what to do if you suspect there might be disturbance; or if layer boundaries are unclear, or... any of those real-world things that are supposed to come with "experience"...
but most of the documentation systems i've seen (forms, context sheets, profile/plan drawings, tomographic photography, etc.), except maybe durham's "reflexive" context sheet, hasn't had anything systematically asking for evidence for possible disturbance; and if there was, i'm still not sure how to record it: obvious rodent holes, etc., but even then hodder wrote something about getting rid of such "gobbledygook" to "produce a drawing acceptable to the discipline" & there's wheeler about being "decisive and interpretive" & petrie saying nothing you weren't sure about should be recorded...
& then nobody's usually going to include anything about disturbance in most journal articles, so... it sort of disappears from the record & is hard to evaluate how serious a problem it is & how what is done about it...
my concern is that even single context planning really only records a surface, not what happens inside; so even if you suspect maybe something fell down a frost crack, or was dragged down by worms, or roots, or whatever, could you really document evidence for that (paths of worm holes, or root casts, for example), or only document your suspicions...?
there was also an attempt by triggs a few years back to do "matrix seriation" on a well-documented historical site to try to identify intrusives, but i'm not sure he was so successful...