Two thoughts, since publishing this:
– Site This is an archaeological problem, and much more clearly open to further research and synthesis than some of the others. Indeed it is an area in which many are working. Richard Morris’s Churches in the Landscape has already done much synthesis; and every year more analyses of specific sites are published in which landscape analysis reveals very much indeed.
– Inexplicable archaeologies This is an extra one. Parish churches often contain archaeological discontinuities that are hard to understand. In mine, for example, one column of the north arcade has been lengthened by, say, 8-10 inches by simply inserting a stretch of wall in mid-capital/column. This presumably is a response to a setting-out error, and is also a neat proof that the (simple) north arcade is a decade or two later than the (more complex) south one. Perhaps, too it means that the masons started at the east and west ends simultaneously and only realised their error when the two side met. So this one, at least, is not inexplicable. But they often are. For example a whole group of local churches have odd extensions at the west end, towards the tower, which must be to do with construction of the tower but which don’t happen elsewhere and don’t really make sense. 9 out of 10 of such problem areas are presumably, like mine, the result of bodges; but the problem comes when a whole group of misalignments/cut back walls/etc defy explanation, and that some of these may be clues to lost structures rather than incompetent/rushed workmanship. Here, again, there may be room for further investigation, for example in the patterns of use and abandonment of doorways, which when drawn together may reveal patterns hitherto unnoticed.