>I was always told the churches are oriented east-west so the south wall should really be facing south...... does anyone know what's going on here?<
It's true that about 30% of British mediaeval church dials decline and do not face exactly South - and hence the church itself doesn't face the cardinal points of the compass. So many of the Christian churches in Britain are (more or less) aligned East-West (and burials are also so aligned) that it is quite often presumed that this was a necessary design criterion in the construction of churches and so, in a few cases, it almost certainly was; but as persons of today's technical age, I feel we can all too easily wrongly assume that the builders of those times had an interest in and desire for, accuracy similar to that which we have today. For my part I am not at all clear that there was a real desire for accuracy in church alignment - whether or not it was feasible with the technology of the day. Just as with many other constructions of the same period we frequently see an ambivalent attitude to questions of this sort. Even 100-200 years later than the time of the mediaeval churches, the timber framed buildings of the Tudor period were erected using green timber which, very soon after completion, developed the distortions that we see today. In the sixteenth century they certainly understood such problems and they had the means by which to avoid it (as they did with boat building for example) but even with houses for quite wealthy patrons it was not deemed to be important enough to store timber for long periods before use. These building imperfections were just accepted. There would have been many building workers all involved on different parts of a church at the same time and even in prestigious constructions like cathedrals problems were usually solved as they were encountered rather than by any form of real planning. The case of the Duomo in Florence is a case in point. There the building was started many years before the problem of how to support the dome was contemplated, let alone solved. It was only during the later building that they installed the meridian line probably more to measure the settlement of the walls than to get things aligned (and it amply demonstrates the misalignment of the transcepts as one can see today) . The same relaxed approach applied to the lanterns in Durham and Ely cathedrals. You also see it in early stained glass where some panes are adjusted to fit at the end of the assembly of the window rather than to have all made to the same uniform size from the outset. Furthermore when you also consider that in those times there was much less attention paid to the construction of foundations than there is today it is perhaps not surprising that we see strange shifts in alignment as a building develops and extends over varying ground conditions. Indeed it may well be that those buildings with the larger shifts are more a consequence of avoiding unsatisfactory ground conditions or even of making use of hitherto unknown rock layers than ever it was of a desire at all times to maintain a rigorous East/West alignment. It has been said that sometimes there may have been a deliberate intent in the design to acknowledge the drooping head of Christ. However I suspect that it is more likely that such ideas are really local 'explanations' - or excuses - devised to justify the way the construction developed, especially since God was then thought to be intimately involved in all things. As I said, some 30% of vertical dials on UK church walls decline (that is to say 'face') slightly East or West of South and it has been suggested that this might be used as a means of establishing, albeit approximately, the extent of non-alignment of churches with the East-West line. If it really was an acknowledgement of the drooping head of Christ then perhaps one would expect many (or maybe even most) such declinations to be in one direction. In fact for dials that do decline and where we have records of measured values of declination we see very roughly that 60% of dials decline slightly East and 40% decline slightly West. I suspect this is close enough to 50:50 for us to reject the 'head of Christ' hypothesis. However I have not researched the issue. Turning to the matter of the techniques that were available to builders of the early period, it was relatively easy to establish the North-South line from the shadow of the Noon sun. However, I don't know if you have ever actually done this but to depend on a direction found by waiting for the shadow (of say a vertical four foot cane) to be at its shortest and marking that out can easily yield an error of plus or minus 2-3 degrees, especially if this is done in summer when the sun is high in the sky. If the cane is not quite vertical then it is easy (in a single measurement) to see errors of 4-5 degrees. If a suspended magnet had been used to establish North-South in some cases then there is a changing error between magnetic North and True North which in Britain can reach large values but which is of the order of 5-10 degrees and occasionally more. Figures in the range 2-6 degrees are typical of many of the misalignments in churches of the mediaeval era and this may be the explanation for some of them. An interesting question though. Patrick
