mega-what / whole horizon analysis

Prehistoric Ritual Horizons: Overall Fit

Tap/Click pics for big ones, names for site pages. Prehistoric luni-solar trajectories.
Ancient cultures seem to have regarded a place as a Sacred Place when it was a Place where Heaven and Earth could be seen to meet in Harmony at the Horizon. Thus they had an effective luni-solar horizon calendar. Having seen that, we can understand that prehistoric monu­ments were not just about alignments but marked optimal observ­ing positions.

These examples demonstrate how they required an overall fit between landscape and sky at every sacred place.

Panoramic view from Farnoge Court Tomb Farnoge Court Tomb c.3500BCE:

This first site shows a clear north-south axis with a dip to one side and a single hill to the other. South is in a saddle. Westwards, a single hill has the equinox on its top. The south lunistice zone is on the flat slope leading up to it. The north lunistice zone spans a slight rise on the other slope. Eastwards, the south lunistice zone spans a dip, ending on a hilltop. The equinox is at the lowest point of the Eastern horizon and the north lunistice zone ends on another hilltop. North is at the intersect of local and far horizons.

Panoramic view from Ballynamona Court Tomb Ballynamona Court Tomb c.3500BCE:

The second example is a version of single dip to one side of the north-south axis with two hilltops on the other side. It also shows that sometimes the measuring tool is the landscape profile below the horizon rather than the horizon itself. South is indicated by the land / sea intersect and the western horizon forms two peaks. The northern one has major standstill on its top, the southern one is lunar mid­point and the dip between them is a week or so before spring equinox. Eastwards is a bay and though the sea horizon is flat, the shape of the land forms a dip, with the equinox in the middle. Across the bay, distant hills mark summer sun rises and the lunistice zone. North is at the intersect.

Panoramic view from Maulinward Stone Row Maulinward Stone Row c.1200BCE:

This third one is a rather extreme version of two hilltops to both sides, or single dip to both sides if you prefer. South is in a saddle. Westwards, the south lunistice zone spans a hilltop and runs down the ridge. The north lunistice zone runs up a slope and ends at a hilltop. Between these is a dip that bottoms at the Equinox / Winter Cross-Quarter mid­point. Eastwards, the south lunistice zone spans a hilltop and ends at the basal step. The north lunistice zone spans a distinct block of high ground. Between them is a dip that bottoms at the Equinox / Summer Cross-Quarter mid­point. North is roughly indicated by a dip.

Panoramic view from Dunbeacon Stone Circle Dunbeacon Stone Circle c.1200BCE:

The fourth example is an extreme version of two hilltops to one side, single rise to the other. It shows how local ground as close as c.20m may be used to form the appropriate overall fit. North is on a distant hill but the right hand slope of it, as is south. They are rough indicators only, being subservient to the fit of both sides. The eastern hill has summer cross-quarter at its northern foot and winter cross-quarter at its southern one. Notice how the equinox has no mark itself but accurate ones a quarter-month before and after it. The west is very local but fits nonetheless. Accuracy for that sector is provided by an outlier.

Home Site Lists Glossary

© Michael Wilson.