Is the Bulford Post Alignment to the Midwinter Sunset?

Look along the posts: https://timdaw37.github.io/bulford-posts-3d/

Set Midwinter, Sunset, Full orb. Then Midsummer, Sunrise, First gleam. The question is which of those two pictures you trust.

Midwinter Sunset

Midsummer Sunrise

Stonehenge has the same argument. The axis works both ways, and for a generation the tourist photograph has been midsummer sunrise up the Avenue. Parker Pearson and others have put the weight on the other end: you walk in from the north-east and watch midwinter sunset drop through the Great Trilithon. Ruggles has been careful to say the architecture is precise enough to pick the solstice in space, and that first or last gleam is the closer convention, not that one festival owns the monument.

Harding, Leivers and Silva, in PAST 113, publish the Bulford posts as a solstitial pair — midsummer sunrise and midwinter sunset, about 120 m apart, around 2950 BC. Harding found the two structural pits first and only then asked what they pointed at. That is the right order. The question here is only which way the line is better.

The 3D model is the published plan, not the Wessex GNSS. From that plan the posts run 48.31° / 228.31° true. On Bulford’s own skyline, 2950 BC, the six like-for-like events are:

Event	Convention	Azimuth	Miss from the posts
Midwinter sunset	Full orb	228.55°	+0.23°
Midsummer sunrise	First gleam	48.90°	+0.58°
Midwinter sunset	Half orb	229.03°	+0.72°
Midsummer sunrise	Half orb	49.29°	+0.97°
Midwinter sunset	Last gleam	229.53°	+1.22°
Midsummer sunrise	Full orb	49.72°	+1.41°

Full orb at midwinter sunset is the closest of the six — a quarter of a degree, inside a solar radius. First gleam at midsummer is next, half a degree, one disc. The rest are worse.

Two warnings, which do not cancel the table.

You must not mix conventions. First gleam at one end and full orb at the other are different ways of watching. If the same limb is used both ways, winter last gleam is worse than summer first gleam. Reciprocity was never going to give a second bullseye on one straight line: refraction and the solar radius push rise and set apart, and the south-west horizon is not high enough to cancel that.

The picture also flatters the winter disc. The south-west skyline is a local rise at about 880 m, so the sun sitting on it looks planted. The north-east gleam is 5.7 km out. Your eye and the azimuth are not quite the same measurement.

Ruggles’ published rule of thumb is that claimed sightlines of half a degree and better in the old lists were the ones most easily explained as selection, that British prehistoric orientations show no evidence of precision much greater than about 1°, and that a couple of degrees is still a fair orientation in a group of sites (Ruggles 1984, 304–306; 1997, 207). On that scale the midwinter full-orb miss is allowed, the midsummer first-gleam miss is allowed, and neither is a prosecution. A single pair in a crowded pit field still has to beat coincidence.

We do not have the GNSS. Two independent digitising jobs of the PAST plan agree to 0.2 m, which is inside the pit and not the solar miss. The miss may move when the survey comes out. Until then, the model says: if you have to pick one, pick the sun standing on the south-west ridge at midwinter.

Have a go with the table in one hand and the model in the other.

Harding, P., Leivers, M. and Silva, F. 2026. A newly discovered solstitial post alignment in the Stonehenge landscape at Bulford. PAST 113, 2–5.

Ruggles, C.L.N. 1984. Megalithic Astronomy. BAR British Series 123. Oxford.

Ruggles, C.L.N. 1997. Astronomy and Stonehenge. Proceedings of the British Academy 92, 203–229.

3D plan: https://timdaw37.github.io/bulford-posts-3d/