Watching the Sun in Architecture

The Posts That Watched the Sun: What Hezingen and Yeavering Add to the Eostre Case

A new short article by Richard Sermon in Time and Mind, published online in June 2026, does something Victorian-era mythographers and modern skeptics have both failed to do for the goddess Eostre: it puts a physical, dateable, measurable structure on the ground next to her name. Two West Germanic sites, roughly two hundred miles and probably several decades apart, both show rows of upright wooden posts aligned to track the equinoctial sunrise. One of them sits under a mound of coins and jewelry that archaeologists have called offerings. The other sits under Bede's own account of the calendar he says the Anglo-Saxons kept before Christian conversion. This is exactly the kind of convergence that the Reclamation Series has been built to notice.

This post gives us a second, independent evidence stream for Eostre that isn't etymological at all. Hezingen's post row and Yeavering's Structure E either point at the equinoctial sunrise or they don't, and they do, which is a very different kind of proof than a cognate word list. It also hands you a clean, documented case of the appropriation-not-erasure pattern (Paulinus preaching at Yeavering's own sunrise posts in 627 AD) that slots right next to our Doctrine of Discovery and Christmas material.

Documented: what the article actually shows

Two sites, same evidence class, worth separating cleanly before any interpretation gets layered on top.

Hezingen, in the Twente region of the eastern Netherlands, near the German border. Metal detectorists found a concentration of early medieval coins and jewelry there in 2019. Excavation between 2020 and 2021 uncovered a thirty-meter row of eighteen post holes running due west to east, with a large boulder at the eastern end. The coins and jewelry, mostly dated between 590 and 700 AD, were deposited at this row over many years, not buried once as a hoard or grave goods. The excavators' own conclusion, quoted directly in Sermon's article, calls it a pre-Christian cult site where the finds "should be considered offerings." They go further and propose the row functioned as a solar observatory: on the equinoxes, and only on the equinoxes, a vertical post's sunrise shadow and sunset shadow form a single straight line. The row was built to catch that alignment.

Yeavering, in Northumberland in the north of England, close to the Scottish border, was the seventh-century royal seat of the pagan and later Christian kings of Bernicia. Brian Hope-Taylor excavated it between 1953 and 1962 and identified an "east-west axis of structural symmetry" respected by at least seven major buildings, including two great halls. The axis ran through a genuinely strange wedge-shaped structure, Structure E: nine concentric arcs of timber foundations radiating out from a single post, widely read as tiered seating for an assembly or grandstand. Sermon's contribution is to correct the azimuths (earlier researcher Richard North had used a flawed geometric plan and an artificially raised horizon) and show that three sunrise alignments at Structure E, once measured correctly, line up with the summer solstice and with day-counts that could regulate a lunisolar calendar. That calendar is the one Bede describes in De Temporum Ratione, written around 725 AD, where he names the lunar month Eosturmonath after a goddess, Eostre, whose month falls in what the Roman calendar would call April.

The two sites are contemporary, seventh century AD, and belong to closely related, mutually intelligible West Germanic-speaking cultures: Angles, Saxons, and Frisians. Both show immigrant or resettling populations building their new ritual architecture directly onto much older prehistoric burial mounds already present on the land, apparently to borrow ancestral legitimacy for a claim that wasn't actually ancestral. The article is careful, and this matters, to flag where its own evidence is soft: Hope-Taylor never had radiocarbon dating, the internal phasing of Structure E is inferred rather than proven, and North's original alignment calculations were simply wrong until Sermon went back to the primary excavation plans.

Pattern: what this does to the Eostre chain

Your Easter research already traces the sound and concept chain: Aset, Isis, Ishtar, Astarte, Eostre, Easter. Until now the strongest evidence for that chain has been philological and comparative, Bede's text, cognate goddess names, seasonal placement. What Sermon adds is a second, independent evidence stream that doesn't rely on etymology at all: architecture built to do a specific, testable job. A row of posts either lines up with the equinoctial sunrise within a fraction of a degree, or it doesn't. Hezingen's does. Structure E's corrected azimuths do. That's not a shared root word or a scholar's interpretive reach. That's a compass bearing, cut into the ground fourteen centuries ago and still checkable today.

This is also a clean, documented instance of the pattern you've traced elsewhere in Playing With Fire: solar and lunar tracking as the literal infrastructure of a goddess cult, not a later poetic gloss on one. The article's own math is worth sitting with. A twelve-month lunar year runs about eleven days short of the solar year, so an intercalary month has to be inserted every two to three years to keep the two calendars from drifting apart, the same problem every lunisolar calendar on earth has had to solve, the Hebrew calendar included. Sermon proposes that Structure E's tiered arcs could have let observers watch the first full moon on or after the spring equinox, check its rising position against the alignment's known azimuths, and use that single reading to decide whether the coming year needed an extra month. If that holds, Eosturmonath wasn't just a name on a list Bede compiled from hearsay. It was the output of an actual working calendar, kept by people standing at a specific post, watching a specific horizon, in the exact window your sound chain already places her: late March into April, the season of the reawakening light your solar-symbolism thread returns to again and again.

There's a second pattern worth naming, closer to your suppression-and-appropriation research. Michael Bintley's earlier work, cited in this article, argues that Christian missionaries in the seventh century consciously appropriated the "cult posts" at sites like Yeavering rather than simply destroying them. Bede records Bishop Paulinus arriving at Yeavering in 627 AD and preaching there for thirty-six days, baptizing the local population in the river Glen, at the very site where the sunrise posts already stood. That's a documented case of the same absorption-not-erasure mechanism you've mapped through the Doctrine of Discovery material and the Christmas chapter: don't tear down the sacred infrastructure, stand your own message inside it and let the old alignment do quiet work under a new name.

Green-hat speculation: where the evidence stops and the shape starts

Everything above is Sermon working from primary excavation reports and Bede's own text. Past that line, a few things are worth naming honestly as pattern-recognition rather than proof.

The article itself stays appropriately cautious about a direct causal or migratory link between Hezingen and Yeavering. The azimuths differ by roughly four degrees, and Sermon is explicit that this could reflect two related cultures independently solving the same observational problem rather than one site copying the other. That's the more interesting possibility for your purposes, honestly. Convergent solar-tracking architecture appearing twice, in two populations who spoke mutually intelligible dialects but lived in genuinely different landscapes (lowland and already-depopulated Twente versus upland and still-populated, still-British Northumberland), reads less like diffusion and more like the same underlying human relationship to the equinoctial sun expressing itself wherever a small population needed to keep a working calendar. That's the pattern your book keeps finding: not one central source radiating outward, but the same structural relationship to the sky recurring independently because it solves the same problem everywhere it's tried.

It would be speculation, not established fact, to claim Hezingen's post row was specifically dedicated to Eostre. The excavators themselves only raise the possibility, citing your colleague Rudolf Simek's competing reading of Eostre as a fertility goddess rather than a sunrise goddess, and note that a ninth-century Old Saxon baptismal vow names Thunor, Woden, and Saxnot, not Eostre, as the gods being renounced in that region. Sermon's case for Eostre at Hezingen rests on the timing (an equinoctial alignment) and the etymology (the "east" root in her name) lining up together, which is suggestive, not documented.

Where this sits in the larger project

What makes this article valuable for the book isn't a single new fact so much as a new kind of fact: physical, azimuth-measurable, independent of the philological chain you'd otherwise be relying on alone. It's the difference between arguing a word probably meant something and pointing at a fourteen-hundred-year-old post row still oriented, within a fraction of a degree, exactly where the argument says it should be. That's the kind of evidence an engineer's pattern recognition was built to trust, and it's exactly the register Playing With Fire keeps reaching for: not belief, not received tradition, but the sky and the ground still agreeing with each other long after the people who built the alignment are gone.