Saint Michael Sword: Are the cathedrals really on a straight line?
Legends claim that seven prominent European sites dedicated to St. Michael form a perfectly straight “sword” across the map, implying a deliberate medieval alignment. Commenters unpack why this is almost certainly a product of map projection quirks, selection bias, and the sheer number of St. Michael sites, noting that the locations are not cathedrals, predate the Mercator projection, and sit on pre-existing strategic or sacred landscapes. Several point to statistical concepts like the birthday paradox, ley lines, and the look‑elsewhere effect to argue that such striking patterns are inevitable when you have thousands of religious sites to choose from.
Nature of the “St Michael” Sites
- Commenters note the seven locations are mostly monasteries, sanctuaries, or islands, not cathedrals in the technical sense (seat of a bishop).
- Several are on pre‑Christian or obviously constrained sites (rocky tidal islands, pagan temples, defensive hills), so their exact locations were not freely chosen.
- St Michael is an extremely common dedication: hundreds of churches in England alone, many on elevated or liminal sites and often built over earlier pagan sanctuaries.
Geometry, Projections, and “Straightness”
- Participants stress that “straight line” is ambiguous on a globe:
- Geodesic (shortest path on a sphere/ellipsoid).
- Rhumb line / constant bearing (straight on Mercator).
- Lines in other projections (Plate Carrée, equirectangular, azimuthal).
- The alignment looks best on a Mercator map but is not a true geodesic; distances from a great‑circle line are non‑trivial.
- Some point out Mercator’s special property: straight lines represent constant compass bearings, useful for navigation but not “natural” geography.
Coincidence, Probability, and Selection Bias
- Many highlight the Texas sharpshooter fallacy and look‑elsewhere effect: with thousands of churches and many map projections, remarkable alignments are inevitable.
- Rough probabilistic arguments:
- Any 50 km‑wide band across Europe covers ~1% of the area; with ~1000 St Michael sites, dozens of such bands will contain many sites.
- Choosing 7 from thousands dramatically increases the chance of finding an apparent line.
- Others criticize simplistic probability estimates (e.g., GPT‑generated) as mis‑framing the question and ignoring that the line and subset were chosen post‑hoc.
Historical Feasibility and Intentionality
- Strong skepticism that medieval builders intentionally aligned these sites using a projection they did not have (Mercator post‑dates most sites by centuries).
- To argue intentional design, one would need:
- Evidence of an old projection where the line is straight.
- Historical records of such an alignment or of a “St Michael’s Sword” concept.
- Some ask when the legend of the line first appears; cited references suggest a very recent (20th‑century) origin, not medieval.
Broader Reflections and Further Work
- Several compare this to ley lines and other pattern‑seeking (apophenia) in landscapes and texts.
- Suggestions for more rigorous work:
- Build a full database of St Michael sites and search for the line that covers the most points.
- Monte Carlo simulations of random points to quantify expected alignments.
- Despite skepticism, many find the geography, history, and symbolism around the sites genuinely interesting.