How Were Medieval Shield Wooden Cores Constructed?

The Short Answer

The best current evidence suggests that some 12th- and 13th-century European shields were made from multiple solid wooden planks or boards set side by side. Alder is strongly documented in the Szczecin and Seedorf shields. Linden is documented for the Konrad von Thuringia shield at Marburg.

• the painted Szczecin shield was made from thin alder planks, but the exact plank count is disputed between six and seven;
• the second Szczecin shield preserves two alder planks, with an estimated original total of about five;
• the Seedorf shield used four lengthwise alder boards with bevelled, edge-glued joints;
• the Konrad von Thuringia shield at Marburg is catalogued as linden, while its reported board count and thickness remain source-qualified.

This is enough evidence to reject a vague idea that medieval shields were simply generic wooden boards. It is also not enough evidence to publish a single universal formula for all medieval shields.

Evidence Categories Used Here

Szczecin: Thin Alder Plank Construction

Two 12th-century kite shields from Szczecin, Poland, are among the most important surviving pieces of evidence for early medieval wooden shield construction. They are especially valuable because they are actual archaeological finds, not later artistic depictions or modern reconstructions.

The painted Szczecin shield is dated to the late 12th century. The current National Museum in Szczecin catalogue dates it to 1176-1200. The shield is made from alder and preserves evidence for a one-layer side-by-side plank structure. The individual planks are narrow and thin by modern expectations: approximately 6-17 cm wide, with thickness reported around 4-8 mm depending on source.

The Second Szczecin Shield

The second Szczecin shield is less complete, but it helps prevent another mistake: assuming that all shields from the same context had one fixed board thickness or plank count.

Only two planks survive. Dowen et al. give the surviving length as 97 cm. The two surviving planks are recorded as 13.8 cm and 12.2 cm wide, with thickness around 7.5-9 mm. The wood is again alder, identified by microscopic analysis.

Seedorf: Bevelled, Edge-Glued Alder Boards

The Seedorf shield, associated with Arnold von Brienz and dated around 1200-1220 in the Shieldwright notes, is one of the strongest comparisons for early 13th-century shield construction. It is especially important because it does not simply duplicate the Szczecin pattern.

The Seedorf core is recorded as four lengthwise alder boards. The mating edges were bevelled to increase the bonding surface, and the construction is described as edge-glued. This is a different joint geometry from the right-angled plank edges reported at Szczecin.

Marburg: Linden, But Qualified Claims

The shield of Landgrave Konrad von Thuringia at Marburg, inventory 3177, is central for early 13th-century heraldic shield evidence. It is dated around 1230/1240 and belongs to a very important Marburg group.

The current institutional record identifies linden wood, leather, chalk ground and polychromy. It gives the dimensions as 88 x 72.5 cm. Those material and object-identification details are strong evidence.

The more detailed wooden-core claims are less secure. Secondary technical literature reports seven to eight linden boards and about 10 mm thickness, with the board count explicitly visibility-qualified. Older Warnecke-attributed wording reports 10-12 mm thickness and split/core-derived linden, but the relevant Warnecke 1884 pages have not yet been directly read in the current research archive.

What The Evidence Does Not Prove

The strongest lesson from these shields is variation. The evidence does not support one universal medieval shield-core formula.

It does not prove that every 12th- or 13th-century shield used alder. Marburg points strongly to linden. It does not prove that every shield used the same number of boards. Szczecin, Seedorf and Marburg each point in different directions. It does not prove one standard thickness. Even the two Szczecin shields differ.

What This Means For Shieldwright Reconstruction

For workshop purposes, the evidence suggests several practical test paths rather than one finished answer.

A Szczecin-inspired experimental core would reasonably test thin alder planks, square/right-angled mating edges and a side-by-side arrangement, while keeping the six-vs-seven plank issue visible. Such an experiment should not be advertised as the exact original Szczecin construction unless the unresolved plank-count contradiction is later explained.

A Seedorf-inspired experimental core would reasonably test four alder boards with bevelled mating edges and structural edge-gluing. This is one of the clearer surviving examples for joint geometry, though the adhesive chemistry still remains a separate question.

A Marburg-inspired experimental core should be more cautious. Linden is well supported, but board count, thickness and joining need source-qualified handling. It may be practical to test linden boards around the reported thickness range, but the result would be a Shieldwright reconstruction hypothesis, not a proven Marburg recipe.

The most historically honest approach is not to ask, "What was the medieval shield core?" but "Which construction is best supported for this object, date, region and shield type?"

Provisional Construction Comparison

Conclusion

The surviving evidence shows that medieval shield cores could be carefully made multi-board structures, not crude slabs hidden under paint. But the evidence also resists a single answer.

Szczecin gives us thin alder plank construction with square internal edges and an unresolved six-vs-seven plank contradiction. Seedorf gives us four alder boards with bevelled, edge-glued joints. Marburg gives us linden and an important early 13th-century heraldic shield, but its board count, thickness and joining details still need direct primary verification.

For Shieldwright, that is a strong beginning. We can build historically informed experiments from this evidence, but we should keep the labels visible: what is proven, what is interpreted, what we choose to test, and what we still do not know.

That honesty is not a weakness in the reconstruction. It is the point.