Did Medieval Shields Use Gesso? Grounds and Preparation Layers on Surviving Medieval Shields

A Note on the Evidence

The surviving evidence does not support a single universal shield-ground recipe. It reveals a family of related techniques that varied with place, date, materials, purpose and the kind of surface the shield maker wanted to create.

Grounds and preparation layers are among the least visible parts of a medieval shield, so this article separates several kinds of evidence with particular care. A conservator's analysis of a paint cross-section, a museum catalogue description, a scholar's reading, a proposed reconstruction and an open unknown are not the same kind of claim, even when they use the same word: gesso.

The sections below run roughly in date order. That order is a way of organising the shields, not a story of progress: a modelled relief ground already appears around 1200, and the most complex layer system described here belongs to a late-medieval surface. Several of the richest shields were also funerary, processional or votive objects rather than shields made for fighting, and they were more likely to be finished lavishly and to survive.

What Is a Shield Ground?

A ground is the prepared layer between the structural surface of an object and its final decoration.

In its simplest form, the sequence might be:

• wood → covering → preparation ground → paint

But surviving medieval shields show a range of different arrangements:

• wood → black organic layer → lime-white layer → paint
• wood → parchment → chalk ground → paint
• wood → parchment → chalk ground → silvering → paint
• wood → parchment → thick modelled ground → metallic leaf → paint
• wood → canvas → gesso → paper → leather

Some shields were even decorated with relief elements built from textile stiffened with the same sort of mineral preparation used for the painted surface.

The ground therefore had several possible functions. It could smooth the irregular surface beneath it, conceal wood grain, parchment texture or seams, provide a pale, bright base for pigments, or receive metallic leaf. In thicker applications it could also be shaped, stamped or modelled into relief.

Some medieval shield grounds were not simply primers. They were part of the sculpture of the shield.

Late Twelfth Century: Two Shields from Szczecin

Two late-twelfth-century kite shields survive from the early medieval stronghold area of Szczecin, and they must be kept apart. The painted shield is in the National Museum in Szczecin, inventory MNS/A/22250/1, dated c.1176–1200. The second shield, known as the unpainted shield, was only recently identified and is published together with the first by Dowen and colleagues; the museum record for the painted shield does not describe it.

The painted shield has the only analysed twelfth-century layer sequence among the shields in this article. Irena Jagielska examined paint samples under the microscope and by XRF.

• wood → black organic layer → lime-white layer → coloured paint

The lime-white layer is an observed layer. What it was for is a separate question. It lies under the colour and could have served as a pale base for it, but the analysis does not establish its function, and the function of the black layer beneath it is not established either. What matters here is what the samples do not show: no covering and no chalk or gypsum ground of the kind found on later shields.

The second, so-called unpainted shield is useful for a different reason. Calling it "unpainted" does not prove that medieval craftsmen left its wood deliberately bare, nor does its present state show that it never carried a facing, preparation layer or painted surface.

The absence of surviving gesso is not automatically evidence that gesso was never present.

Thin mineral and organic surface layers are among the parts of a shield most vulnerable to abrasion, moisture, burial, biological deterioration and later conservation.

Seedorf, Around 1200 to 1220

The shield traditionally associated with Arnold von Brienz at Seedorf provides much stronger evidence for a prepared ground over a covering. Johann Rudolf Rahn, who examined it in 1878, recorded parchment on the front with a chalk preparation. The conservation report by Matthias Senn and Franz Moser (1991/92) gives the layers in more detail. I know that report through Kohlmorgen and Dowen and colleagues, who both cite it.

The original appearance was also considerably richer than the shield's present muted colours suggest. The Swiss National Museum reports that analysis showed the now grey-green field was originally bright blue and that the lion was covered with silver leaf.

Two points matter here. First, the sequence: the wood provided structure, the parchment a continuous surface, the gypsum and chalk layers the painting ground, and pigment and silver the visible heraldry. Second, the lion: already around 1200, the preparation material was being built up into relief, not only spread flat beneath the paint.

Already around the beginning of the thirteenth century, therefore, the evidence points to something much closer to medieval panel-painting technology than to paint simply brushed onto boards.

Konrad von Thüringen, Around 1230 to 1240

The shield associated with Konrad von Thüringen is historically described as a funerary or memorial shield associated with his grave. Its museum record names a chalk ground beneath the polychrome decoration, and Hefner-Alteneck described the front as parchment-covered. The reverse was also covered, gilded and painted.

Its raised lion is a separate element of pressed leather, not a relief modelled in the ground. It is relief decoration of a different kind from the gesso lion at Seedorf.

That later damage shows the chalk preparation behaving as a physically distinct layer within the decorative system. Again, the structure was not merely wood followed by paint. It was substantially closer to:

• wood → parchment → chalk ground → decorative surface

Around 1250: The Vögte von Keseberg Shield

Many of the shields that follow come from the group that once hung in the Elisabethkirche in Marburg, and much of what is known of their surfaces comes from Jan Kohlmorgen's descriptions. The shield of the Vögte von Keseberg, which Kohlmorgen dates around 1250, is reported as having one piece of parchment on each face, followed by chalk ground and paint.

The precise composition of that chalk preparation and its binder has not been analytically established, but the existence of the ground itself is well supported.

The phrase "chalk ground" should not be turned into a precise modern recipe when the surviving evidence does not provide one.

The craftsmen clearly prepared the surface. Exactly how much chalk was used, what animal glue or other binder was employed, how many coats were applied, and how the mixture was worked may remain unknown.

Later Thirteenth Century: Nordeck, Liederbach and Arnsberg

Three Marburg shields that Kohlmorgen places in the second half of the thirteenth century show how differently a prepared surface could be used.

The Nordeck zu Rabenau shield has a parchment or parchment-like covering and a white chalk ground on the front. Kohlmorgen reports neither paint nor chalk-ground traces on the reverse. Because the rear covering survives only in fragments, that absence may be loss rather than choice. If the reverse was left unprepared, preparation may have been applied according to the needs of each face rather than automatically to the whole shield.

The shield attributed to W. von Liederbach shows another technique. Both faces carry chalk ground, and Kohlmorgen considers that the front was probably silvered all over before the coloured decoration was applied. Parts of the heraldry could therefore be created by leaving the metal itself visible while other areas were painted over it:

• wood → parchment → chalk ground → silvering → paint

A heraldic "white" or "silver" area need not have been painted with white pigment. It could actually have been metallic.

Because silver tarnishes, a surviving medieval shield may look very different today from the brilliant object originally carried or displayed.

The shield attributed to the Counts of Arnsberg has thin parchment-like coverings on both faces. On the front came chalk ground, followed by a dotted, modelled relief about one and a half millimetres high. The field was painted red. The heraldic eagle was silvered, but not every pale part of the bird was metallic: fine features such as individual feathers, legs and tail were executed in opaque white paint, while darker details were added separately.

• wood → parchment-like facing → chalk ground → modelled relief → silvering and paint

A "silver" animal could combine actual metallic leaf with painted white details.

1292 to 1308: Heinrich I of Hesse and a Relief of Stiffened Linen

The lion shield of Landgrave Heinrich I of Hesse is dated by Kohlmorgen to 1292–1308, a range taken from Heinrich's years as landgrave rather than from the object. One published caption instead names Heinrich "der Junker", his son. Both faces are reported as parchment-covered.

On the front was a gilded chalk ground. Above this was an elaborate heraldic lion made from two layers of linen impregnated with chalk ground, shaped and pierced to form the animal (Kohlmorgen). Helmut Nickel, writing in the Metropolitan Museum Journal, describes the same construction as a lace-like overlay of gesso-stiffened linen over a gilded background.

Seedorf's lion shows that a relief ground was nothing new by this date. What Heinrich's shield adds is a different way of making relief: here the preparation material stiffens a textile that is cut and shaped like a sculpture.

The shield maker was simultaneously painter, modeller and textile worker. The decorative surface was effectively built.

Around 1300: Welfenberg, Stromberg and Steinau

Three Marburg shields that Kohlmorgen dates around 1300 have grounds that were deliberately worked to create texture, light and relief.

On the Welfenberg or Helfenberg shield, the chalk ground carries modelled decoration. The background was worked into a damask-like grid with a four-petalled flower in each square and gilded, while the raised wolf-iron charge stands about two millimetres above the surrounding surface.

The Stromberg shield is reported with parchment on both faces followed by chalk ground, engobage or modelled preparation, silvering and paint; some later lists call the covering white-tanned leather instead. Its heraldic birds were built up in substantial relief; one shield catalogue gives the raised chalk mass as reaching approximately thirteen millimetres. Even if that measurement requires confirmation from conservation records, the surviving object clearly shows substantial modelling.

The Steinau genannt Steinrück shield carries a modelled chalk preparation with elaborate background ornament, including complex geometric forms and medallions with female heads, around its three heraldic wheels. The preparation was not something the viewer was supposed to forget existed: its relief was an intentional part of the shield's appearance.

The photograph at the top of this article, taken by Ludwig Bickell in the late nineteenth century, is catalogued as this shield. Two later copies of the Steinau shield also exist. The heavy losses in the photograph suggest the medieval original, but I have not been able to confirm from the record which object was photographed.

The Fourteenth Century: Schenk von Schweinsberg and Muschenheim

Two Marburg shields attributed to the Schenken von Schweinsberg survive. The older, which Kohlmorgen dates 1300–1320, is described with parchment on both faces and chalk ground beneath its painted decoration.

The younger shield, around 1350, gives an especially clear published sequence:

• parchment → chalk ground → silvering → paint

The shield of the Herren von Muschenheim, also around 1350, repeats a closely related sequence. One piece of parchment is reported on each face, followed by:

• chalk ground → silvering → painting

Its red field and silver chevron were therefore produced as part of a layered decorative system rather than by painting directly onto wood.

Across the Marburg group, repetition of parchment, chalk preparation, modelling, metallic leaf and paint is difficult to dismiss as an isolated workshop experiment. It represents a mature decorative technology. But it still does not justify claiming that every medieval shield everywhere was constructed this way, and the Marburg shields survived because they hung in a church, many of them as memorials.

Around 1320: Karl von Trier, Outside the Marburg Group

The shield associated with Karl von Trier, Grand Master of the Teutonic Order from 1311 to 1324, is not one of the Marburg shields. It came from Burg Reifenstein in South Tyrol and is now in the Tiroler Landesmuseen in Innsbruck. It is described in the heraldic literature as a Grand Master's processional or display shield. Its date of around 1320 rests on its painted heraldry; around 1900 Freiherr von Rahden argued that the board was an older shield painted later, and Bernhard Engel rejected that.

Its covering is described in three different ways: Engel wrote of a boiled cowhide on each face, the museum record gives a leather covering, and Kohlmorgen reports parchment on both faces. Which is right is not settled.

The front is painted on a yellow field, and Engel describes the reverse as painted too: by his time the outline of a lost lining could be read only from the paler paint it had covered.

That last point is a useful warning: in descriptions of shields, "ground" can mean the colour of the field as well as the prepared layer beneath the paint.

Textile Grounds and Later Re-Covering at Sion

The Bartolomei shield from Sion demonstrates another complication. Investigation revealed an earlier medieval shield beneath a later transformation.

The original mid-fourteenth-century shield had a linen or canvas surface carrying its heraldic decoration. In the later fifteenth century it was reactivated and another painted linen covering was applied over the earlier one.

The precise ground beneath these painted textile layers has not been established. That uncertainty is itself important: textile on a medieval shield does not automatically mean linen → gesso → paint. There may have been such a preparation, but unless analysis demonstrates it, it should remain an open question.

c. 1376: The Black Prince's Funerary Shield

The shield among the funerary achievements of Edward of Woodstock, the Black Prince (died 1376), at Canterbury Cathedral is a funerary shield, not a fighting one, and it has one of the most unusual recorded sequences.

Guy Francis Laking described a light wood joined down the centre, covered on the front with:

• wood → white canvas → gesso → paper → leather

He also describes the fleurs-de-lis as embossed leather. Later accounts call the wood poplar; the shield has since been scanned and conserved at Canterbury, but I have not seen a published report of those results. The exact relationship between the layers still needs that conservation evidence, but one conclusion is secure: gesso was not necessarily the last preparation immediately beneath paint. It could occur inside a much more complicated composite surface.

Medieval shield makers were solving structural and decorative problems rather than following a modern textbook recipe.

The Oña Shield D 60: A Late-Medieval Stratigraphy

Two shields in the Real Armería in Madrid, D 59 and D 60, come from the cloister of the monastery of San Salvador de Oña near Burgos. Their shapes belong to types of the twelfth century and around 1200, and older literature dated them accordingly. I place them last because the most detailed study of them, by Álvaro Soler del Campo (1996), dates their surviving decoration much later.

So the age of the wooden shield and the age of its surviving decoration are separate questions. The shape points to around 1200; the layers that were analysed point to the late Middle Ages or later.

Even on one shield, there was not necessarily one universal ground recipe.

D 60 is therefore the most detailed layer analysis in this article, and it shows how different the faces of one shield could be: a glue-bound system under gold leaf and azurite on the front, an oil-bound one under red on the reverse. But it is evidence for a late-medieval, probably votive surface. It is not evidence for how shields were prepared between the twelfth and fourteenth centuries.

Chalk, Gypsum, Gesso, Stucco: What the Words Can Carry

The question "Did medieval shields use gesso?" is partly a question about words. The sources for these shields use several of them, and they are not interchangeable.

• At Seedorf the conservation report distinguishes two materials in one ground: a gypsum layer on the parchment and a chalk layer above it. That shows that "gypsum" and "chalk" can name different layers on the same shield.
• For the Marburg shields, Kohlmorgen writes of chalk ground (Kreidegrund) and of engobage for the thicker modelled paste. Neither term comes with an analysis of the material.
• Laking, writing in English about the Black Prince's shield, uses "gesso".
• Soler del Campo, who analysed the Oña layers, writes of gypsum stucco.
• Some descriptions use "ground" for the colour of the heraldic field, as with Karl von Trier's yellow ground.

All of these are modern words, from the nineteenth century onwards. None comes from a medieval technical description of a shield.

Unless a shield has been scientifically analysed, its description should keep to the terminology its evidence actually supports.

So, Did Medieval Shields Use Gesso?

Yes. Surviving medieval shields provide analysed and published evidence for mineral preparation layers that modern sources describe variously as gesso, gypsum, chalk ground and stucco.

But that answer needs qualification. There was no single medieval shield-gesso system demonstrated across all surviving shields.

• The analysed twelfth-century sequence from the painted Szczecin shield has a lime-white layer over a black organic layer, with no covering and no chalk or gypsum ground in the sampled areas.
• Some shields have chalk preparation.
• Some have gypsum or gesso, and at Seedorf both gypsum and chalk are reported in the same ground.
• Glue-bound grounds are reported at Seedorf.
• One late-medieval shield, Oña D 60, shows different binding systems on its two faces.
• Some grounds were thin painting preparations, and others were built into substantial relief, already at Seedorf around 1200.
• Some supported paint, some supported silver leaf or gold leaf, and some were used to stiffen and model textile decoration.

And for a number of surviving shields, the preparation remains unknown.

What the Surviving Shields Actually Show

Taken together, the shields I have studied suggest a recurring manufacturing principle rather than a universal recipe.

The painted face of a high-quality shield was often constructed as a sequence of specialised layers. The wooden core provided strength and shape. Parchment, leather or textile could isolate and unify that core. A mineral preparation could then create a controllable decorative surface — polished smooth, modelled into relief, used to stiffen textile, covered in silver or gold, or painted with heraldic colours.

The finished shield was therefore not simply painted carpentry. It could combine the crafts of the woodworker, parchment worker, painter, gilder and modeller in a single object.

Perhaps the most useful reconstruction model is consequently not:

• wood → gesso → paint

but:

• structural core → facing or covering where evidenced → preparation appropriate to that object → relief or metallic treatment where evidenced → paint → later coatings or interventions where evidenced

That sequence should never be imposed on a shield for which the evidence says something different.

A Note for Historical Reconstruction

For modern reconstruction, the evidence argues against selecting one convenient "medieval gesso recipe" and applying it to every medieval shield.

A plausible medieval technique is not automatically the documented technique of a particular surviving object.

That distinction is central to experimental archaeology.

Conclusion

The surviving evidence changes how a medieval shield should be imagined. Under the heraldic lion, eagle, cross or chevron there could be an entire hidden technological structure: parchment stretched across timber, layers of chalk or gypsum bound with organic material, relief built from prepared textile or thick mineral paste, silver or gold leaf, and finally coloured paint. On other shields there was far less, and on many the answer is simply not known.

Centuries of abrasion, tarnishing and material loss have flattened much of that complexity. Originally, some of these shields must have been remarkably bright and three-dimensional objects.

Their grounds were not merely preparation for the artwork. They were part of the artwork itself.