Reconstructing Heather-Reinforced Earth: Material Understanding and Embodied Knowledge in Ms. Fr. 640

Xin Gao
Fall 2025, HIST GU4962: Making and Knowing in Early Modern Europe: Hands-On History

Abstract

This project reconstructs a sixteenth-century recipe for earthen wall construction recorded in BnF Ms. Fr. 640 (fol. 14r). The manuscript describes tamping moistened earth between wooden boards and strengthening it by intermixing it with branches of heather or similar things. Through this process, early modern builders combined local materials with practical, embodied knowledge to create durable structures. Using small-scale reconstruction, the study investigates how soil, water, and heathers behave during mixing, layering, and compaction. The making process is treated as a reconstruction experiment that reveals the sensory judgments, gestures, and material reasoning embedded in the original recipe. By integrating textual analysis, experimental making, and reflective documentation, the project demonstrates how artisanal practice functioned as a form of embodied knowledge.

1. Introduction

The recipe for constructing earthen walls recorded in the French manuscript BnF Ms. Fr. 640 (fol. 14r, “For walls of earth and rustic construction”) describes a technique in which moistened earth is mixed with branches of heather or similar materials and tamped between wooden boards. Like many technical instructions found in early modern artisanal writings, the recipe reduces a complex craft process to brief cues, encoding shared knowledge only partially captured in textual form. The recipe reflects a broader epistemic feature of early modern making: key operational judgments were not transmitted primarily through writing but through sensory experience, embodied practice, and continuous interaction with materials.¹

This study treats the recipe as both a technical procedure and a record of embodied knowledge. Through small-scale reconstruction, it examines how early builders may have assessed moisture content, manipulated earthen mixtures, integrated plants, and modulated tamping force, forms of situated judgment implied but not articulated in the manuscript. Reconstruction thus becomes an interpretive method, transforming the recipe from a static text into a dynamic making process.

Methodologically, the project integrates three complementary components:

1. Textual analysis, which translates, contextualizes, and interprets the recipe within the framework of early modern craft theory;

. Experimental reconstruction, using soil and dried heather to reproduce the operational sequence and observe the material behaviors; and

3. Reflective documentation, which records tactile, visual, and acoustic feedback encountered during mixing, forming, and tamping.

This combined approach makes it possible to access dimensions of knowledge inaccessible through textual analysis alone, especially the multisensory judgments required to interpret indeterminate terms such as “moistened earth” or “beaten & tamped down.”

Situated within the interrelations among text, material behavior, and embodied practice, this study demonstrates that reconstruction serves not only to recover historical techniques but also to reconstruct the epistemic modes through which early modern craftsmen understood and transformed the natural world.[^] In doing so, it brings into view the traditional intelligence embedded in early modern earth construction, wherein human sensories, natural materials, and environmental conditions jointly constitute the technical system.

. Textual Analysis of the Recipe

2.1 Material logic and regional practice in sixteenth-century Europe

The recipe on fol. 14r begins with an interesting and remarkable observation: swallows, the author notes, teach us this technique by building nests from earth reinforced with plants. This parallel between animal and human construction provides an entry into early modern ecological thinking. The manuscript frames earth and plants as elements of a shared environmental logic, in which the structural behaviors of natural materials could be learned through observing animal “engineers” in nature.

This ecological framing provides an interpretive foundation for understanding the material choices that follow. Across sixteenth-century Europe, earthen construction formed a coherent body of embodied knowledge shaped by local environment, material availability, and craft traditions. Rather than relying on abstract standards, builders operated through regionally grounded material logics.

This material logic can be observed in sixteenth-century Italian military engineering. Giovanni Battista Belluzzi’s Trattato delle fortificazioni di terra (c. 1550) offers a systematic account of earthen fortification techniques that rely extensively on local plants, including brushwood (frasca), bundled shrubs (stipa), vines (vimini), and timber stakes, to stabilize soil, regulate moisture, and accommodate settlement.²

In the heathland ecologies of western and northwestern Europe, regions spanning France, Scotland, Ireland, northern England, and parts of Scandinavia, nutrient-poor, acidic soils supported dense growth of Calluna vulgaris (heather).³ Such environments produced a material landscape in which earth and locally available plants formed the primary resources for vernacular construction. In this context, heather functioned as a regional analogue to the plant reinforcements described by Belluzzi.

Historical evidence confirms the widespread and sustained use of heather within vernacular earthen systems. In France, heathers were routinely mixed into rammed earth.⁴ In Scotland, Ireland, and northern England, heather and bracken were incorporated into walls and roofing systems.⁵ In Scandinavia, layered mats of heather provided structural and moisture-buffering functions in sod buildings.⁶

Besides the practices, the Toulouse manuscript Fr. 640 offers an instance of explicit textual evidence concerning heather use in earthen construction. On folio 14r, it clearly describes mixing heather (originally in French: bruyère) with earth, placing the mixture in one-foot layers, and compacting each layer with specialized wooden tools. The recipe is structured through a sequence of operative verbs, revealing how material states were assessed through tactile, visual, and auditory cues rather than through abstract measurement. Heather is not presented as an incidental inclusion; it appears as an intentional component mixed into the operative sequence of rammed-earth construction, implying a function tested and confirmed through practice.

BnF Ms. Fr. 640, fol. 14r

For walls of earth and rustic construction* ¹

Swallows have taught us this craft, making their nests out of mud mixed with wisps & stalks of hay or straw to make it bond. Therefore, in places where stone & brick are lacking, one can use earth to make partitions & walls. And for this, light earth, which does not form clumps when plowed, but which is as if intermixed with arene, holds first rank, because it can be beaten & tamped down better. It is true that one needs to moisten it & la cut it into the shape of sods with a ditch-spade, and thus place & arrange it. This one lasts longer and there is not as much construction work, and not dryness does not make it crack & split. But, because such kinds of earth are not found everywhere, those who are on good & fertile land, after marking out with a measuring line the width & length of their foundations, drive into the ground in along the edges, on this side & that, long ch poles & or chevron beams to support boards between which they throw the earth, making each layer one foot thick or thereabouts, intermixing it as if S.S.S*² with branches of heather or similar things, then with beaters of three different forms they tamp it & beat it. One is called the mallet, which has a triangular form like A, and with this, one first tamps the earth. Next one uses one which is made of blocks of wood pointed at the tip & helved to a large stick, & this one is for pressing the earth well at the extremities and edges of the wall, which adhere to the boards, & is called.*³

The other one is called the bat, which is for flattening & beating the earth for the last time, as shown in .C.*⁴ Then one makes another layer of earth & heather and beats it as was said, & continues thus until the wall is complete, which one covers with heather & then with earth. Some m intermix rows of bricks in the said wall. They also make the wall tapering, giving a width to the foundations according to how high one wants to raise the wall. Which, when old, whitens, & thereby shows that it has saltpeter in it. That is why, when they fall down, gunpowder makers profit from them.

Notes: [1] On earth fortifications at this time, see Daniela Lamberini, Il Sanmarino. Giovan Battista Belluzzi architetto militare e trattatista del cinquecento (Florence: Leo S. Olschki, 2007), which discusses a manuscript treatise from ca. 1550 by architect Giovanni Battista Belluzzi (1506–1554).

[2] Abbreviation for stratum super stratum (Latin: “layer upon layer”) used in alchemical writings, among others.

[3] Word omitted (no blank space). The author-practitioner seems to have forgotten to name this tool. Presumably, he is referring to the middle instrument in the marginal drawing, which he may have intended to label “B.”

[4] Though none of the instruments in the margin is marked “C,” the author-practitioner is presumably referring to the bottom drawing.

2.2 Embodied Material Knowledge Encoded in the Text

Unlike modern materials science, sixteenth-century earthen construction relied on no standardized tests or quantifiable parameters. Instead, its knowledge system was fundamentally embodied: builders assessed material behavior through touch, resistance, moisture, sound, and the visible behavior of soils and plants under manipulation. Fr. 640 captures this sensory-material epistemology not through explicit explanation, but through the operations, adjectives, and evaluative cues embedded in its language.⁷

The vocabulary of the recipe, such as “moisten,” “cut,” “intermix,” “place,” “arrange,” “beat,” and “tamp,” functions as an archive of embodied operations. Each verb encodes a bodily technique while simultaneously pointing to a material state that must be recognized and judged. Table 1 summarizes these categories of embodied knowledge, ranging from sensory judgement, to tool-mediated actions, to assessments of structural performance.

Table 1: Summary of embodied knowledge categories in Fr. 640

At the same time, the manuscript embeds implicit material logic. Descriptions such as “light earth, which does not form clumps,” “moisten it,” or “cut it into the shape of sods” indicate how builders inferred mechanical and material properties through embodied engagement. Table 2 maps these textual cues to the material judgements they imply. In this system, knowing how soil-plant composites behave mechanically was inseparable from the sensory experience of working them. This embodied epistemology provides the conceptual basis for the experimental reconstruction undertaken in this study.

Table 2: Textual evidence mapped to sensory judgement and material logic

3. Experimental Reconstruction

Beyond this, however, the entry leaves much unsaid. It does not specify material proportions, nor does it clarify what is meant by terms such as “intermixing,” “tamping,” or “moistened earth.” This textual indeterminacy is not an omission but reflects knowledge that was expected to be acquired through making. Reconstruction, therefore, becomes a method of interpreting not only the recipe itself but also its silences. By making the mixture, this project seeks to test how earth and heather interact with hands and tools, and how, in turn, the maker’s movements are shaped by their behavior.

3.1 Materials and tools

For the reconstruction, I selected a clay-rich soil (soil a) and sieved it to remove particles larger than 5 mm, producing a more uniform and workable base material. When necessary, I incorporated a small proportion of sand-rich soil (soil b) to adjust texture and reduce excessive plasticity during mixing. Dried heather branches served as the organic reinforcing component. I cut locally available heather into lengths of approximately 30 mm to approximate the dimensions implied in the manuscript and to facilitate even distribution within the mixture. Water functioned as the sole wetting and binding agent and was added incrementally as I mixed, until the material reached a consistency suitable for compaction. I then placed the prepared mixture into wooden molds measuring 20 × 20 × 10 cm and compacted it using wooden tools. These included a tamper (used both for tamping and flat beating), a triangular tool (functioning alternately as a pointed edge tool and a triangular beater), and additional flat beating implements (see Figure 1).

Figure 1: Materials and tools

3.2 Procedure: Making and Knowing

3.2.1 Soil Preparation

The manuscript distinguishes between two categories of soil: “light earth” and “fertile land.” I understand light earth as a soil intermixed with sand, characterized by a loose and lightweight structure and exhibiting limited plasticity when moistened. By contrast, fertile land in this context does not refer to agriculturally productive soil, but to an earth with greater inherent cohesion, offering builders a viable alternative when light earth was unavailable. I therefore interpret fertile land as a clay-rich soil with lower sand content and higher natural plasticity.

During the preparation process, I first manually broke apart large clods of soil and then sieved the material to remove particles larger than 5 mm, eliminating coarse stones and organic debris. This process transformed the initially irregular and heterogeneous soil into a finer, more uniform material that was more workable for the subsequent stages of mixing and compaction.

Figure 2: Soil preparation process

3.2.2 Heather Preparation and Incorporation

The manuscript’s instruction to “intermix it with branches of heather or similar things” provides a clear indication of the intended material category, yet offers no detail regarding the size, form, or specific preparation of the plant matter. Drawing on both historical and contemporary construction practices, where plant materials such as straw are routinely incorporated into earthen mixes, I inferred that such materials would have undergone basic preparation prior to use. Accordingly, I employed dried heather branches and cut into segments of approximately 30 mm. This length approximates the structural role implied by the term “branches” in the manuscript while ensuring adequate dispersibility within a small-scale model.

Figure 3: Heather preparation process

3.2.3 Mixing and Establishing Workability

After initially combining the soil and heather, I adjusted the moisture content of the mixture to achieve a workable consistency suitable for forming and compaction. The manuscript refers only to “moistened earth,” a term that encompasses a wide range of possible conditions and offers no technical specification. I therefore determined the appropriate level of hydration by attending to the observable behavior of the material and to sensory feedback obtained through manual handling.

When the mixture contained excessive moisture, it adhered strongly to my hands; when insufficiently wetted, the soil failed to adequately coat and integrate with the plants, compromising both the continuity of the mix and its response to compaction. By incrementally controlling the addition of water and repeatedly kneading and pressing the material, I brought the mixture into a cohesive, plastic, and non-sticky state (see Figure 4).

Figure 4: Moisture content and test

(1) Low moisture: cohesive in hand but crumbles on impact

When squeezed in the hand, the mixture holds together lightly and does not stick to the skin. However, when dropped freely, it immediately breaks apart into loose granules. This indicates insufficient hydration and low plasticity.

(2) Moderately medium moisture: partial cohesion with minor stickiness

The mixture shows slight adhesion to the hand and contains small, partially hydrated aggregates. When dropped, it breaks into a combination of granules and small clods.

(3) & (4) Excess moisture: overly sticky, adheres to tools during compaction

The initial hydration for testing was too high. During ramming, soil adhered to the wooden rammer, forming lumps on the tool surface. This indicates over-saturation, reducing compaction efficiency.