Transparent Paper as a Medium of Copying and Design in the Early Modern Architectural Workshop

Anna Bortolozzi

Abstract
This article explores the use and function of the lucido technique in architectural workshops from the Renaissance through the late eighteenth century. By examining evidence from written sources and key drawing collections, the study compares the copying practices of architects with those of artists. The findings reveal that transparent paper was not appreciated as a copying medium in Europe’s architectural workshops until the mid-eighteenth century. When employed, transparent paper was primarily used for copying figure and ornamental drawings that were challenging to transfer using the pricking technique. The paper argues that the marginalization of transparent paper in architectural practice was possibly due to the coating process and the characteristics of the substances employed – vegetable oils and resins – which were incompatible with the working environment of architects. It was only with the commercialization of machine-made wove transparent paper in the early nineteenth century that architects and engineers began to systematically adopt this medium.

Introduction

[1] The architectural treatises and manuals written in the sixteenth and early seventeenth centuries were conspicuously silent on the techniques employed by architects to copy drawings. The historiography of drawing often assumes that Renaissance architects painstakingly copied designs line by line using compasses and rulers – a time-consuming method. Yet several semi-mechanical copying methods, including pricking techniques, calco (stylus incisions), counterproofs, and carta lucida (transparent paper), had been described in artists’ handbooks and practised in Italian and French workshops since the early fifteenth century.¹ Their primary purpose was to expedite the copying process and ensure fidelity to the original work, especially before a piece of art left the studio. They could also serve as tools for exploring compositions in reverse, incorporating additional details, and making other design changes, and so they were essential in the design of new artworks. Indeed, even if scant evidence of the use of semi-mechanical copying methods has been found in collections of architectural drawings, the close connection between artistic and architectural education in the Renaissance, and the frequent overlap between artistic and architectural practice, are reasons to believe that architects were well versed in the semi-mechanical copying methods favoured by painters and other practitioners such as engravers.

[2] The relative lack of interest in the material dimension of architectural drawing, which extends to its historical media and supports, is evident in the literature. Recent publications have begun to remedy this, emphasising the value of examining architectural drawings with a refined material lens.² As a contribution to the understanding of architects’ working processes and of the materiality of their drawings, this study considers the evidence of the use and function of the lucido technique in architectural workshops from the Renaissance to the end of the eighteenth century.³ In this period transparent paper was not industrially produced and commercialised; architects had to prepare it themselves by coating sheets with vegetable oils and resins.⁴ Over time, paper treated with oils and resins becomes brittle, suggesting that numerous once-extant tracings may have perished from neglect or natural degradation. The scarcity of tracings from the Early Modern period in drawing collections may also have been compounded by their primarily utilitarian nature and the traditional undervaluation of copies.⁵

[3] Interestingly, with the introduction of machine-made transparent paper in the early nineteenth century, architects and engineers began to systematically employ this medium. Transparent paper allowed lines from one drawing to be selectively re-used in subsequent drawings. It offered the opportunity to compare solutions visually, to make the process reversible, and to see the architectural project as a pile of overlapping, interrelated layers, thus presenting architects with an unprecedentedly versatile design tool.⁶

Transparent paper and other semi-mechanical copying methods

[4] The earliest instructions for the use of transparent paper as a copying medium in architectural practice date back to 1750, when Louis-Charles Dupain de Montesson (1715–1790), a military engineer and geographer, included it in his drawing manual La science des ombres.⁷ An original placed under a transparent sheet of paper could be traced precisely and effortlessly, without any damage to the original. After having traced the desired image onto the transparent sheet, the copyist could either prick and pounce or trace it with a stylus onto a new working surface. Transparent copies could be also laid onto another sheet of paper. In Montesson’s account, transparent paper (papier huilé) was a method to make an exact copy of a plan drawing, together with the techniques of pricking (piquer), tracing on a pane of glass (contretirer), and tracing with an intermediate sheet of paper rubbed on the back with chalk (calquer).⁸ Of the four semi-mechanical copying methods described by Montesson, surviving collections indicate that since the Renaissance the pricking technique was the one most practised by architects. Pricking involved placing the original on top of the paper intended to receive the copy and transferring the principal points by pricking through the original with a very fine needle (Fig. 1).⁹ The draughtsman would then outline the copy with chalk, pen, and straight edge. Nicolas Buchotte (1673–1757), a French military engineer and contemporary of Montesson’s, said the method was suitable for orthogonal drawings such as plans, elevations and sections, but not for topographic maps or perspective and decorative drawings.¹⁰

1 Bernardo Buontalenti (1531–1608), Wall fountain with aquatic plants, detail. The sheet has been folded and pricked to transfer the drawing on the right-hand side to the left-hand side. Black chalk, pen and brown ink on white laid paper, 431 × 224 mm. Le Gallerie degli Uffizi, Florence, GDSU 2325A (photo: © Gabinetto Fotografico delle Gallerie degli Uffizi)

[5] The contretirer method, also known as tirer à la vître, involved placing the drawing to be copied on a pane of glass with a sheet of paper on top of it. The light passing through the glass ensured the original drawing was fully visible, making it easier to copy. The first author who advised architects to use glass for copying a drawing "line by line" was Augustin Charles D’Aviler in his 1691 Cours d’architecture.¹¹ In the 1754 edition of his Les règles du dessin et du lavis, Buchotte reproduced the image of a wooden frame containing a pane of glass used for tracing drawings "a la vître" (Fig. 2).

2 A wooden-framed pane of glass for tracing "à la vître". Nicolas Buchotte, Les règles du dessein et du lavis, Paris 1754 [first pub. 1722], pl. 1 (photo: Universitäts-Bibliothek Heidelberg, DOI: https://doi.org/10.11588/diglit.37119#0239)

According to Buchotte, this method was useful for copying perspective drawings, architectural ornaments, garden parterres, and figure drawings with asymmetrical, organic lines that were difficult to copy using pricking.¹² He also noted that tirer à la vître could not be used with thick paper of the type generally favoured by architects – "grand Aigle", "grand Colombier", "Nom de Jesus" – because the lines of the original would not be visible.¹³ These three qualities of paper corresponded to the three largest sizes of sheets available at the time in France, measuring 24 × 35 pouces (c.610 × 890 mm), 21 × 31 pouces (c.530 × 790 mm), and 18 × 25 pouces (460 × 630 mm) respectively.¹⁴ To address this limitation, a thin type of paper called "à la Cerpente" or "à la Serpente" that was coated with oil of turpentine was recommended for tracing drawings that could not be pricked.¹⁵

[6] The last method mentioned by Montesson, known as calco, involved incising the original drawing with a stylus after rubbing the back of the sheet with chalk (Fig. 3a-b).¹⁶ To avoid colouring the original, a sheet of paper rubbed with chalk could be inserted between it and the copy. Montesson said calco was a less commonly method because of the potential damage it could cause to the original drawing.¹⁷ If tracing was necessary, he recommended making an exact copy of the original on transparent paper and then tracing the copy with a stylus instead of the original.¹⁸ The hesitancy about the calco technique seems to have been general among architects, as little evidence of its use can be found.¹⁹

3a-b Pierre Bullet (1639–1716), Study for a "Colonne Ludovise", a monument to Louis XIV in the form of a victory column, c.1680–1690: (left) detail of the recto under raking light with the indented lines made by the stylus used to transfer the design onto another sheet; (right) detail of the verso rubbed with black chalk. Graphite, pen and brown ink, brush and brown wash on three joined sheets of white laid paper, 1.239 × 300 mm. Nationalmuseum, Stockholm, NMH THC 6858 (photo: Cecilia Heisser / Nationalmuseum, Stockholm)

[7] From the eighteenth-century sources, it would seem the choice of the copying method depended on the size of the sheet (large versus small), the subject matter (plans and elevations versus perspectival and decorative drawings), the drawing technique (straight edge versus free-hand), and on the desired level of accuracy and speed of the copying process. The material characteristics of both original and copy also played a role in the choice. Additionally, the preference for a specific transfer process was influenced by the value attached to the original and the need to preserve it. Ultimately, the novelty of Montesson’s instructions was not his description of the lucido technique per se, but his sanctioning of it as a suitable method for copying orthographic representations – then the most common convention in architectural drawing.²⁰

Tracing in the workshops of Renaissance artists

[8] While architects’ and engineers’ drawing manuals were significantly late to acknowledge the use of transparent paper as a copying medium, artists’ handbooks described the technique as early as the fifteenth century. Artists’ handbooks also provided various recipes for making paper transparent, using common substances found in artists’ workshops, such as vegetable oils for binding colours and turpentine for varnishing.²¹ Cennino Cennini (Trattato della pittura, c.1400) said to coat paper with linseed oil, and recommended it be left to dry for several days.²² Raffaello Borghini (Il Riposo, 1584) found walnut oil superior to linseed oil because of its smoothness.²³ From the seventeenth century, art manuals suggested either warm oil of turpentine (known as Venetian turpentine), a natural balsam made by distilling the resin of larch trees,²⁴ or a mixture of oil of turpentine and plant oils.²⁵ In order to take a correct outline, whether the original was on paper, panel, or wall, Cennino Cennini’s instructions were to put transparent paper over the figure or design, fastening it lightly at the four corners with a piece of red or green wax. The figure or design would be immediately visible through the transparent paper. Cennini said to trace the outlines and extremities of the original design in ink with a fine nibbed pen or a small miniver brush, and then, having removed the transparent paper from the original, add the highlights and rilievos.²⁶ The copyist could then either prick and pounce (spolverare) the transparent sheet or trace it with a stylus onto a new working surface (calcare), if necessary with the aid of an intermediate sheet rubbed on the back with chalk.

[9] Tracings served a variety of purposes in artistic practice. They were used to consider compositions in reverse, sometimes with details added or with other changes; to create records of artworks before they left the studio; and to replicate works when required.²⁷ Despite the abundance of written sources about the use of transparent paper in artists’ workshops, few tracings from the Early Modern period have survived.²⁸ The earliest preserved examples of artists’ tracings in drawing collections are known from the early sixteenth-century Netherlands. Among them are two notable pieces. One is a compilation sheet featuring studies of "houses and castle formations".²⁹ It seems to have been done for training purposes and involved tracing selected models from the Errera Sketchbook by Herri met de Bles (c.1500–1560). The other depicts the Descent from the Cross and derived from a painting by the workshop of Gérard David (c.1460–1523), with the composition reversed.³⁰ In the sixteenth century, the practice of replicating paintings using lucidi became widespread in artist’s studios and beyond.³¹ Historical sources indicate that even low-skilled artisans, copyists, and amateurs used this technique to make fraudulent copies of famous artists’ works.³² Unfortunately, this sometimes resulted in severe damage to the original paintings.

[10] In the same period, engravers frequently used oils or resins to impregnate drawings before etching them – an alternative method to the counterproof for transferring a design, reversed, onto the plate.³³ Because of the pressure made by the stylus to indent the lines of the drawings, sheets were severely damaged or more often than not destroyed. The 19 transparent papers made by the French artist and engraver Étienne du Pérac (c.1520–1604) for the series of etchings issued in 1575 as I vestigi dell’antichità di Roma are rare examples of drawings that survived the process.³⁴ In his 1645 treatise on the art of intaglio printmaking, the French engraver, painter, and architect Abraham Bosse (1602–1676) expressed concern for the preservation of the original and advised the alternative method of making a traced copy to transfer the design onto the plate.³⁵ Tracings surviving from printers’ workshops show that paper was often treated after the drawing was done, not before – this is hinted at when only the surface of the sheet with the tracing was coated, while edges remained untreated. Drawings by the Italian draughtsman and architect Francesco Antonio Bufalini for Insignium Romae templorum prospectus exteriores interioresque (Rome, 1684) and by the Swedish draughtsman and military architect Erik Dahlberg for Suecia antiqua et hodierna (Stockholm, 1690–1714) are examples of this practice.³⁶

Tracing in the studios of Renaissance architects

[11] The first examples of drawings on oiled paper that can be tied to architectural practice date to the 1490s. In this group are some 50 tracings in the Houfe Album³⁷ and a handful of other tracings in the Metropolitan Museum of Art (Fig. 4)³⁸, the Uffizi (Fig. 5)³⁹, and the Ashmolean Museum⁴⁰. What all these traced copies had in common was that they were taken from drawings by the architect and engineer Francesco di Giorgio Martini (1439–1501) and his circle. Measuring c.300 × 250 mm, they featured ancient monuments, architectural details (including classical capitals, bases, and entablatures), and classical sculpture. Their different drawing styles, however, indicated they were the product of at least three different hands. Useful elements of comparison are offered by the drawings in the Houfe Album depicting triumphal arches, especially alongside the corresponding motifs in the drawings of the Metropolitan Museum and the Uffizi.

4 (left) Anon. after Francesco di Giorgio Martini, Arch of Septimius Severus, c.1501. Pen and ink, brush and brown wash on tracing paper, 281 × 257 mm. The Metropolitan Museum of Art, New York, inv. 80.3.632, Gift of Cornelius Vanderbilt, 1880 (photo: The Metropolitan Museum of Art, New York) | 5 (right) Aristotile da Sangallo (attr.) after Francesco di Giorgio Martini, Arch of Janus (left) and Arch of Septimius Severus (right), c.1501. Pen and brown ink on tracing paper laid onto another sheet, 370 × 280 mm. Le Gallerie degli Uffizi, Florence, GDSU 6711A (photo: © Gabinetto Fotografico delle Gallerie degli Uffizi)

The Houfe draughtsman used a straight edge and brown wash technique, while the Metropolitan Museums’s and Uffizi’s copyists drew free-hand using pen and ink only. The Uffizi’s copyist omitted the inscription from the attic of the Arch of Septimius Severus, which was present in the Houfe drawing, and used etched lines instead of ink wash to convey the sculptural value of the bas-reliefs. The Metropolitan Museum’s copyist accurately reproduced the inscriptions on the attics (incorporating additional inscriptions from the lower part of the arch), but his rendering of the sculptural elements of the arches was coarse. The Uffizi tracing, which combined half of the Arch of Janus and half of the Arch of Septimius Severus on the same sheet, stood out from the group in terms of composition: not only did it allow for an immediate visual comparison of the two arches, but it may also have been a deliberate choice by the draughtsman to save time and effort when copying.

[12] The extent of this set of tracings is not known, but it should be noted that they do not differ from other contemporaneous copies made on laid paper. Because the paper was treated with oil, Arnold Nesselrath suggests the Houfe Album drawings were made in anticipation of a never-finalised architectural treatise by Francesco di Giorgio Martini.⁴¹ Yet the lack of indentations on the Houfe tracings suggests at least another equally compelling hypothesis: the various sets of copies of Francesco di Giorgio Martini were executed as model drawings for the workshop or for collecting purposes.⁴² The popularity of Francesco di Giorgio’s drawings of antiquities in the first decades of the fifteenth century is borne out by the numerous copies scattered in other collections.⁴³ Significantly, several of the copies were done using transparent paper, a medium that facilitated the production of series of copies from one prototype. Traced copies were often then transferred to laid paper and finalised with ink wash to give them the appearance of finished drawings.⁴⁴

[13] Other examples of early sixteenth-century traced copies from model books of antiquities can be found in the Larger Talman Album in the Ashmolean Museum (Fig. 6).⁴⁵ The work of an anonymous draughtsman between c.1520 and 1550, the tracings feature a composite capital, two highly ornamented Corinthian capitals with putti, and two bases. The two Corinthian capitals appear to be exact copies of two prototypes in a model book of antiquities by the Master of the Oxford Album, a collaborator of Jacopo Ripanda (c.1465–1516?).⁴⁶ In the eighteenth century, the five traced copies were trimmed and pasted into an album of drawings owned by the collector John Talman (1677–1726).

6 Anon., Three capitals and two column bases, c.1520–1550. Pen and dark brown ink on three pieces of tracing paper joined together, 320 × 205 mm. Ashmolean Museum, Oxford, WA1942.55.157.2 (photo: © Ashmolean Museum, University of Oxford)

[14] The last of these sets of traced copies, originally a bound volume, is held by the Drawing Matter Collections in London.⁴⁷ Dating to the mid sixteenth century and consisting of 64 oiled sheets drawn on both sides in pen and brown ink, the volume had traced copies of plans of contemporaneous churches, classical buildings, tombs and mausolea, classical fragments, capitals and entablatures, machines, anatomical studies, trophies, and other subjects, often with inscriptions (Fig. 7).

7 Studio of Giovanni Antonio Dosio (attr.), Sketch of milling machine, measured ornamented cornice, ribbon/label ornament, c.1550. Pen and ink on tracing paper, 300 × 200 mm. Drawing Matter Collections, London, inv. 2159.63 and inv. 2159.6364 (photo: © Drawing Matter Collections)

Similar designs for machines can be found among the drawings of Francesco di Giorgio Martini; the sources of the other tracings remain unknown. The style suggests that at least two people had a hand in the book, arguably the architect Giovanni Antonio Dosio (1533–1610) and a mid-sixteenth-century draughtsman associated with Antonio da Sangallo the Younger’s studio.⁴⁸ The accumulative character of several of the sheets and the lack of alignment of the drawings with the edges suggest the book was created as a compilation of models for the workshop and not for commercial purposes.

Tracing in the eighteenth century

[15] Almost no architectural tracings are known from the seventeenth century. In this period copy drawings of ancient and modern architecture continued to be made and collected, but not on transparent paper.⁴⁹ Instead, it was in the eighteenth century that transparent paper appears to have resurfaced as the copying medium of choice, not only in architectural practice, but also in the pursuit of scholarly work and publications. This resurgence was likely driven by a renewed interest in the study of antiquities and the subsequent demand for copies from untapped collections of manuscripts and drawings. In a letter dated 1726, Francesco Scipione Maffei, a writer and historian, declared his intention to copy an ancient papyrus preserved in Geneva by soaking a thin sheet of paper with "olio di sasso", the petroleum found in the Sassuolo district near Modena.⁵⁰ Maffei also explained that at the time the practice of tracing manuscripts using transparent paper "was granted as a favour in all royal libraries".⁵¹ Maffei’s antiquarian interests resulted in a series of 58 tracings of ancient buildings, now preserved in the Bibliothèque Carré d’Art in Nîmes.⁵² The tracings were likely purchased or commissioned around 1750 by Maffei’s pupil and secretary, Jean-François Séguier, from Renaissance manuscripts preserved in Rome, including Giuliano da Sangallo’s Barberini Codex and Pirro Ligorio’s Antichità.⁵³ Included in the same series were three tracings of contemporaneous buildings, including Bernini’s Roman church Sant’Andrea al Quirinale and the Jardins de la Fontaine in Nîmes.⁵⁴

[16] Maffei was not the only eighteenth-century antiquarian to commission copies of Giuliano da Sangallo’s drawings of antiquities on transparent paper. In the 1780s, in anticipation of his Histoire de l'art par les monumens (1810–1823), Jean-Baptiste Séroux d'Agincourt asked the architect Léon Dufourny to make tracings from the Codice Barberiniano Latino 4424 in the Vatican Library, the so-called Sienese sketchbook in the Biblioteca comunale degli Intronati in Siena, and the Codice Resta in the Biblioteca Comunale in Palermo.⁵⁵ Séroux d'Agincourt appreciated Renaissance drawings as objects of study and historical documents that could help him to recall the image of post-classical monuments. Transparent paper was therefore recognised by him as the most suitable medium for making exact copies of the originals.⁵⁶ A number of Dufourny’s tracings are preserved in the collection of the École nationale supérieure des Beaux-Arts in Paris, while a large corpus of tracings by other artists is preserved in the fourteen volumes of preparatory material for the Histoire de l'art par les monumens, donated by Séroux d'Agincourt to the Vatican Library.⁵⁷

[17] Another series of tracings of a similar date that drew on historic sources was a volume titled "Modern and Antique Ceilings Coloured", now in the RIBA Collections in London.⁵⁸ Produced in the office of the architect Robert Adam (1728–1792), the volume has 60 tracings of different sizes (ranging from c.270 × 270 mm to c.550 × 320 mm) depicting Roman stucco ceilings discovered on the Esquiline and Palatine hills, wall decorations and grotesque works, and plans of ancient mausolea (Fig. 8).

8 Agostino Brunias (c.1730–1796) after Francesco Bartoli (1675–1730), Antique lunette from the Palatine, 1760. Pen and brown ink and watercolour on tracing paper. RIBA Collections, London, vol. 54 (formerly vol. 78), "Album of copy drawings, maybe from published sources, by Pietro Santi Bartoli, Francesco Bartoli and others, by the office of Robert Adam", fol. 20 (photo: © RIBA Collections)

Their prototypes can be found in drawings and prints issued by Pietro Santi Bartoli (1635–1700), Francesco Bartoli (1670–1733), Francesco Contini (1599–1669), and Anne Claude Count de Caylus (1692–1765). Notably, several of the tracings depicting ceilings were coloured with blue, red, yellow, and green watercolour, faithfully reproducing the original drawings by Pietro and Francesco Bartoli.⁵⁹ The close resemblance between the traced copies and the exquisite neoclassical ceilings designed by the Adam’s office leaves no doubts about their function as models.⁶⁰

[18] Transparent paper’s growing popularity in the later eighteenth century as a medium for multiple sets of copies of classical motifs is further demonstrated by the numerous tracings made by Jacques-Louis David (1748–1825) and his pupils.⁶¹ The Louvre holds three albums of tracings by David from his study period in Italy and Rome between c.1775 and 1780.⁶² Drawn with different media, including pen and brown ink, black chalk, and grey wash, David’s tracings were copies of contemporaneous print works devoted to antiquities, including Pierre d’Hancarville’s Antiquités étrusques, grecques et romaines tirées du cabinet de Hamilton (1806) and Johann Joachim Winckelmann’s Monumenti antichi inediti (1767), as well as other untraced sources.

[19] David’s use of transparent paper as a copying medium seems to have been embraced by his pupils, as confirmed by two large albums of tracings by the Swedish artist Per Krafft the Younger (1777–1863), now held by the Royal Swedish Academy of Fine Arts in Stockholm.⁶³ The method of tracing by means of transparent paper was also soon an established practice at the École des Beaux-Arts in Paris, where students in the third year were required to produce traced copies of classical monuments and other antiquities.⁶⁴

Transparent paper and architectural practice in the eighteenth century

[20] The surviving evidence suggests it was in the first half of the eighteenth century that architects began to use transparent paper in their ordinary practice. This would be consistent with the historical sources, which acknowledged transparent paper for the first time as an alternative copying medium in 1750.⁶⁵ Unlike the Renaissance architects before them, though, eighteenth-century architects did not limit their use of transparent paper to copying classical prototypes: they used it to trace original project drawings by contemporary architects and to trace their own project drawings.

[21] One of the earliest substantial groups of architectural tracings from the period is held by the Nationalmuseum in Stockholm.⁶⁶ The some 600 sheets of tracings were made between c.1720 and 1780 in the studios of the Swedish architects – and successive Superintendents of the Public Works – Carl Hårleman (1700–1753) and Carl Johan Cronstedt (1709–1777), and were subsequently incorporated in the collections of drawings gathered by the two architects, now known as the Tessin-Hårleman Collection and the Cronstedt Collection.⁶⁷ Hårleman and Cronstedt used transparent paper to copy drawings by French and Italian masters during their educational travels on the Continent in the 1720s and 1730s (Fig. 9).⁶⁸ They also used it to copy drawings by other Swedish architects stored in the Superintendent’s office or in private collections such as Carl Gustaf Tessin’s. Hårleman’s and Cronstedt’s tracings feature a range of subjects, including plans, elevations and sections of residential and religious buildings, architectural elements, interior decoration, sculpture, antiquities, furniture, garden designs, fountains, infrastructure, coaches, and theatrical sets and costumes. However, as large part of the sheets depicted interior decoration, it would seem Hårleman deliberately chose transparent paper over pricking, the better to copy the extravagant, asymmetrical designs of the French Rococo style (Fig. 10).