The project arose from a group project of five students from the Expert class Type design (EcTd) course 2014–2015 of the Plantin Institute of Typography in Antwerp, led by Dr. Frank E. Blokland. The subject of the study was the 18th-century Belgian punchcutter Jacques-François Rosart. Walda, Michel and Lukas, three alumni of the course, further examined Rosart’s punches, matrices, foundry type, and related prints in collections in Antwerp and Haarlem.
The digital fonts presented include a range of text and display types, decorative capitals, and ornaments. It is undoubtedly the largest and most comprehensive collection of revivals based on the work of a single punchcutter to date.
Jacques-François Rosart was the most important punchcutter in the Netherlands in the 18th century, after Johann Michael Fleischmann.
Rosart was born in Namur, a city in the part of the Low Countries that nowadays is known as Belgium, in 1714, into a family of goldsmiths – not uncommon for punchcutters and engravers of the time. In 1740 he arrived as an autodidact in Haarlem, Holland. Rosart produced many fleurons and innovations on music type. He also produced Latin, Flemish and Hebrew alphabets and an extensive range of ornaments.
Rosart was relatively unsuccessful. The great printer Joh. Enschedé in Haarlem, preferred the work of Fleischmann (1707–1768) to that of Rosart. However, Enschedé was interested in his larger ‘display’ fonts, fleurons, and music type. Rosart wrote about the Enschedé type foundry: ‘I am the one who cut a large number of their fonts ‘[…]’. I hope to show the whole of Europe proof of my capacity.’
Unfortunately Rosart went bankrupt and had to sell his workshop, but he did not give up his practice and settled in Brussels in 1759. The Austrian government (which back then ruled present-day Belgium) set up a book trade and Rosart was fortunate to start a foundry under the patronage of the Duke of Lorraine. Over the next decade, Rosart published two specimen books, one in 1761 and one in 1768. Rosart had found success and sold his types to foundries in the Netherlands, France, and Germany before his death in 1777.
Although the Dutch Baroque style is clear in some letters, the types Rosart produced are classiﬁed as ‘transitional’. They combine elements of Renaissance archetypal models with that of the ‘Modern’ style, i.e., referring to aspects of writing with a broad nib and flexible pointed pen respectively. The ‘Garamond’ model, referring to the types of Claude Garamont (1499–1561), is typical of the Renaissance style, while the types from Giambattista Bodoni (1740–1813) are typical of the ‘Modern’ style. Rosart’s types contain a mixture of these styles. As with all kinds of time, Rosart cut each body size (today we would say ‘point size’) individually and named them according to the standards of his time, for example: ‘Moyenne’, ‘Cicero’, ‘Paragon’, ‘Nonpareille’ and ‘Grand Canon’. However, this does not mean that he has by definition used a different model for each body size. After all, frameworks can be scaled.
Images from the 1768 Rosart Specimen.
Box of original matrices ‘Garamonde Romaine’, stored in the Museum Plantin-Moretus.
Grand Canon Romain from the 1768 Rosart Specimen.
The very first starting point of the research and subsequent digitization was the Grand Canon Romain, as this larger model (about 30 pt on today’s computer) meant that the finer details of the letterforms were more visible.
Further sources for the study of Rosart’s idiom were two type specimens published by Rosart in Brussels in 1761 and 1768. In addition, the impressive standard work ‘Fonderies de caractères et leur matériel dans les Pays-Bas du XVe au XIXe siècle by Charles Enschedé’ published by the Enschedé type foundry in 1908, and the extensive and revised 1978 reprint (under the careful supervision of Bram de Does) were important sources. Another unique formative source for our design were the original punches and matrices of Rosart, which are preserved in the collection of the Museum Plantin-Moretus in Antwerp, Belgium.
When the Rosart Project started at the Expert class Type design course, the students soon came to the decision to split into two groups to design a text and a display version. The reason for this was that the Grand Canon Romain was originally not cut for text sizes. Nevertheless, the group responsible for the text-sized version used the digitization of this model as a starting point. While the group responsible for the display-sized version discussed and emphasized the delicate details of Rosart’s larger models, the ‘text group’ focused on tweaking the parameters to make the model suited for ‘reading’ sizes. As a first step, the contrast, the difference between the thick and thin strokes, was reduced. Moreover, the spacing was loosened and the proportions adjusted to better match the conventional look of typefaces for body text.
By the end of 2015 the Rosart revival project had forward momentum, but with so much of the focus having been on historical characteristics and interpretation, the text and display versions the group had produced only included a basic set of upper and lowercase letters, basic figures and punctuation marks. The time consuming effort of producing a fully functional typeface (that fits present day typographic needs) still lay ahead of them.
With the Expert class Type design course completed, three of the alumni, Walda, Michel, Lukas and the course director Frank, agreed to continue the Rosart revival project. Michel, who had already worked extensively on the ornaments, investigated further on the Open and Titling Capitals and collaborated with Walda to design Rosart’s Flourished Alphabets, while Lukas further developed the text and display styles.
Lukas reviewed the digital font files the two groups had produced and started to carefully expand the character set. While adding the basic characters for a bold version it became clear that certain characteristics of the Roman would have to be adjusted. At this point the workload began to grow considerably as more material was found, more characters were drawn and more glyphs needed to be modified, as well as the fact that both groups had yet to start designing the italic.
While gradually expanding the weights and character set of the text and display versions, the group searched for additional archival material to support their designs. Michel and Walda were fortunate enough to have access to the original punches at the Museum Plantin-Moretus and the Noord-Hollands Archief (Haarlem, the Netherlands), while Lukas explored recently digitized titles from various sources including the library of the Museum Plantin-Moretus. A key reference in this next stage was August Johan von Rösel’s ‘De Natuurlyke Historie der Insecten’, a rare example of Rosart’s printing types for reading sizes and a beautiful specimen in its own right.
August Johan Rösel von Rosenhof, ‘De Natuurlyke Historie der Insecten’, Haarlem by C.H. Bohn en H. de Wit, 1765–1788.
Images: Collectie Stad Antwerpen, Erfgoedbibliotheek Hendrik Conscience.
Lukas adds: ‘writing about the type-design process is difficult, as you work you continuously make spontaneous decisions without documenting them. At a certain stage you can see a form and a direction for the typeface and the process can only really be understood by comparing the outcomes with the original source material.’
In the Grand Canon Romain, for example, the capital letters are relatively large and much bolder than the lowercase. This was something quite fashionable in types during the Baroque period. As this kind of discrepancy would look erroneous in a contemporary typeface, the weight of the capitals was reduced to better match the lowercase. However, one advantage of the apparently uneven weight distribution in the source material was that it helped to imagine what more bolder weights should look like, at least for the capitals.
In the eighteenth century bold weights of type had not yet been invented, during this period to emphasise texts or darken words, black-letter types were used. It goes without saying then that Rosart never intended for his types to be bold and this made the creation of the bold weights all the more difficult as Lukas was adding weight to intricate counters and darkening areas within the anatomy of the type. Some shapes that read clearly in a lighter weight do not work well when made bold and with the expressive and playful forms of the Baroque, especially in the italics with their relatively steep angle and expressive curves, one quickly runs into problems.
Maintaining the finesse of the original model in bolder weights (while keeping counters and apertures open enough for use at smaller sizes) was a serious challenge. Because the italic forms are slightly more condensed than the roman, the darker and bolder certain shapes become, the more difficult it is to draw them in coherence with the original. Some letter-shapes, for example the italic ‘z’ (see Petit Canon Italique No. 78 for example), are rhythmic and complex, and it is hard to bolden them without losing some of the elegance and character Rosart intended.
All too often details are faded in the historical prints due to the technical limitations of the period, the quality of the paper, the consistency of the ink or the letterpress itself and the myriad of variables in the time that passes from then to now. In the art of reviving type, it is the punches that often offer unique insight into not only the tacit skill of the cutter, but the crucial decisions and nuances within the designs, intentional or not. The team had the good fortune to have access to the original punches cut by Rosart and to study these in detail and take photographs to work from.
Original punches of the Titling Capitals cut by Rosart – Photos: Walda Verbaenen.
Some details had to be modified either to avoid dark spots or to improve the distinctiveness of certain characters in smaller sizes. One example is the shape of the h of the italic cuts. In the original it has a ball terminal on the inside. (see Petit Canon Italique No. 785, digitized from ‘Fonderies de caractères […]’). This shape tends to look like lowercase letter b, which can be confusing especially when used in small sizes, to mitigate this for contemporary readers the new italic text version ‘h’ is based on the letter ’n’, making the shape more open and is easily distinguishable. For the sake of historical accuracy the original shape of the h (with the ball terminal) is included in the digital fonts as a stylistic alternate.
Petit Canon Italique No. 785 from the 1767 Ploos van Amstel brothers Specimen.
Petit Canon Italique No. 785 from the 1752 Rosart Specimen.
Petit Canon Italique No. 785 from the 1780 J. de Groot Specimen. – Joh. Enschedé en Zonen, ‘Fonderies de caractères […]’, 1908.
Developing the italics presented its own unique design challenges. The spacing of the asymetrical shapes of uppercase A, V and W for example – usually the white space between letters would be evened out in order to get homogeneous patterning and when text is set in all caps these asymetric shapes create white gaps at certain letter combinations, which can't be compensated by kerning. This was something the team had to get accustomed to and accept as a stylistic peculiarity of Rosart. The whole revival process revealed several similar challenges, sometimes the designers simply had to make compromises or accept inconsistencies, but it is exactly these incongruities that define the character and charm of Rosart’s types.
As Lukas puts it ‘sometimes these references led to even more confusion, like when you found a totally different design of a certain shape in another specimen. Basically the whole revival process was accompanied by the comparisons and careful considerations of how far things can differ from the original without changing the spirit of the design... a design which seemed to evolve or devolve depending on which specimen we looked at.
In the interest of contemporary typographic standards, certain things like the length of the serifs in the Roman had to be standardised while trying to maintain the overall appearance of the original material. This is also the case in the italic, the expressiveness of varied slopes and different angles of lettershapes had to be slightly reduced in the new version for a more consistent and contemporary look.’
Collection of various S shapes enlarged to the same size. The style of the serifs and the width differ. The large diversity of the shapes sometimes made it much harder to decide in which direction the digital version should take.
Different shapes of letter g – in this example its apparent Rosart didn’t always stick to a specific model. Left: Le Parangon Italique No.788 (Ploos) – 1784/1790 Right: Le Petit Canon Italique No. 785 (De Groot) - 1780.
Comparison of two different capital letters U. Notice the difference of the transition of the stems to the serifs. Left: an almost sharp connection. Right: a rather soft transition.
In summary it can be said that when designing a revival typeface it isn't always helpful to look at a vast quantity of archive research. From a design perspective, this project is a good example of how historical material, while always informative, can limit design and even contradict itself at times. At exactly the point where the designers assumed that they had seen all of ‘the Rosart’, something new popped up, which looked different again. This seemed counterproductive at times, because it questioned all progress and put doubt around early design decisions. Generally, it seems best to adhere to a certain specimen as the main reference, rather than using too many at the same time. Nonetheless, the project provided a unique insight into not only the types of Rosart in general, but also how he evolved certain ‘base’ models for roman and italic in new directions throughout his career.
Original punches cut by Rosart – Photos: Walda Verbaenen.
(Collection Enschedé, Noord-Hollands Archief, Haarlem).
Walda and Michel, who digitized some of Rosart’s Flourished Alphabets, have done an in-depth research into the historical material of Rosart as a basis for their designs. They dived into the collections of Museum Plantin-Moretus and at the Noord Hollands Archief / Collection Enschedé. Thanks to this opportunity, they could have a look at the original punches and printed material and captured everything (photo / scan) which proved to be very valuable to their research and to make decisions about the shapes of the serifs.
Comparison between Rosart (Black) and Pierre Simon Fournier (Red).
The four chosen series of punches of the Flourished Capitals were carefully photographed in detail. Other sources for this research were various specimens, in particular the ‘Typespecimen of J.F. Rosart, 1768’, ‘Proef van letteren, Enschedé 1773’, Épreuve des Caracteres de la Fonderie de la Veuve Decellier, 1779’, and ‘Typefoundries of the Netherlands, Enschedé, 1978’. Very helpful was also Charles Enschedé's report in ‘Fonderies de caractères […]’.
The punches and type specimen of these capitals in the Collection Enschedé were examined by Michel and photographed by Walda during their several visits.
Original punches of the Titling Capitals cut by Rosart – Photos: Walda Verbaenen.
Capitale de Paris – Rosart’s Titling Capitals.
Comparison of letters Rosart cut in Haarlem (left) and several years later in Brussels (right).
Open Capitals punches cut by Rosart, Collection Enschedé. Photo: Walda Verbaenen.
Work on the open and titling capitals is still in progress. For this, Michel is examining the type specimen and punches. By comparing the ﬂourished capitals with the titling capitals made in Haarlem and Brussels, Michel found out that the ﬂourished capitals were created on the framework of the titling capitals.
Black-Letter typefaces were probably developed especially for the German market, which the name «Duijts» suggests. Duijts is an archaic version of the dutch word “duits”, which means “deutsch” (German).
The revival of the Textura was much easier than the revival of the text and display types, because the model Rosart followed for the Black-Letter types was much clearer. There are only a few printed specimens and very few in larger sizes, which made the research and revival process pretty straightforward. The only difficulty was the lack of certain shapes to extend the characterset. In order to get an impression of how missing characters could appear, some printed specimens of Fleischmann’s Black-Letter were used as reference.
Within the process it became obvious that it would make sense to create two versions, one for large applications with fine hairlines and a high stroke contrast and a lower contrast version for smaller sizes.
Paragon Flamande – Black Letter / Textura – No. 900. Specimen of the Widow Decellier, 1779.
Text Flamande – Black Letter / Textura – No. 822. Specimen of the Widow Decellier, 1779.
Michel started his research into source material in the Museum Plantin-Moretus for the extensive digital series of ornaments. He then found out that the type-foundry artifacts of the historical printing company Johan Enschedé have been preserved and are now archived at the aforementioned Noord-Hollands Archief. After his first visit, Michel was overwhelmed by the vast amount of historical material, such as the punches cut by Rosart, as well as printed type specimen and even related written documents from the 18th century.
To really touch and feel this material was a great experience: examining and photographing the punches that Rosart personally delivered was undeniably awe-inspiring. After collecting a lot of material, Michel started digitizing the first ornaments, strongly inspired by their beauty and detailed execution.
One could question what the value of a revival is in a time in which more and more typefaces are developed. More speciﬁcally: what is exactly the added value of Rosart’s type? In his own time he was, for example, not considered of the same level as his famous contemporary Fleischmann. Also, centuries later Daniel Berkeley Updike was far from positive about Rosart’s œuvre in ‘Printing Types: Their History, Forms, and Use’ (1922). However, reviving Rosart’s work undoubtedly adds to the richness of the present-day digital typographic palette because of the speciﬁc unique qualities of his œuvre.
The booklet was developed by the five alumni of the Expert class Type design in 2015. It is not just a presentation of the outcomes of the Rosart Project: it is a concise manual and guide for everyone who wants to investigate historic type-foundry material and wants to produce a revival. The texts are in English and the illustrated booklet contains 44 pages (143 × 210 mm) and 500 copies were printed.
Le Rosart is Lukas Schneider’s revival of the types cut by Rosart, made with the support of Ray O’Meara. The contemporary interpretation is informed by extensive research into period sources. In addition to consulting various type specimens, this included studying Rosart’s original punches and matrices which are held at the library of the Museum Plantin-Moretus in Antwerp. Just like the work of his rival Fleischmann, Rosart’s designs exhibit a vertical contrast axis, long bracketed serifs, and bulbous terminals. The digital version adopts these features, together with baroque characteristics such as the ‘g’ with antenna, the bearded ‘G’, or the exposed fangs in ‘E’.
Revolver Type Foundry’s Le Rosart was conceived to meet the typographical and technical needs of today. The fonts are drawn in two optical sizes. The majestic Display styles are complemented by a humbler Text variant. Optimized for smaller sizes, Le Rosart Text is distinguished by sturdier hairlines, blunt serifs, and a more generous set width. The Display and Text families both include a range of ligatures, alternate forms, various sets of numerals including fractions, arrows and other symbols, as well as elements for composing braces. They each come in five weights with roman and italic styles, and are available for desktop, app, e-book and web licensing. Trial versions for testing purposes in layout applications are available on request.
For more information please contact: email@example.com
Besides various roman and italic styles, Rosart also cut a few blackletter types. A revival of Rosart’s types from the 18th century wouldn’t be complete without a Textura, or Flamande, as it was called in the Netherlands at the time.
In a similar fashion to his romans, Rosart’s blackletter types are characterized by a vertical contrast axis and sharp silhouettes, with relatively wide capitals and spruce details. Details like the ‘s’ that flexes its ball terminals and especially the decorative “feelers” on the extenders of ‘b’ or ‘p’ reveal its origin in the baroque period. Research of Rosart’s Flamande proved to be difficult, as only few printed examples have survived. In order to fill the gaps, the blackletter types cut by Rosart’s contemporary Fleischmann were used as a secondary reference.
Le Rosart Textura was produced in two styles. Named A and B, they are distinguished by the amount of contrast. The one with finer hairlines is suited for larger applications, and the darker version for smaller sizes. With more than 600 glyphs per style, the fonts support most Latin-based European languages. Le Rosart Textura is available for desktop, app, e-book and web licensing. Trial versions for testing purposes in layout applications are available on request.
For more information please contact: firstname.lastname@example.org
A set of 20 digital fonts redrawn after the original materials designed by Rosart. These fonts are a reinterpretation and were designed by Lukas Schneider.
A single digital font in two variants based on Rosart’s blackletter types from the 18th century. The fonts were revived by Lukas Schneider. These will be available soon.
The Rosart Ornament fonts contain about 865 pictograms and ornamental graphics originally designed by Rosart. These fonts were carefully revived by Michel Paré.
This series of Flourished Capitals consists of 4 digital fonts redrawn after the original designs of Rosart. The digital version was thoroughly revived by Walda Verbaenen and Michel Paré. These will be available soon.
© 2020 – Alumni of the Expert class Type design 2014–2015
at the Plantin Institute for Typography, Antwerp:
Ianthe Bato, Artur Frankowski, Ray O’Meara, Michel Paré, Lukas Schneider and Walda Verbaenen. The website was designed by Lukas Schneider
All rights reserved. No part of this website may be reproduced, distributed, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embodied in critical reviews and certain other noncommercial uses permitted by copyright law. For permission requests, write to the publisher at the email addresses below. Errors and omissions excepted.