Reviving A Historic Typeface

Further Reading on SmashingMag:

• Hands On The Sigmund Freud Typeface: Making A Font
• Understanding The Difference Between Type And Lettering
• Weird And Wonderful Typography – Yet Still Illegible
• Hands-On Experience: The Rehabilitation Of The Script

Later, when I got involved in typeface design, I realized that the process of reviving a typeface is comparable to the reconstructions done by paleontologists when they imagine how creatures long extinct might have looked. Even more fascinating is that both processes usually start with isolated findings and incomplete evidence, but imagination and informed speculation come to the rescue and help to fill in the missing pieces. If we look closely at the history of paleontology, we can see how both time and imagination have played a major role in the development of the science, for only these two components can enable us to go back in time to see these awe-inspiring creatures.

Both paleontology and typeface design seemed to be completely unrelated to me until they mixed in late 2008, when I had the experience of reviving a typeface used in a book printed long ago. What follows is the history of that, with my findings from the process of digitization and the result of the revived typeface.

Background

The Book

It is widely known that the remains of creatures that lived millions of years ago have been preserved thanks to the process of fossilization. As infrequently as a well-preserved fossil is found, so too is a well-printed and well-preserved 17th-century book discovered. This is the case of the source of my revival: a carefully composed but poorly printed copy of La Cicceide Legitima by Giovanni Francesco Lazzareli (third edition), a book of poems recounting the deeds of a man nicknamed Ciccio in Italy in the late-17th century. The book appears to have been printed in Venice in 1694 in a printshop named Herz.

Spread of La Cicceide Legitima showing the extent of deterioration of the book.

I came across this book in a second-hand bookstore in my hometown, Bogotá, Colombia, in 2003. I found the idea of reviving a very old typeface very attractive, but I didn’t realize how hard it would be, working with a sample in this condition. However, the thought of the freedom of interpretation this would give me helped me decide to use this book, dismissing two other better printed but less interesting candidates.

The book, a volume of 228 pages and nine signatures and measuring 9.8 × 16.5 centimeters, is printed on what appears to be highly absorbent, ordinary handmade paper. Three type sizes were used to set the text, and the one in small pica (about 10 points in size) features both roman and italic style. All of the pages appear to have been carelessly printed, the evidence of which includes excessive pressure, worn-out types (printing offices at the time would use a set of matrices for decades or even centuries) and unjustified or moved characters.

Historical Context

The history of a typeface is incomplete without some consideration of the context in which it was used. The one we’re interested in was produced in the early Baroque period (17th century), concurrent with a significant decline in the quality of books and typefaces produced in Italy. But it was also a period of great achievements in typography in other nations such as France and the Low Countries.

Europe in the 17th century falls in what we now call the Early Modern period, characterized by the Baroque cultural movement, the French Grand Siècle dominated by Louis XIV, and the beginning of modern science and philosophy, including the contributions of Galileo Galilei, René Descartes, Blaise Pascal, Isaac Newton. Europe was also torn apart by warfare throughout the century as a result of the Thirty Years’ War, the Great Turkish War, the end of the Dutch Revolt against Spain and the English Civil War, among others, while Europe’s colonization secured the Americas as a major economic resource of the new empires.

An intricate ornamental pattern found in the book, reminiscent of works from the Baroque period.

The Typeface

Precisely identifying our typeface was especially difficult, due in part to the ubiquity of some printing types at that time, as pointed out by M. Carter, 1961:

“Community of typefaces becomes particularly evident in Germany soon after 1500, and those faces that were common in Germany are found also in Italy, the Low Countries, England, Scotland and even, during the first thirty years of the century, in France, a country that has rarely imported type or matrices. Mr. A. F. Johnson has done some hard work on the German types of 1500 to 1540, and has left it on his record that as many as ten or twenty presses had founts made of the same set of punches during those years.”

On the other hand, some features of our typeface, such as the low contrast and the notably tilted stressing, correspond to an early (but not Venetian) roman type, of the kind known as “old style.” A few other features (such as the pointed shape of the right stroke of the roman “g” and the unusual treatment of the same letter in the italics) make the typeface hard to identify with more accuracy; although certain treatments — such as the horizontal crossbar of the “e” and the more dissimulated pen strokes — point to a French or Dutch roman, close to the typefaces cut by Garamond and Van den Keere. Another reason why the typefaces used in the book could be French is because Italian printers began buying matrices from France in the middle of the 16th century. Additionally, Garamond’s romans, like a handful of its predecessors, served dozens of printers in several countries.

A sample of three roman types with similar characteristics: Minion (blue), Garamond Pro (green) and Arno Pro (red). The overall proportions, stress axis and serif shapes, all features of garalde roman types, were borrowed for the revival.

If pressed to pinpoint this typeface, I would say it is a small pica roman old-style type, probably French, cut in the 16th or early-17th century, featuring medium extenders, medium contrast between thicks and thins and a good optical weight for long text settings.

The Process

Working From the Inside Out

Due to the low quality of my samples, I started with great uncertainties about the actual shape of each letter. All I had were tiny models blurred by the excessive pressure during printing, so I decided to select the best of every uppercase and lowercase letter, as well as every number, in order to make reasonable decisions.

I straightened and superimposed every sample as a separate layer in an image file so that the common areas tended to be darker than the eccentric forms produced by the spread of ink on the paper. I called the resulting shape the “maximum”: an area of ink potentially spread in all directions that would contain the DNA of the typeface.

Scans of several samples of the same letter were put on layers in order to find the maximum ink spread.

I looked for the “skeleton” of each letter (theoretically located in the middle of each stroke) in order to start drawing from the inside to the border of the ink spread (outwards). This skeleton would become one of my very few certainties during the entire process of revival.

The skeleton of the typeface emerged from the blurry letterforms. Like a digital paleontologist, all I would have to do afterwards was add the flesh and skin.

Optical Size and Weight

Upon finding the skeleton of each letter, I felt more confident to start drawing. The next logical step was to decide the visual weight of the original typeface — or at least of the one I wanted to create. This was done rather arbitrarily, just making sure to keep the width of the vertical stems of the lowercase inside the “maximum”, thus making the visual weight optimal for reading at 10 points in size. I decided to start with one fifth of the x-height, because I learned this has been one of the most established practices of typeface design in the last few centuries.

These would turn into the muscles and tendons of the new type.

Up to this point, no details at all were present. Perhaps the only additional decision I made was the result of thinking about what sorts of shapes could have been derived from the carving of a piece of metal just a few millimeters high with burins and files. According to Fred Smeijers, the shapes were the natural result of technical limitations:

“Not only do the tools invite the punchcutter to make the second n, but also this shape is easier to handle in the rest of the process of typeface design and punchcutting. It has no straight lines and no sharp corners. And the absence of these hard elements makes the form of the second n more acceptable to the human eye than that of the first. This more subtle shape has notable visual margin of tolerance. Hard straight lines make us wonder whether they are really straight or not. If they are indeed not quite straight, this is awkward to look at. So the punchcutter avoided such niggling questions and situations by building in a kind of visual doubt: no straight edges, no sharp corners. The forms become easy to handle, easy to mix and to bring into balance with each other.”

So, I decided to avoid sharp corners and straight edges, just as a punchcutter four centuries ago would have done. This decision proved very useful in helping me to distribute the visual weight of the letters at the baseline and at the x-height, thus producing a horizontal effect that was perhaps useful to achieving a nice word shape and, therefore, a comfortable reading experience.

The lessons of broad-nib pen calligraphy provided a good starting point for the proportion between the x-height and the stem width, since the model was initially obscured by the excess of ink and pressure.

At this point, I had drawn the lowercase and uppercase letter and the numbers. But the page had a certain blurriness overall that I found disturbing. The new typeface looked worn out and overused, just like the original, and that was not the effect I was going for. That made me realize that most of the smaller details needed special attention.

From Blurry to Sharp

The final shapes of the serifs and stem connections emerged from my analysis of existing types, some of them revivals and others not, such as Adobe Garamond Pro, Minion and Quadrata. From a careful observation of Adobe Garamond Pro, I realized that most of the round connections in my typeface were too round. Minion showed me the grace of sharper details; I also learned from it that some serifs needed harder and crisper edges to look more convincing. The process of borrowing details from similar typefaces is comparable to taking the scales, skin texture and color from living species during the process of reconstructing a dinosaur.

The evolution of the serif details: first draft (blue), intermediate stage (yellow) and final version (red).

The Caps Dilemma

Everything was progressing until I noticed a subtle yet important difference between the original typeface and my rendering: the caps were remarkably darker in the book. So, I decided to make my capital letters darker than the lowercase letters in order to stay true to what seemed to be the intention of the original designer and the convention at that time. This is a principle of optics: the larger the letterform, the darker strokes need to be in order to compensate for the excess of white in the counters. This principle has been followed since the invention of printing and is still used today. I exaggerated it a bit here to achieve an older-looking style.

Darker caps (right) are typical of old-style printing types and one strongly visible in my source book — something worth preserving in the revival.

My Additions and Contributions

Just as living species depend on mutation and adaptation to survive, typefaces too depend on features that enhance their performance in their natural environment. This seems to partly determine the degree of failure or success of print typefaces in the real world.

Even though many typefaces with features similar to those of Legitima must exist, many of them seem to me just too polished to capture the special atmosphere that old metal typeface gives to the page (perhaps with the exception of Quadraat and Adobe Garamond Pro). Legitima was designed from a 10-point original to work best when printed at the same size.

A certain awkwardness or imperfection present in the original was preserved, too, apparent in the bulkiness or fullness at the points where the strokes change direction. Among the other details, the uppercase letters were left purposefully heavy, and the drawing of the curves was meant to the recall the effect of the burin and file on old metal type.

Sketches of Legitima. The additional weight where the strokes change direction (top) as well as the diversity of angles in the italics (bottom) are some of the characteristics that were preserved in the digital fonts.

All of the features mentioned above, plus the slightly concave strokes (stronger at the top and bottom), contribute to making Legitima a very legible text type, rooted in the traditions of 17th-century Europe but with great expressive potential for our time.

Conclusion

The original typeface as it was printed compared to the final version of Legitima.

During the process of designing Legitima, I learned that reviving a typeface is not so much about bringing old shapes back to life as it is about preserving the qualities that justify its existence in today’s digital world. Merely tracing contours could be done by a machine, but bringing the spirit of a bygone age into the 21st century is inherently human, adds value to our time and contributes to preserving a cultural heritage that would be lost without the sensibilities of the designer.