Intercut wood typeface project typeface design and printing blocks; Spring 2006–present (work in progress)
(formerly known as protoTYPE)

Intercut print sample / type specimen 

By exploring dynamic approaches to modularity, appropriation, deconstruction, non-linearity, and formal abstraction, the aim of this project is to develop an experimental typographic system for letterpress printing which is both visually and conceptually interesting. In this system, the role of letters as visual representations of language becomes secondary to their power as expressive formal structures.

As such, it is to be decided whether or not the letters of the alphabet will enjoy the world outside of their Language Communication factory where they have been working diligently for so long now. However, I like to imagine this project as a unique opportunity for them to frolic freely—even if just for a fleeting moment—in the sunny fields otherwise occupied by the cryptic brush strokes of abstract artists.

My hope is that in this context people will be more apt to recognize the forms of letters so often trumped by their communicative functions. In some ways, these proverbial goblets were designed to not hold any wine at all.

 

Introduction

This project began as a semester-long undertaking for an Advanced Letterpress class taught at the MassArt Press by the honorable Al Gowan during my final semester at MassArt (spring, 2006). The assignment, as given to me, was relatively open-ended. Basically, I needed to come up with a letterpress-related project which I could develop over the course of one semester.

I knew from prior experience with my senior degree project, that I very much enjoyed semester-long design projects which I was allowed to define for myself. Eventually, I decided roughly to develop a project exploring the production of wooden printing type.

Though I feel I have reached the end of one stage of this project, I have way too many ideas to consider it finished. Thus, I consider most of the work here more as a proof of concept than a completed effort.

 

Research

cover of the book American Wood Type: 1828-1900 by Rob Roy KellyTo initiate the project, I researched a large amount on the design and manufacture of wood type, as well as its related (but now primarily non-existent) industry. An invaluable source for information and inspiration along the way (even still, as I type this) was Rob Roy Kelly's definitive book on the subject, American Wood Type, 1828–1900.1

Several topics in the book are of particular interest to me, and strongly influenced the development of my own wood type project. One such topic is the manufacture of wood type. In reading about traditional wood type (and examining it in detail, first hand), I gained much more respect for the conventional production methodologies utilized in making it.

More often than not, large display type is cut from cured and polished end-grain maple slabs.

slab of end-grain maple

Cutting is usually done with a pantograph router device. Pantographs mechanically scale down the motions of a movable arm as its operator traces an oversized letter template. By attaching a routing drill, the letters are cut as they are traced.

diagram of a pantograph router machine

Any fine details of the letter which can't be cut with a router (eg, the sharp corners inside the A, N, M, etc) are finished by hand before the block is ready for printing.

traditional wood type block (capital R)

An alternate approach to wood type production which was gaining in popularity at the end of the 19th century is die-stamping. The general concept of die-stamped wood type is as follows: using great amounts of pressure, a blank wood block is stamped with a metal die. The dies are constructed so that each letter's non-printing counter forms are raised, thus indenting the wood and leaving only the clean positive letterform remaining at the correct height for printing.

diagram explaining die stamping

Despite the great relative speed at which this die-stamping could be done, the metal dies were expensive to produce, especially considering that separate sets of dies were required for each size of font produced.

This die-stamping was interesting to me, however it was a very particular variation on the idea that truly affected my own project. In 1889, a man named George Setchell devised a modular variation on the method for the William H Page Wood Type Company. In this process, sectional dies were mixed and matched to create shapes that existed on more than one letter. For example, a die for the curved left half of the O could also be used to create the same shape of the C and G, etc. Page claimed that this sectional method reduced the amount of dies needed for a full alphabet by one third.2

One design-related subject of note is the frequency at which wood type designers of the 1800s would sample or even blatantly copy another person's work. Indeed, it was commonplace during the years which wood type excelled for a popular design to be duplicated, stretched, ornamented, or generally reworked in some form or another by several people simultaneously. In fact, it was not unheard of for a type designer to re-appropriate someone else's copy of one of their own designs.3 In this sense, it could be said that the design process of many wood types occurred as a full circle.

an example of Reverse typeAnother interesting design topic is that of Reverse types. Reverses, as might be guessed from the name, are made by simply reversing the letters out of a solid or decorated field. By simply placing together a line of letters (often with additional decorative end pieces, etc), printers are able to create cohesive, displays with relatively minimal effort.4

sample of multi-color Chromatic typeIn a similar vein as Reverses—though more complicated to print—are Chromatic types. Chromatics are designed to be printed in two or more colors from numerous sets of blocks. They are harder to print since they require that multiple colors printed from separate blocks must align perfectly with each other.5 When done correctly, however, the results can be amazing (as was demonstrated with the William H Page Wood Type Company's Specimens of Chromatic Wood Type, Borders, &c of 1874—often acclaimed as containing the most superb wood type specimens ever printed).6

I certainly was aware of many of the topics described here before beginning my project. However, it wasn't until reading Kelly's full account of their history that I gained full respect for them.

 

Design

type specimen of a Gothic typeface publshed in 1838Taking all the ideas I had gathered in my research on Chromatic types, Reversed types, sectional die stamping, etc, I began thinking about my own project. The progressively multi-faceted aspects of many of these approaches to type design (for their time) inspired my experimentation process, ultimately resulting in my contemporary interpretation of 19th-century wood type design concepts.

In Kelly's book I found a typeface which I chose to use as a base to start experimentation and design of my own alphabet. The fact that I started with an existing design has several significant implications. In terms of practicality, avoiding development of a design from scratch helped save valuable time which I was sure to need further down the road for the physical production of my blocks (remember that I originally was aiming to complete this project in one semester). In addition, starting with a decidedly 19th-century wood type style was a good way for me to pay my homage, visually, to the wood type designers of the past.

More interestingly, the idea of working off the back of someone else's design was an everyday occurrence in the traditional 19th-century industry of wood type (as is mentioned above). As a matter of fact, the typeface I worked from—as it appeared in an 1838 specimen from George Nesbitt—had already undergone appropriation at least once before I got to it, from a more condensed version originally published in 1837 by William Leavenworth.6 Thus, my seemingly lazy choice of a starting point is very much appropriate for the topic! 7

The underlying typeface I chose to begin experimentation with is a simple yet historically representative Gothic. By choosing to start with a very simple design, I not only ensured that I would have more flexibility when it came to developing my own design, but also that the logistics of technical production could be kept as straight-forward as possible (at least to begin with).

A specimen of the original typeface was scanned and a useful subset of characters (namely A, B, C, D, E, F, G, H, I, M, N, O, P, R, T, and U) were vectorized for further digital experimentation.

At this stage, I began tweaking the original letterforms to correct what I interpreted as design flaws or inconsistencies. Eventually this lead me (under the influence of the sectional die process, mentioned above), to precisely systematize many of the formal elements which could recur in multiple letters (eg, the rounded sides of circular letters or the straight vertical strokes forming the stems of other letters):

recurring shapes highlighted on multiple different letters

Next, I decided that as well as being able to print the letters of my alphabet, I also would like to enable the printing of each letter's reverse shapes. The concept of Reverse printing (as mentioned in my research above) is certainly nothing new. What is unique about my approach is that I eliminated any and all surrounding forms which would otherwise have given the reversed letter shapes a formal sense of context. With this lack of surrounding counter forms (especially in letters with flat sides) an interesting amount of figure-ground ambiguity was created:

letters and their counter forms

To increase to flexibility of what could be done with this pairing of letter and counter forms, I broke the counters down further still. This deconstruction created an even more abstract and varied range of unique shapes to serve as a palette from which I could compose in almost infinite ways.

collection of typographic shapes used for experimentation

Next I began to experiment with the effects which could be achieved by mixing and matching different pieces of different letter forms and counter forms, joining disparate pieces and removing others to push the limits of legibility and figure-ground ambiguity.

experimentation with forms and counter forms in black and white

The next progression was the addition of color…

experimentation with forms and counter forms in 3 colors

…and before not too long, I found it quite pleasing to throw all preconceptions of legibility and traditional typographic communication out the window, experimenting with a wide range of aspects like non-linearity, transparency, etc (all of which can be achieved with letterpress printing). Indeed, even from the incomplete alphabet of shapes I had created, I opened many windows for the use of typographically derived forms as compositional elements. These could function, depending on the technique of their application, more like brush strokes than like letters.

abstract experimental composition using typographic shapes

It is important to note that at this point all sketching had still only been done digitally. All of these experiments were simply to test the possibilities of what might be done with similar blocks in a real-life printing environment.

After designing a few digital test compositions, I was pleased with the wide range of possibilities of what could be done. I did feel, however, that I could afford to go a step further by increasing the complexity of the letterforms. I did this by grafting large slab serifs on to the existing sans serif letters.

addition of slab serifs

With an updated set of slab serif characters, I paid a visit to Wentworth Institute of Technology (where I would be producing the blocks) for technical advisement. It was here that some restraints were drawn to my attention. Most significantly: the fact that my letters were to be drilled out of solid blocks with a circular drill bit. The same way that you cant fit a square peg into a round hole, a round drill bit can not cut squared angles.

This created problems when it came to shaping any sharp corners in my letters, especially now that I had introduced many more right angles with the addition of the slab serifs.

diagram explaining the limitations of a circular drill bit

I could have kept the squarish designs and cut each corner by hand, but opted (in the interest of time and sanity) to simply soften up my design by rounding the angles. I altered the letters, making sure that no corners existed that were any tighter than was possible to cut with the ⅛" drill bit I would be using.

rounding of corners

comparison between original typeface and my adaptationBy this time, I am happy to say, my letters had changed enough where the underlying structure of the original 1838 sans serif was still recognizable, but the face had taken on a character of its own as a rounded slab serif.

Since my time and resources were limited, I decided to choose a small subset of my newly designed letters to produce as actual blocks. For this, I chose letters which were useful for spelling a wide range of words as well as for examining how various different shapes might work as blocks (many of the most uniquely shaped letters are also the least common in use). The final choice for letters: A, E, G, H, N, O, and R.

All of the letters and their related counter shapes would eventually exist on their own separate blocks, but to reduce wasted materials (and time spent at the router) I arranged the total of 27 separate pieces on to twelve 4×5 inch blocks. These could later be sawed apart with a fine band saw. To make this sawing a bit easier, I included a minimum amount of extra space between each piece.

layout of 27 different pieces 12 blocks, with cut lines

With a newly altered design arranged precisely (and the end of my semester closing in quickly), I was ready to get the production of the physical type blocks underway.

 

Production

The device I used to cut my blocks is a digital CNC router. The particular machine lives in one of the shops at the Wentworth Institute of Technology (across the street from MassArt). It is more-or-less a computer-driven plotter with a Dremel drill attached to it.

CNC machine

The first step in being able to direct this amazing machine to cut my letters was to convert them from the 2-dimensional drawings I had created in FontLab and Illustrator to 3-dimensional digital files. This was done using a 3-D modeling program called Cobalt.

building 3-D blocks in Cobalt

One might hope that if you have a digitally-driven router and a digital 3-D file, the only thing left to do would be to hit a print button. I'm sure this is possible in many cases, but it was not so for me. Before the 3-D digital models could control the CNC router, they had to be converted to what is called G-code. G-code is essentially a string of instructions telling the CNC machine what to do, and in what order; it defines the path which the router drill must follow to create a 3-D object. For my project, the G-code was created by a program called Unigraphix NX (which has since, in successive versions, been renamed simply to NX).

G-code path of a drill in Unigraphix NX

After many headaches and technology hurdles which I don't particularly wish to relive here, I was ready to make some blocks! But first I needed to choose the materials from which I would cut.

Unfortunately, on my meager student budget, wood similar to that traditionally used for wood type can become quite expensive very quickly (as much as $19 per block, in the 4×5 inch size I needed). For a minimum number of blocks of such material, I would have ended up paying as much as $200 or more (out of pocket).8

After weighing possible alternatives, I found a suitable replacement—at least for my experimental purposes—in Speedball brand linoleum printing blocks (available at your local Pearl art store for $2 per block). Linoleum is much, much softer than end-grain maple, and therefore is much less durable under printing. However, its softness does reduce resistance to drilling, allowing for much faster production times (a valuable quality when working on a tight schedule).

To test the productions setup, a few test blocks were cut from cheap foam as well as composite wood fiber. Though blocks made from these materials aren't any good for printing, they are helpful in setting up the router… as cheap as linoleum blocks are, pink foam is cheaper!

pink foam test block (capital A)

After the machine was all set up, it was time for the moment of truth: the drilling of my first printable wood type block.

video of block being drilled

video of block being drilled

Soon after successfully producing one single block of type (a capital R), my semester ended, and thus my entire undergraduate career. So when it came time to show what I had developed over the span of an entire semester, I only had one block and a few test prints to present as a physical manifestation of all my work. Luckily, my instructor knew all the effort and time that had gone in to getting that one block, so there weren't any problems.

single type block (capital R)

After the semester ended, however, I was not satisfied with just one block. There was no way that I would just let all that planning, design, and thought end at one block!

Unfortunately, since the school had ended, my access to the CNC machine at Wentworth was far less than reliable (to be honest I was extremely lucky to get as much help as I did).

Luckily, I was put in touch with an independent fabrication and prototyping shop in Somerville, MA. Here I was able to continue my production (with the amazingly generous help of Brent Martin) and complete all 27 of the blocks by the middle of August, 2006.

complete set of 27 pieces, including counter forms

 

In use

pile of Intercut blocks

Intercut type blocks set up on press

Intercut letters print

Intercut type blocks set up on press

Intercut counters print

 

 

Intercut printIntercut printIntercut print

 

Intercut print

 

 

 

Future

While do feel I have reached a sort of a milestone of accomplishment with this project, I still consider it a work in progress in some form or another. I still have tons of ideas about what I could possibly do with this project—some more realistic than others—and I have no desire to permanently shelve them.

So here, in no particular order, is a list of things I would like to do in the future based on my work so far:

  • Retrofit a Vandercook to have an electromagnetic press bed. This way (if I had my type mounted on the correct material) I could freely push the blocks around and compose with out having to take the time to lock everything up, as is customary with letterpress printing. Then printing could be much more fluid and easy to experiment with quickly: simply position your type by hand, flip a switch, and voilà! Your composition is instantly locked in place until you flip the switch off.
  • take a much longer time in the design phase, and develop my own typeface from scratch (ie, not work with an existing design as a base)
  • explore even more complex letter shapes
  • design a Chromatic type that relies on certain specific interactions caused by overprinting; consciously design the tertiary shapes that will occur when certain pieces overlap
  • deconstruct not just the counters, but the letters as well
  • develop a modular system for 3-D anaglyph typographic design and printing—not just stuff that vibrates when viewed through the red and blue glasses, but that actually recreates a 3-D effect
  • learn to use a pantograph router
  • create blocks that actually physically interlock with each other to create new shapes for printing (as opposed to having them simply bump up against one another)
 

Thanks

Before concluding my summary of the project thus far, I would like to thank the people who have supported me along the way; without their help and encouragement, none of this would have been possible.

From MassArt: Al Gowan, for overseeing this entire project;Keith Cross (my letterpress dad), for help finding post-grad routing opportunities;Brian Lucid, for the encouragement to push in progressive directions and setting me up with the folks at Wentworth; and all my classmates in the Advanced Letterpress class, for all the input and press-side goofball sessions.

From the Wentworth Institute of Technology: Josh Aigen, for all the help converting my 2-D designs into 3-D digital models; Sam Montague, for the advice on production, materials, and setting me up with the right people; and Peter Rourke, for the help setting up the router and ensuring I didn't route any letters into my arm.

From Kimo Inc: Brent Martin, for all the help with 3-D modeling and router setup and basically everything; and Kimo Griggs, for opening his shop to a broke student and offering some of his best workers' time.

Also:Dennis Ichiyama at Purdue University's department of Art & Design, for all the advice, encouragement, and beautiful prints of the original Leavenworth Gothic (!); Richard Zauft at Emerson University, for the design input and putting me in touch with the people at the Hamilton Wood Type Museum;Scott Polzen (one of the only other people I have met who occupies his free time with wood type production), for the mutual nerd-dom and the lovely sample prints.

Certainly: Mr Rob Roy Kelly (rest in peace), for the continuing inspiration and priceless contributions to the field of graphic design.

And finally: my parents, for putting me through college in the first place.