Frank Blokland: Harmonics, Patterns and Dynamics

These are live blog notes from held in Zurich

Usual disclaimer for live blogging: These are informal notes taken by Dave Crossland at the event, and may or may not be similar to what was said by the people who spoke on these topics. Probably if something here is incorrect it is because Dave mistyped it or misunderstood, and if anyone wants corrections, they should email him immediately ( – or post a comment. Thanks!



Frank Blokland: Harmonics, Patterns and Dynamics

The regularization, standardization, systematization, and unitization of roman and italic type from the Renaissance origins until the Romain du Ro: Harmonics, patterns and dynamics in formal typographic representations of the Latin script

This is my PhD research. I started it 6 years ago, and I blog about my work at

You will find there the research question and subsequent questions: “To what extent are the the harmonic…..”

Its a difficult quesiton, and evidence for such regularization must be found to start. That is what I’ve done, I will show you my models, and the research on Granjon and Garamond I did at the Plantin Moretus Moseum. Nothing durvives from the italian renaissance but there is a direct connection to the French.

Then there are follow up questions, “How exactly then did these regularizations…” and “Moxon and Fournier do not seem to mention forenamed regularizations?” and for me the most interesting idea, “Did they preserve optical preferences from handwriting origins, or did they create new ones due to the forced canonicalisation of letterforms?”

Mythical Eye

Stanley Morrison said “Hving learned and memorised the TRUE proportions of Moille, Paciolo and others, the punchcutters and printers relied on thir eyes and not upon measuing tools”

Pierre Simon Fournier ‘These gentlemen woudl have been to a single rule, the eye is the supreme judge’

Adrian Hutt said “Some indefinable talent in the best punch cutters” – but if we all believe in this, yet can’t define it, does it really exist?


What is typography? How is it differnet to callifraophY? In calligraphy the space ˆs divided by pen strokes; there is no care about line gaps, he can prevent collision on a letter by letter basis.

This is Jenson! The first ‘roman’ typeface. Tsch, you should know this 🙂 There is a clear division of each line, the page image is made by assembling these rectangles.

The spacing between characters is always correct, irrespecive of the sequence of the rectangles.

Dynamics Rectangle

This is the 1455 gutenberg bible type. I wondered if there was a relation fo the sizes. No one knows why its “Em square” or why its written as E and m. This could be the origin,

[image of Emssp]

I could find this with a root 2 rectangle or a golden section rectangle.

This is applied to Silmbach’s Jenson, and you find the SAME system!

The same in Guternberg and Jenson. Jenson went to see Guternberg, he used textura type, and its AMAZING you can find the same rule in the gothic and roman type.

Em/En Square

Tgs=us is the Adobe Garamond, he worked for the Manutius publisher, and the Hypno… poliphili. Griffo took the Jenson model, and I made from this

[the \mmhpH diagram]

You can contruct from this model gothic and renaissance type.

I made the DTL LetterModeller with help from colleagues at URW. I can change the ascenders, descenders, pen width, flatten the curves, and make the step from renaissance to gothic type with this step.

It is a very small step!

Its the same model, with some tweaking of the pen width, the arches.

You can export the shapes generated from the program and use them as a basis for your own type.

[ \mmphB ]

This gas a similar set of proportions. When you look at the serifs, and compare to the previous model, you can see I am here fitting to include the serifs. Is this fitting the model to the evidence?

This is a display type from van der Keere, the same time as Garamond, working for Plantin, and he made a variant of a Garamond type, with shorter ascenders and descedners, and I can find the same proportions with the exception of the ‘e’ – and where it is tweaked to fit the model it works better.

It could be that it was cut later on and then combined.

When you make the ltters more condensed, the ascedners become shorter, in this model. If you have less space inside the letters, you need less space around the letters, between and above/below.


Type is a decision from handwriting to movable type. How was this decision made?

You read that this handwriting was the basis for Roamn type. BUt there is a large step from one to the other. This is a Mainz bible from 1452, likelyt he basis for the gothic type by Guternberg. This is a small step from one to the other.

I don’t think its a coincidence that printing happened in Germany at that time; the system worked immediately familiarly.

Here are 4 ‘quendam alium’ strings, a book scan of Jenson type, 2 letter model instances, and Adobe Jenson.


This is Huston Shuffer (?) from 1457, you can see the clear system of stem intervals. This is a simple model showing the same, the nnnnnnn, the hominum and hominum

Then I took Jenson’s type and put it on the same fencepost rhtyhm.

I thought, if I was Jenson, I would have to do a lot; make the shapes, make the spacing for casting the type. Could it be that Jenson who knew the Gothic system just applied that system?

Anyone here who has designed type, you know that its a lot of work to design the letters and the spacing. You can look at a catalog for ideas of how to solve solutions. But he couldn’t.

You can do things so easy today with computers, the problems of casting. People say today that its hard, but really type design is very easy! 🙂 Don’t tell anyone though, its great marketing.

Fence Posting

You can apply the rhythm to all the letters, more or less.

This is Griffo’s type, Monotype’s Polihilos, from the Hypno … and its still the same system.

So that is the base lowercase forms. What about Capitals?

Jensons Capitals on top, the Trajan type and a Carolingian miniscule hand.

Morison doesn’t provide much information on how these are related.

[quote of Morison]

I placed the caps on teh same grid, and You can see they fit in the same grid. Type designers STANDARDISE things, make a SYSTEM, and I believe that is also what he did.

You can see the H and N are quite wide, you may see it too wide. Griffo came along and made them more condensed, but he was tweaking the base system.

As soon as you have a system, you can add a little bit of exceptions to this, and you see this in the red line.

So here you can see the system applied.

This is Monotype Bembo. This is the betters from the Edna (?) also published by Manitius. This tweaks the letters more.

These are origins from Jenson, looking for differences between the original paper Jenson and Adobe Jenson.


I seem to be the first person to put the letters on a grid.

THis is Times New Roman, Bembo book, and you can see they fit in the same grid.

The monotype machines required a grid too.

Catisch writes how Roman stone carving seem to have been designed on a grid. Was this researched in the renaissance? Don’t know. But the roman carving was studied.

You must tranfer the written forms to the punchs, and you need a grid for this.

There is a 5.5×5.5 (6×6) unit grid. The rhym of stems isn’t seen as important in contemporary type design, but this is crucial. If you make the serifs on the right longer you can make the stem rhtym more regular. [NOT SURE I GOT THAT PART RIGHT]

Here you see the Garamond letters fitting the Romain du Roi grid.

Unit arrangement

Could you fit from left edge of left step to left edge of right stem? or center of counter to the sidebearing edges?

Here are 4 images

1. the letters ‘omnibus’ spaced on fenceposting of mmmmmmm

2. fence posting with unit correction 16 n units

3. fence posting unit correction with 32 n units

4. this is adobe jenson fitting.

**My 8 year old daughter can space Jenson as well as Slimbach can with this model!**

You can do spacing with no knowledge of typgraphy. Anyone can apply the system.

Cadance Units

Here is the original Adobe Jenson spacing and the ‘Cadence’  spacing, and you can see it comes very close.

Here is the original Adobe Garamond spacing and the ‘Cadence’  spacing, and you can see it also comes very close, not as close but its good.

As I said, the Romain du Roi is shown as the first grid made type, but Moxon’s has a grid. he doesn’t show the lines but I find the same grid.

So it exists at least 100 years before the RdR, and I suspect it goes back earlier.

Monotype had an 18 unit machine, but you can see the factory spacing and the cadence spacing are close.



The same. So what you learn is ‘optical’ is just conditioning to the system.


So I went to Plantin Moretus and looked at actual metal type. I find the same system.

In 1959, there were letters cast by the PM museum and the spacing is quite uneven. You can lie the types down on the table and see this

[image of q o n h g on their sides]

And you can see the type from the 16th C has the IDENTICAL widths. This means the casting was systematized and SIMPLIFIED.

You can place the matrices in rows with the same width


And you can find these shared widths in the original matrices. The brass was made like a line of chocolate bars, a bar with the same width that is the sliced into the matrices.


There is a small film on the site, we used the matrics that are insured for €100,000 and we were just playing with them 🙂

There are no original hand moulds but we used a reproduction

Set Patterns

So, roman type evolved since the renaissance. 15thC, 16thC, this is van der Keere, the proportions then suddently change in the 17thC, the baroque period. This is DTL Elsivier, van Dijke’s type, and this is Fleischman. The larger x height it typically dutch to make it more economical.

But then the system is broken.

The baskerville at the bottom is made on the 18 unit monotype system.

When you bought a set of matrices, you got a set of foudnry type to set the register of the mould, so you don’t have to set the spaces optically.

This is a totally difference approach to what I find from mreasuing.

Today you draw a letter, then find the space for it. I think the leters were drawn on a grid, so you didn’t have to design the spaces. So the serifs were determined by the grid. Today its arbitray, you design the lednght of serifs and adapt your spacing to them. That only started in the 18th C.

This is a belgian punch cutter [name?]. Updike isn’t positive about his work in ‘Printing Types’ but you can see he didnt look to stadnardise his matrics. he gave his casters set patterns.


So, what is type design?

Its applying systems.

These are systems of the 15thC. If you dont konw the original typees, you will always be infleucnd by them as they are so core to the forms we recognise. especially for text type.

You have a aharmoinc systema nd a proprotional sytem. the you have contrast and weight, 2 differnt things. then the rhythmic system, and the formalisation of pen shapes into type shapes.

Then finally the idiom, your personal choices. That is the difficult part. You can reconize Zapf or Frutiger’s type. But if you can define what you see? No, what makes Zapfs letters different to Ungers? To define this you must be able to define the pattersn underneath in detail, so you can see what the type designer has done.

All that adds up to type design.


When you can do this, then you can put it into the computer. So in 10 years, we’ll have a a program to make a type in the style of Unger 20%, 50% Zapf, and 30% your own. You go out and have a beer and come back and tweak it a bit.

Then you say type design is so difficult so you can sell your fonts and amke a lot of money.

But if you can define it, you can make a program of it. And if you can’t define it, it doesn’t exist.


Q: You make something from a system, you tweak it… can you say the same of human physionomy? Someone must have had the idea of a grid. You can tweak them across races, but there are these proportions. How intentional are these systems? Did you find clear indication in text?

A: I am full of doubts. In science nothing is true until its proven. Well, there is nothing written on this as first hand evidence. All ideas have no proof on this topic. I think I have more proof, I have physical proof in casting applying the measurements. I have doubts, yes, do I think its likely? More and more. If I look at OpenType and PostScript, I talked to Adobe about how much their technicians were working with artist, and it was a close collaboration. Same at URW. Why would that be different in the 15thC? If you look at the step from Bottico (?) to Jenson, I think this helps explains the gap in the step. I am open to a better theory, like a scientist.


This entry was posted in Knowledge. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.

Post a Comment

Your email is never published nor shared. Required fields are marked *

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    Markdown is turned off in code blocks:
     [This is not a link](

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>


  • What we do

    Understanding Fonts is a type design training business. If you'd like an event in your college or city, let Dave know: