collective writing/typeface project

abi's picture

A little background: a friend (Easton West) and I were sitting around on a summer day last year and we were discussing representing a 'community' through a typeface as well as the impetus or reasons to embark on making a typeface (historical reference or perceived problem). The idea came to mind to try collecting writing samples and making an 'average' of them in order to create a sort of average conception of the alphabet, and afterwards to use this as a basis/skeleton for a typeface.

Here is the current incarnation:

Even though I had assumed the end result would be fairly clear, what was most surprising thing for me was that a fairly legible sans-serif emerged from the cacophony of individual voices. The question now is whether it is worth it to pursue the project further and begin to refine a skeleton to eventually base a typeface upon.

Any thoughts?

hrant's picture

Interesting, and not unattractive.


Reed Reibstein's picture

I've thought that something like this would be a really cool experiment -- good luck! I'll be interested in seeing what further refinements produce.

hrant's picture

BTW, you might like to check out Vlad Atanasiu's fascinating -and
incomparably rigorous- work on allographs in Arabic calligraphy.


Quincunx's picture

Very, very interesting.
I think it is defenitely worth it to pursue the project further.
If you do, make sure to keep us posted via this thread!

Florian Hardwig's picture

Hi Abi,
sounds interesting indeed! Nice idea. It reminds me of the famous Frutiger picture with the layered letterforms, ‘revealing the quintessence: Univers’ ;-)

samples were collected under controlled conditions.
I’m quite curious what these conditions were … Do you mind sharing?

representing a ‘community’ through a typeface
As far as I can tell from your samples shown, this community is pretty homogeneous, no? Homogeneous, i.e. American. If your experiment would include some German writers for instance, I’d bet there were descenders with uc J, no (or lesser) bars with uc I, etc.pp.

average conception of the alphabet
Yes, it is. But where does it come from?
I would say that it’s the forms we learn and internalize at school, ± individual wearing to some extent. This picture shows 3 layered fonts, all made for the use on primer worksheets for teaching handwriting—US style (bottom); in comparison to one of your samples (top).

(Schoolhousefonts’ DMOA Print (Zaner-Bloser style) & DMOB Print (D'Nealian style), Jess Latham’s Print Clearly. Adjusted size and kerning, 60% opacity)

abi's picture


Thanks for the Vlad Atanasiu reference, I'll be sure to look into it.


As for the sample collection methodology there were a few basic 'controls' to the experiment, (but really calling them controls is a bit of a misnomer) everyone was asked to sit down with the same kind of paper, with the amount of space given to everyone else and using the same pencil. The use of pencil (fairly soft lead) allowed for us to gauge pressure and this was factored into the end result as well (people that press hard when writing made darker samples and thereby represented more in the final result). In the preliminary tests I had implemented more controls and instead of a blank space they were asked to put each letter in an individual gridded unit, but we dropped this because it seemed to provide a subconscious limit on the amount of space you could use, I liked the larger blank space because listing out the alphabet is a fairly automatic thing so therefore people could get into their own rhythm easier. If you have any suggestions as to the refinement of this methodology that would be great.

I think the question of where the 'average' conception comes from is worth a detailed look. The current incarnation was gathered from a decidedly diverse group of people, we had tried to get as much range in how people write as possible. Upon retrospect now it would have been interesting to do this with a class of grade 1 students, and compare the overlaid result with the reference material (Perhaps that is a future possibility). I also heard about this student in Germany who was doing a thesis project collecting the the 'primer worksheet' fonts from around the world, I should check in on how she is doing on it.

Florian Hardwig's picture

Abi, thanks for the explanation of your methodology.

listing out the alphabet is a fairly automatic thing
Yes, and of course this is a legitimate approach. But then, it ain’t the way we usually write text, is it? With this method of isolated letters, one after another (and, at that, the ‘probands’ concious about being examined), you’re likely to get more ‘idealized’, conventionalized characters—and I don’t even think about ligatures or ‘joined script’.

If (and I say if, as there is absolutely no need to do so) you want to collect the ‘plain normal handwriting’ (There is no such thing, btw. Or thousands, for every single person, depending on oh-so-many circumstances), there is no alternative to collect random handwriting samples that were not produced for this purpose.

the question of where the ‘average’ conception comes from is worth a detailed look.
Well, if you are taught that the ‘I’ has bars, you’ll write it like that. As a child, and later on. Certainly some people will personalize/emancipate and drop (or exaggerate) the bars—but the school model remains the norm, concerning a certain region and a period of time.

I also heard about this student in Germany who was doing a thesis project collecting the ‘primer worksheet’ fonts from around the world
Are you talking about Linda? Ha, small world of typography! ;-) I did a similar project, we exchanged some stuff. If you’ve a deeper interest in handwriting styles used in primers, feel free to contact me.

abi's picture

haha, yes I am actually referring to Linda, I've never met her but my roommate knows her well, they met while on exchange to KABK. Small typography world indeed.

I gathered the samples in that way in order to look at the conception of the alphabet, a conscious writing. The more idealized versions of the characters and the tiny variations of that ideal are what I sought in gathering the samples in the isolated manner.

hrant's picture

Another thread just gave me an idea:
Could you recompile separate results for men and women?


Florian Hardwig's picture

@hrant: Which thread?
Yes, there are some formal elements women tend to use, and some you’d more frequently find examining male handwriting. Yet, any absolute statement about that would be penned by graphology, wouldn’t it?
Or, with the words of Sten Nadolny: »Aus Geschriebenem siehst du ja nicht einmal, wer spricht, Mann oder Frau!« [“From script you can’t even tell who is speaking, man or woman!”]

Nevertheless, closer inspection of that would be intriguing, too.
And while doing so, don’t forget the lefthanders!

hrant's picture

> any absolute statement about that would be penned by graphology

Any absolute statement should be penned as fascism.
But interesting non-absolute statements can only help.

> don’t forget the lefthanders!

Good idea!


Syndicate content Syndicate content