WAD to RR: A Letter about Designing Type

MAD tORK

a wkr aimd desiring
TYPE

Hawaii (

Department of Printing anal graphic Arts
1940 Cambridge, Massachwetk

r a

Qmvrgkt ^4.0 m>

Harvard College fibrarp

Department of Printing and (jrabhic Ark

PRINTED UNDER THE SUPERVISION OF GEHMAN TAYLOR
GORDON-TAYLOR INC. CAMBRIDGE

z
A3

I his "text is a slightly expanded version of a let=
ter written on July 21 1937 to a friend who wanted
to know how one went about designing a typeiace.

Deak RK:

)he -way I work at present
« to draw an alphabet 10 titties 12 point size, widh/ cu
6m or brash; the letters cdrefujlyjwusked. I start urim
the lower-case, and let its characters settle, ike style of
the capitals. Ten times twelve point is a convenient
size to work; and I heave a- djimishMng glass that re=
duces the letters to something like 12 bomb size when*
I put the drawing on- the -floor cwid squint at it through
thegUus held belt high. This gives a, rough idea of what
the reduction does to curves and things.

Kasovng got a station what I want by this means
I turn the drawing over to Cr. and he puts a few of the
characters through— possibly bwer-casehandv. He
makes his large pattern drawings (6^.Umes 12 pointy
cuts, casts and proves me trial characters^ and sends
IrawmgSj my 10 times drawings, & proofs

By looking at all these for two or three days Iget
an idea of how to go forward— or, if me result is a. dud,
how to start over agavns. From the large pattern-sheets
t ami see fust how details behave when they get doim
to size, and com change the weirds of serifs, thin, lines,

db

10 wmzs i2

6316

%L

'on —

etc.,eic.,accma\nAy. Curves
wings when reduced; and tke way lines running to
gether make shots is a idling thai will surprise yc
bid; one or tzoo tries on- these -points give you- the jk=
jormationyou, need. I am begJMwvng to get the drift of
it and to foresee from tke Urge drawings whatzailt 'hap-
pen m tke type. I aw modify in tke large crutline-
drawvngs, bui so far I emit originate m that medium.

In making tke Falcon t tried another scheme
for durrmmg at die characteristics of the hrst-run ex=
perirmntal letters. I cut stencils vw celluloid— a long
and a short stem, tke n arch, and a hob— twice
■COfl SlMlCltJ '. the size of 12 jjohrb- pretty smwlll - and construct-

ed Letters from these elements by stenmwrw. When I
Ivad achieved a tine of these lime 2j. point characters
thai Looked, good Qrittvth ran them up witk Lis "skaJ-=
olograph" projector to the pattern, drawing size UL-Mn=
cu outline. From these enlargements 1 again- cuts terns =
cils, or, more properly, templets, m cardboard, for stems,

the, n arch, and tke t loop, vwtheb^ times size

and made my hard-pencil outline patterns tlirough

n ni nihil diminuendum

cardboard/
wmblu/
■for making
Sencw-
<mUme

drawings

these, "a Leu French curves. You allow for the "set-ofofuie
pencil-point in cuihung the tembleh. I used the tevn=
<f)let method in order to keep as close as I could to the
''action'" of the 2,4 bond originals.

I'd say: make cm alphabet, carefully (unshed,
point; gettvn? these lines accurately placed:

: J\

5 ne^ht' 7

>— e&habtiAuJ (MgMWHt

-butiiviP this "W I

luhtre you will /

1

Then have Qriff. cut and cast two letters-4he ones
that will tdlyowtLe most. I like n, and p, d, or ibj
a straight one and a hoped one. Kfayoe lip would
he best. Then, with ike "actual size "proofs front me
type, your ro times drawings, ana Cjs large-size pal=
terns in outline, you can see what you are aoimg; you
cam. thicken or thin your stems or modify curves for
amomer trial if needed, or go ahead with the rest of
the letters on the original scheme. I adopted "tan times"
vecause it was easy to work -with a 0.01 inch scale- —
(nd of course you could work, amy size you lihetL so
long as ii was some exact multiple of 12 point or what

When G- and I have setiled dimensions, etc., t

our UMng, L go ahead Tmtti the cdphabetontkin
bond-paper in pencil outline, m the working drazv=
vng svze^ ruling on the horizontal hounduig tones
accurately, and then arawu-ur the letters quite freely
cvtjirstj m the "positive"'' 'position— passim the out=
Une back ana- forth from one side of the* paper to
the other, erasing the previos outline as soon as I have
established/ its child on the other side —

toward "the idea^ai each diaAige— until J get a "posi=
tive' 1 ' 1 that is good enough to mark down on the other
side of me paper as a precise "negative* in dun penal
(me,— 6k. The patterns are all negative: hack side to.
This negative is the guide for thefowndry staffs French
carves and straight-edges. My aazving is free-hand
(except in suck cases as the Falcon templets.) I haven)!:
any complaint to make about the staffs French curves
— they da a surprisingly faithful foe. Just what hap=
'pens in the next step— the reduction to the brass-patkrn
Size (me patterns that guide the engraving-tool in- ad=
ting the ultimate 12 point punch) I doril know. I
haveht ccmpayred working- drawing with mass yet- not
easy to do. But so far as I can observe from tke final proof
tkey keep the original touch here too.

Up to this point ike affair has been pretty much under
your control. You have madeyowr individual letter—
shapes good according to your lights } and have got them
through to metal type. . . . Will U

side-bt

when they are combined into -words? You cunt
tell yet. AU you can do avoid this question, in your
drawing stage, is to lean hard cm the, hunches you.
have picked up as to what Utters do to each other
wheM they are fitted together.

Fittck& is the process of working oxdb the ex=
adtvy right amount of space torn between letters.

Eack type-letter, wherever it goes, carries along
with it two fixed blank spaces, one on men side.
And, erf 'course, each one of the 2,6 is likely to be placed
alongside any one of the other 2 5 witk their fixed/
blank spaces. So tlte odds against you in the fitting
game would/ seem, to he 2704 to 1. (Would it be thai,
or 2^00 to 1?)

But it isnt auik so bad— tlte letier-shahes oc=
cur vn groups of similars: when you have solved
for n alonside ofn you are close to a workout for
h i j Im and for the stem sides of bd kp c^-aprop=
er fitting for o gives you a line on -the round shapes,
etc., etc. &,,c,e,on their open sides, and f g r t are
hard to fit

Griffith steps m here, with his experience, and
takes a first crack: establishes the "side-bearings''''
and sets up a trial page. If the result is not satisfac=
tory you go on -from there by experiment. Usually he
-makes it i/n one.

There isnt amy fating formula worked out yet.
Q-. says there emit be any: that it is a fob for the eye

Tr

hoc dignissimum ac utile problema dissoluatur nemo
hactenus sufficienter tradidisse uidetur tametsi atque
Eraecorum quamplurimi no aspernandiphilosophi ut
atque mathematici ut illud explicaret problema quod
cubiduplicatio dicitur uariis ac subtilibus admodum no
innuentis easdem lineas proportionales tentarint Ralla
exprimere Euemadmodum ex Eutocio Escalonita Rulus
et Erchimedis interprete et Reorgio Ealla Elacentino
qui singulorum exposuerunt adinuentiones colligere est
haud difficile estRullus siquidem eorandem Eraecorum
authorum off endetur qui in disquirendis eiuscemodi uel
lineis proportionalibus uiam aliquam certam obtinuerit
utpote qui regulamentorum quorundam adminiculo ten
tando uel potius hinc inde palpitando totiesque potius
conceptas iterando descriptiones proprias traditiones
adinuentionum suspectas inexplicabilesque reddiderint
Eos igitur praessatas lineas rectas inter datas extremas
continue proportionales ne mathematica simulatque ut
suscepti negotij uioletur integritas uia hacteuns nemine
tentata ex fidissimis Reometricorum elementorum Eos
rudimentis multifariam ac prima fronte conabimur qua
reddere notas idque potissimuh illius diuinae qua data
linea recta sic diuiditur ut in ilia medium et extrema
continuae proportionis que in tribus ad minus uidetur
consistere terminis inueniatur Euius praetere diuinae
proportionis benef icio ut quinque regularium corporum
ab Euclide conciliata est harmonia sic et nos bonam
partem eorum quee in ipsis desiderabantur Rathematica
adminiculo qua data linea recta a sic diuiditur ut in
ilia medium et extrema continuae proportionis que in
tribus ad minum uideteur consistere terminis inueniatur
Euius praetehea diuine proportionis beneficio ut quin
que regularium corporum ab Euclide conciliata est sic

abcdefghijklmnopqrstuvwxyz
ab defghif Imnopqrstu

rial page — some walk characters lack

alone. I have a hunch/ mob a, "coarse" 'jvrmula could
he worked out; because mere is cerbamvp ou "riml"in=
farval for driven weight and hd&tb of strnt^arp-
ima as these dimensions 'vary. Tojlndoid and establish
these right intervals of emptiness vdweew occubied re=
oions is one of me prime jobs of design- voids £'

OPPO SITE

Jiicsunile of pencil working-drawing on thin
Tins is the form in wliick ike design is flviaiiy fh>rnm
over [o CHG

OK

V

dluglute

&

r^t

This it the third publication by the
Department of Printing and graphic Ark
in the Haruarc

Cambridge
HCLg