The Typographic Inception of The Cherokee Syllabary
The Typographic Inception of The Cherokee Syllabary
The Typographic Inception of The Cherokee Syllabary
The adaptation of most writing systems to print has generally been one
of mimesis: typography attempting to replicate the visual appearance of
a specific handwritten form of a given writing system. The typography of
Latin-script incunabula emulated the Carolingian hand with roman, and
the Italian scribal style with italic; Greek typography—from Aldus Ma-
nutius on—attempted to replicate the complex Byzantine hand; Chinese
and Japanese typography were initially based on their long calligraphic
tradition; and so on...
Cherokee typography did not follow that process. In fact, the Chero-
kee syllabary itself—unlike most writing systems—did not slowly evolve
in written form to then eventually be adapted to print. It was invented
single-handedly by a Cherokee man named Sequoyah at the beginning
of the nineteenth century. In a reversal of the usual process, Sequoyah ex-
plicitly developed the syllabary in a shape which would make it ‘suitable
for print’, and deliberately took inspiration from Latin typography for the
design of its characters.
The actual translation of his syllabary into print further involved typog-
raphy in a complex back and forth process involving a number of people’s
perceptions of typographic shapes and technical constraints to establish
its definitive forms.
a e i
ga ka ge gi
ha he hi
la le li
ma me mi
hna
na ne ni
nah
qua que qui
s a se si
da ga de te di ti
o u v
go gu gv
ho hu hv
lo lu lv
mo mu
no nu nv
quo quu quv
so su sv
do du dv
tlo tlu tlv
tso tsu tsv
wo wu wv
yo yu yv
Figure 1 The Cherokee syllabary, with the common phonetic transcription of each character.
5
Figure 3 Undated drawing of the syllabary in both forms, attributed to Sequoyah.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
However, Sequoyah did not settle for these cursive characters for his sylla- Figure 9 The Hicks Syllabary, was
‘enclosed in a latter from Charles
bary. In a process which will be detailed later, Sequoyah decided to adapt Renatus Hicks, second principal chief
his syllabary into ‘characters for print’, radically modifying in the process of the Cherokee Nation, to Thomas L.
McKenney, head of the Office of the
most of the cursive characters to give them the shapes which became the Indian Affairs in the War Department
model for the syllabary as it exists today. on 14 January 1825.‘ Willard, W. and
Sarbaugh, The Early History of the
In 1821, Sequoyah presented the syllabary to the Cherokee Council; Cherokee Syllabary, p.85
it is unclear whether he presented the cursive characters or the characters
for print, or both. Of the two forms of his syllabary, he is reported to have
explained that the new ones ‘would do for print and the old one [cursive]
for writing.’10
After an initial period of rejection—some sources even report that he
was accused of sorcery11—the syllabary began to generate massive inter-
est, and within a matter of a few months the majority of the Cherokee
population knew how to read and write in Sequoyah’s syllabic characters.
The numeral system however never was adopted.
If the cursive syllabic characters were ever used, no document written in
that script, apart from Sequoyah’s chart, seems to have survived. By 1825— 10. Bass. Talking Leaves, no pagination
11. Perdue. Cherokee Editor, p.69
the date of the earliest surviving sample of the syllabary [Figure 8]12—the 12. Walker, W. & Sarbaugh. The Early
Cherokee language was being written in Sequoyah’s ‘characters for print’. History of the Cherokee Syllabary, p.85
7
A B C D
— — — —
Figure 10 Cursive characters and their equivalent characters for print, extracted from figure 3. The new
characters in colums A remain essentially cursive; those in B-D take new and more typographic shapes.
Sequoyah took his inspiration from the Latin alphabet to adapt his sylla-
bary into characters for print. John Howard Payne, a US Consular official,
reporting on the invention of the syllabary related that Sequoyah was ‘struck
with the Bible Book [and studied it] for characters to make use of in print.
He copied out some of the [Latin] letters...’.13 Various other sources, more
or less plausibly, cite other inspirations: an English spelling-book,14 an un-
specified old English book,15 the McGuffey Reader, and a Greek text.16
When looking at Sequoyah’s drawings, we can observe that the new
characters he created are not uniformly modelled on Latin typographic
forms. They display varying degrees of similarity with Latin characters,
and varying degrees of adaptation for print. Some characters remain cur-
sive; some are directly modelled on Latin typographic characters; and
some are hybrids of both vocabularies, with the majority of them leaning
more toward typographic structures.
In Figure 10, the characters in column A are direct adaptations or sim- 13. Bass. Talking Leaves, (no pagination)
plifications of their shapes from the old syllabary and thus remain essen- 14. Anonymous in The Missionary
Herald, Volume 22, Number 2, February
tially cursive, even as characters for print. 1826, p.48
No mention is made of Sequoyah having consulted handwritten docu- 15. Davis. The Life and Work of
ments in the design of either the cursive syllabary or the new one. But as Sequoyah, p.160
16. Ehle. Trail of Tears, p.153. This source
mentioned previously, his cursive characters display formal characteris- does not provide any reference for
tics not unlike the contemporary handwriting style of the time, and may making such a precise affirmation.
Additionally, while at least one Greek
have been inspired by it. [Figures 6, 7 and 8] document was present in the Cherokee
However the characters of the cursive syllabary are cohesive as a writ- cummunity at the time—a Cherokee
man called David Brown having
ing system, and not similar enough to the Latin cursive letters to make translated in 1825 his manuscript
this supposition plausible on its own; they might in fact simply be the Cherokee New Testament directly from
an original Greek text (Bass. Talking
result of the characters of both scripts being written with the same tool. Leaves, no pagination and Davis, The
But one character in the new syllabary makes it probable that Sequoyah Life and Work of Sequoyah, p.166)—it
seems unlikely that Sequoyah would
at least knew of Latin cursive handwriting: the new shape of Ꮰ displays a have had access to it and in the
form reminiscent of a common ampersand, but rotated counterclockwise event that he did, it does not seem to
have had a significant impact on the
[Figure 11]. It is indeed an ampersand—in a form which was never trans- design of his syllabary, since very few
lated into type17 but was very common in handwriting in Sequoyah’s time, characters can be interpreted as being
specifically inspired from the Greek
and was notably used by most of the missionaries attached to the Ameri- alphabet.
can Board of Comissionners for Foreign Missions (ABCFM) living in the 17. The eminent type designer
Cherokee community. [Figures 12 and 13] Matthew Carter was first in noting
this particular shape of ampersand
The other characters, displayed in columns B to D, depart in their new on eighteenth century tombstones in
shapes from their old cursive counterparts. As we progress through the New England. He confirmed to me that
it never occured in print.
columns, the characters become increasingly similar in structure to latin 18. Another character, Ꭶ is intriguingly
roman characters, and the last characters appear to be taken directly from whimsical. It appears to be based on
musical notation; in the old syllabary
the typographic shapes of letters of the Latin alphabet.18 But Sequoyah’s it take the shape of a treble clef, and
adaptation of Latin shapes to his syllabary was not a simple exercise in in the new one, it is replaced by a
character very similar to a dal segno.
mimicry; his interpretation of the Latin characters is a very interesting He might have seen these characters
and idiosyncratic one. either in written or printed form.
Taken individually, one could
Since he was not familiar with the Latin alphabet, he had no notion consider the similarities between each
of the ‘generic shapes’ of the characters: he did not perceive the printed of these characters and Latin musical
notation a coincidence, but their use
characters as stylized renderings of the fundamental structure of Latin for the same syllable seems to validate
letters according to a particular typographic style. He interpreted the the parallel.
9
Figure 20 Romanized Cherokee developed by John Pickering.
typographic aspects of the letters in their printed forms—such as serifs and
Figure 14
ball terminals—as structural parts of the characters rather than variable sty-
listic elements.
Consequently, Sequoyah used these typographic features as tools for
creating his characters—serifs and ball terminals went from being merely Figure 15
typographic features to become inherent parts of a character’s structure. In
some cases, these typographic features even became the sole structural ele-
ments used to differentiate otherwise similar characters:
Figure 16
– The characters Ꭱ and Ꮢ, both based on the Latin R, differ only in the
elaboration of the tip of their leg.[Figure 14]
– The characters Ꭹ and Ꮍ are both represented by a shape similar to the
Latin lowercase y, but are distinguished from each other solely by their Figure 17
different use of serifs and ball terminal.[Figure 15]
– Only the top serif differentiates Ꮒ and Ᏺ.[Figure 16]
– Ꮃ and Ꮤ vary slightly in structure [Figure 17], but to a user of the Latin
script look like two design variations of the same Latin character. Figure 18
During the early 1820s, the same period in which Sequoyah was developing
his syllabary, another writing system for the Cherokee language was being
developed in Boston by the linguist John Pickering.
An alphabetic system adapted from the Latin alphabet [Figure 20],
it consisted of the Latin letters a,d,e,g,h,i,k,l,m,n,o,s,t,u,w,y—used in
upper and lowercase forms, with italics—to which set were added three
new characters created specifically for this Cherokee writing system: Figure 21
11
To this basic repertoire, four marks—identical in shape to the Latin apos-
trophe, cedilla, dieresis and breve—were used as diacritics to modify the
phonetic value of the characters;19 [ ’ ] to mark a glottal stop, [ ¸ ] for nasali-
sation, [ ¨ ] to disolve a diphthong, [ ˘ ] to note shortened vowels. Another
diacritic, similar to the Latin acute, was used above or following a vowel
to denote emphasis on a syllable. [Figure 24]
Figure 24 Three accents of the accents in use. A grammar of the Cherokee language.
While this writing system was favoured by the ABCFM, whose mis-
sionaries were active in developing a literacy program in the Cherokee
community to disseminate their religious beliefs, it was met with little
interest from the Cherokee themselves. Conversely, Sequoyah’s syllabary
was well-suited to the Cherokee language, and since its presentation to
the Council in 1821 was rapidly being taken up by a sizeable portion of
the Cherokee community. Furthermore, having been developed by a
Cherokee man it had become a subject of national pride for the Cherokee
people. In only a few years, it became the de facto embodiment of the
Cherokee language.
As Samuel Worcester, a missionary of the ABCFM who arrived in the
Cherokee nation in September 1825 and who was to play an important role
in the translation of the Cherokee syllabary to print, reported to Rufus An-
derson, assistant secretary of the Board, in a letter dated 27 March, 1826:
‘If books are printed in Guess’s characters, they will be read; if in any other,
they will lie useless. [...] Whether or not the impression of the Cherokees
is correct, in regard to the superiority of their own alphabet for their own
use, that impression they have, and it is not easy to be eradicated.[...] At
their national council, [the Cherokee people] have listened to a proposal to
substitute an alphabet like that of Mr. Pickering, and have rejected it.’ 20
As Sequoyah’s syllabary quickly came into general use, the Cherokee Coun-
cil adopted it as the Nation’s official script in 1825.21 Pickering’s Cherokee
alphabet remained an object of consideration and discussion only among
the members of the ABCFM, and as Worcester became better acquainted
with the Cherokee language, in his correspondence with the Board he
came to be an ardent supporter of Sequoyah’s syllabary and increasingly
critical of Pickering’s alphabet and the ABCFM’s support for it, noting in
19. Pickering. A Grammar of the
particular that he ‘found no use of Mr. Pickering’s vowel ....’.22 Ω
Cherokee Language, p.11-15
Pickering’s alphabetic system never found more than marginal use— 20. ABCFM 18.3.1 v.5, 230
21. Davis. The Life and Work of
essentially by English speakers unfamiliar with the Cherokee language— Sequoyah, p.166
and subsequently fell out of use entirely. 22. ABCFM 18.3.1. v.5, 234
13
Figure 25 (left) Syllabary drawn by Sequoyah for John Howard Payne in 1839.
Figure 26 (above) Syllabary reproduced by Samuel Worcester, in a letter dated 22 December 1825
Figure 27 (below) Model for the punchcutter by Worcester in his letter of 2 September 1826.
WORCESTER’S MODEL FOR THE FONT
‘Thus I think there will be no occasion for matrices for 16 of the characters, 23. Davis. The Life and Work of
viz R D W G P M B A Z E T J K S H L, as the small capitals of the English Sequoyah, p.155
24. Anonymous in The Missionary
fount will answer every purpose.’ Herald, vol. 23, no. 12, p.382
25. Davis. The Life and Work of
Sequoyah, p.165
With these instructions Worcester made design decisions which depart 26. Bass. Cherokee Messenger, p.126
from Sequoyah’s design. He appears careful to try to respect Sequoyah’s 27. ABCFM 18.3.1. v.5, 232
28. Bass. Cherokee Messenger, p.78
drawings, adding in the same letter: ‘I would not have the figure 4 substi- 29. During its short and troubled life,
tuted for the character Ꮞ [...] nor would it be well to use an inverted V for from 1824 to 1886, the foundry which
is now referred to as the New England
Ꮩ, but rather to have a distinct type, as u and n in English’. Type Foundry changed management
However, in Sequoyah’s design of his characters for print several no less than eight times. From its
inception to circa 1829, it was headed
character shapes are consistent and inter-related with other characters. by Baker and Greele. Annenberg. Type
It seems that several of these graphic consistencies were lost on Worcester, Foundries of America, p.203
15
Figure 28
Seq
uoy
ah [
f rom
figu
Seq re 3]
uoy
ah [
f rom
figu
re25
]
Hick
s [fro
mfi
gur
e 9]
Wo
rces
te r [fr
om
figu
re 26]
Wo
rces
ter’
[fro sm
ode
mfi
gur l
e 27
]
but also for the most part on other users of the script including Hicks,
and as a result many of the formal consistencies between the charac-
ters—especially those totally invented characters which bore no formal
reference to Latin letters—as they were initially intended by Sequoyah
were subsequently lost.
This can be observed by comparing, in Figures 28 to 37, Sequoyah’s
drawings of the syllabary [extracted from figures 3 and 25] with those of
another Cherokee man litterate in the syllabary, such as Hicks [extracted
from Figure 9] and then Worcester’s handwriting [from figure 26] and model
for print [from figure 27].
Figure 28 displays those sixteen characters which Worcester perceived
to have been directly inspired by the Latin capital letters and for which he
consequently instructed the punchcutter to use Latin small caps matri-
ces. Eleven of these characters appear, in Sequoyah’s drawing, very close
indeed to their latin counterparts. Of the remaining five characters,
three—Ꮐ, Ꮇ, Ꭺ—are not exactly modeled on the Latin characters:
– Sequoyah’s shape for Ꮪ is not unlike a Latin S, but is drawn rotated 90°.
Worcester substituted an upright Latin S in its stead.
– Ꮓ is drawn by both Sequoyah and Hicks with a clearly upright stem
terminated by two horizontal segment which extend on both sides of
the stem; but Worcester significantly interpreted the character (as will
be further detailed later), drawing it with a slanted stem and short-
ened horizontal strokes joining it at its ends—essentially making it
into a Latin letter Z.
17
Figure 30
Figure 31
Figure 29
S eq
uoy
ah,
und
ate
dd
raw
S eq
uoy in g[
ah, fig.
dra 3]
win
g, 1
8 39 [
fig.
25]
Hick
s, d
raw
ing
,1 825
[fig
Wo . 9]
rce ster
, ha
ndw
ritin
g, 1
8 25 [
fig.
26]
Wo
rce ster
182 ’s m
6 [fi ode
g. 2 l fo
7] r th
ep unc
hes
,
For example, at the end of some strokes, Sequoyah put small triangu-
lar elements which he clearly modeled on Latin beaks. Such triangular
elements in Sequoyah’s drawings are consistently translated [Figure 29] as
beaks in Worcester’s model for print. In the character Ꮖ, this translation
is carried to the triangular shapes but also to the two opposed serif-like
straight segments, transforming Sequoyah’s intentionally asymmetrical
character into a symmetrical one with four beaks.
Worcester’s deliberate restraining of Sequoyah’s vertical serif-like
strokes—those which extend equally on both sides of the horizontal stroke
to which they are attached—is clear in the characters in Figure 30. Their
translation to typographic shapes is inconsistent: in some cases they are
rendered as beak-like structures, like the triangular elements in Figure 29
from which they differ: in the remaining characters, they are rendered
as a short segment extending only on the inner side of the character. It is
even more significant in Ꮓ: Worcester does again reduce the vertical serifs
to beaks, but completely changes the character’s shape by rendering it
as a Z despite its basic structure being clearly related to Ꮴ and Ꮖ in both
Sequoyah’s and Hicks’ renderings.
Several of the characters invented by Sequoyah shared other common
visual elements. However several of these common features were inter-
preted differently from one character to another—particularly in the case
of the more cursive characters detailed in Figures 31 to 33—in their prepa-
ration for print in Worcester’s model for the punchcutter.
In Figure 31 all the characters have an elliptical element of similar pro-
portion in Sequoyah’s drawing, particularly the five first five characters.
However, in Worcester’s model they are given different widths and stroke
modulations, making them look significantly less related. In each charac-
ter, some elements also differ from Sequoyah’s model:
– The sinuous bar diagonally crossing the counter of Ꮻ has been made
more angular and horizontal in Worcester’s interpretation.
– The small diacritic-like sharp hook on the right side of the Ꮕ in Worces-
ter’s rendering is far removed from Sequoyah’s more sinuous stroke
which extended almost to the full height of the character and as wide
as the elliptical part of the character.
– It can also be noted that the significant reduction of the size of the oval
in Ꭳ echoes Hicks and Worcester’s drawings more than Sequoyah’s, but
not significantly so.
19
Figure 36
S eq
uoy
ah,
und
ate
dd
S eq raw
uoy ing
ah, [fig
dra . 3]
win
g, 1
8 39 [
fig.
25]
Hick
s, d
raw
in g, 1
825 [
Wo fig.
rce 9]
ster
, ha
ndw
ritin
g, 1
8 25 [
fig.
26]
Wo
rce ster
182 ’s m
6 [fi ode
g. 2 l fo
7] r th
ep unc
hes
,
Figure 33
Figure 34
Figure 35
Figure 32
S eq
uoy
ah,
und
ate
dd
raw
S eq
uoy in g[
ah, fig.
dra 3]
win
g, 1
8 39 [
fig.
25]
Hick
s,dra
win
g, 1
825
[fig
Wo . 9]
rces
ter,
han
dw
ritin
g, 1
82 5 [fi
g. 2
6 ]
Wo
rce ster
182 ’s m
6 [fi ode
g. 2 l fo
7 ] r th
e pu
nch
es,
raised to the character’s mid-height. Furthermore, the first six characters
possess a similar cursive diagonal angle in Sequoyah’s drawing, but have
their bottom part brought down to the baseline in Hicks’ and Worcester’s
drawings. In the latter’s model for print, the similarities are further re-
duced; the first three characters are still related to each other, and so is the
other group of three, but the formal similarities between the two groups is
almost completely gone: the first group being modulated on a vertical axis,
the second on a hybrid, but mostly horizontal, one.
The closed loops of Ꮼ and Ꭷ in Figure 33 are similar to each other and
also have similar proportions to those found in the characters in Figure 32.
In Worcester’s model, Ꮼ is defaced and illegible, but we can observe that
at least Ꭷ no longer shares similarities with the characters of Figure 32.
The character is rendered in a shape more similar to the cursive shape of
the Latin letter e, as it is also presented in Hicks’ drawing. Moreover, in
Sequoyah’s design, the loop extends up to the standard characters’ height
and the upward curved stroke extends above it. In Worcester’s rendering
the upward stroke of Ꭷ only extends up to the other characters’ height, so
the loop is scaled down accordingly.
The characters in Figure 34 are not related in Sequoyah’s design of the
syllabary, but they are made similar in Worcester’s model for the punch-
cutter. In Sequoyah’s design, Ꮥ shares formal characteristics with charac-
ters such as Ꮳ, Ꮈ and Ꮔ: it has a lunate-shaped stroke terminated on top by
a serif-like segment and under which is adjunct a cedilla-like curved small
stroke. In Hicks and Worcester’s interpretations, however it takes the
shape of an s-like character, to which is adjunct a horizontal segment at
its mid-height. In the model for the punchcutter, it now closely relates in
shape to the characters Ꮪ and Ꭶ which have both been rotated 90° counter-
clockwise to resemble more the Latin s—the matrix of which was to be
used for Ꮪ, as seen previously.
In Figure 35, the characters in the model appear to be based on Worces-
ter’s interpretation and differ significantly from Sequoyah’s characters.
Worcester also altered the vertical span of some of the characters. All the
characters in Figure 36 are based on a central stroke similar to the Latin I.
In Sequoyah’s model, Ꭾ has a curved stroke extending from the center of
its stem to ascend above the characters’ height. In Worcester’s model, the
whole character shape is shrunk down to make the apex of the curved
stroke only reach up to the characters’ regular height.
Figure 37
As already noted in Figure 33, this scaling down of a character was also
applied to Ꭷ. Two more characters, in Worcester’s model, are altered to
fit the standard vertical alignment: Ꮏ is not scaled down but has its top
and bottom segments truncated; the same procedure is applied to the
descending stroke of Ᏸ. [Figure 37]
21
ALTERATIONS TO THE MODEL
When Worcester received proofs of the font in the late spring or early
summer of 1827, the political landscape of the Cherokee Nation had
changed: Charles Hicks had died of illness in early 1827,30 and was
succeded by John Ross in the position of Principal Chief of the Chero-
kee Nation.
Ross, along with George Lowry (referred to by Worcester and some
other sources as Major Lowry, who would become in 1828 31 Assistant
Principal Chief), subsequently became involved in the process of the
creation of the Cherokee font. Not being literate in the syllabary,32 Ross’
comments remained general. Maj. Lowry, however, took a very active
part in supervising the work of Worcester and took a role which would
now be considered art direction; Worcester explicitly expressed his
wish to have the font designed in such a way that it would meet their
satisfaction.33
In a letter dated 12 June 1827,34 Worcester relates Ross and Lowry’s
reaction to the proof of the characters cut according to his model and
instructions. Ross was displeased with the decision to have the font cast
on the small pica body size, and on a small caps height to harmonize
with Latin lowercase characters as Worcester had instructed with Hicks’
approval:
‘...I had [Ross’] opinion in writing that the characters ought to be a size
larger. [He] wishes to have the Cherokee characters larger than the Eng-
lish, because he thought the small pica large enough for perspicacity in
regard to some of the letters, and because, being made larger, when a
Cherokee word was printed in an English line, as might frequently be
done, the difference would more readily strike the eye.’
Lowry, for his part, severely critiqued some of the characters’ shapes.
Consequently, Worcester provided a new model for the characters (now
lost), redrawn according to Lowry’s directives, and specifically high-
lighted some of them:
‘The following are such as I think should by all means be altered, in con-
formity with [Lowry’s] wishes [...], Ꮄ Ꮈ Ꮝ Ꮍ Ꮗ Ꮢ Ᏸ Ꮙ Ꮧ Ꮥ Ꭾ Ꮿ
Ꮨ Ꮚ Ꮛ. [...] Respecting the character Ꮙ [...] the new is not suited to my
taste but Maj. Lowry was very particular respecting that one letter and
I made it perhaps a hundred times before I could suit him. Ꮢ should
have the space between the first and final strokes...’
23
Figure 38 Layout of the Cherokee cases.
Figure 40
Worcester also vaguely indicated that ‘punctuation marks’ were to be 36. Anonymous in The Missionary
provided in the same proportions as in the equivalent Latin small pica Herald, Volume 23, Number 12,
December 1827, p.382
font.42 As to the remaining twenty-nine units, fourteen punctuation and 37. Bass. Cherokee Messenger, p.81
non-syllabic characters—all with the same grammatical function as in the 38. Bass. Cherokee Messenger, p.80
39. ABCFM 18.3.1. v.7, 21(5)
Latin script—can be seen used in Cherokee text settings: period, comma, 40. ABCFM 18.3.1. v.7, 234
colon,43 semicolon, hyphen, double opening and double closing quotes, 41. Boudinott and Worcester. Cherokee
Hymns
exclamation mark, question mark, asterisk, left and right parenthesis, fist, 42. ABCFM 18.3.1. v.7, 234
and a long stroke (similar to the em dash but roughly twice the length of 43. Worcester mentions the colon not
being used in cherokee (ABCFM 18.3.1
the widest character). One additional glyph was no doubt the wordspace v.7, 234) but I saw it being used several
character, and the remaining fourteen boxes likely included, among oth- times in various cherokee texts.
It may have been borrowed from
ers, the five additional spaces (em-, en-, thick-, mid- and thin) tradition- the equivalent english font when
ally included in a font. A few of the boxes may have remained empty. needed.
25
Figure 43
Wo
rce ster
’s m
ode
l fo
r th
Che e pu
rok nch
e e fo es,1
nt 826
[fig
. 27
]
Figure 44
Figure 42 Small capitals of the Baker and Greele small pica fonts.
S eq
uoy
ah,
und
ate
dd
raw
S eq
uoy in g[
ah, fig.
dra 3]
win
g, 1
8 39 [
fig.
25]
Wo
rce ster
, ha
ndw
ritin
g, 1
8 25 [
fig.
26]
The font was cut and cast on a small pica body [Figure 41], as originally
instructed by Worcester and despite Ross’ objection, which seems to
have been subsequently abandoned. Also following Worcester’s initial
instructions, the characters R D W G P M B A Z E T J K S H L were taken
directly from the small capitals [Figure 42] of the only small pica font [Ap-
pendix 3] available from Baker & Greele. More precisely, three fonts in
the small pica body—Small Pica No 1, Small Pica No 2 and Small Pica No
3—can actually be found in the 1827 (and subsequent) catalogues44 of
Baker & Greele; but the three fonts differ in design only in the lowercase,
sharing the same capitals and small capitals.
Nine of the fifteen characters which Lowry specifically instructed to Figure 41 Detail of the cherokee
be recut according to his directives—Ꮝ Ꮗ Ꮢ Ᏸ Ꮧ Ꮥ Ꮿ Ꮨ Ꮚ—were characters, which were cast on small
cap height.
indeed modified and, interestingly, brought back closer to Sequoyah’s
models [Figures 43 and 44]:
27
Figure 45 First appearance of the Cherokee syllabary in print, in the Missionary Herald, in 1827.
Barely five years after its introduction in Cherokee society, the Cherokee
syllabary made its first appearance in print,45 in the Missionary Herald
of December 1827 [Figure 45]—though it was printed not from the types
themselves, but from a stereotype .
The Cherokee Press began operation in February 1828 with the pub-
lication of the first issue of the Cherokee Phoenix [Figure 46], which was
to be published in English and Cherokee weekly—with a few intermis-
sions—for seven years. It also marked the beginning of the prolific pub-
lication of religious books and pamphlets—as well as secular materials
such as legal documents and notices, hymn books, and almanacs—in the
Cherokee script.
While the Cherokee Press was thriving—with up to three presses run-
ning simultaneously,46 conditions in the Cherokee Nation were begin-
ning to deteriorate. In the early 1830s the state of Georgia was rapidly
expanding, leading Georgian settlers to begin to claim land in the Cher-
okee Nation; and crucially, gold was discovered in Cherokee territory.
The situation gradually degenerated and led to unrest. The Cherokee
Phoenix ceased to be published in May 1834, and all the Cherokee Press
printing activities were put on hiatus. In 1835, the Cherokee Press was
seized by the Georgia Guard.47
In 1835, a unofficial group of Cherokee representatives called the
Cherokee Treaty Party signed the controversial Treaty of New Echota, by
which the Cherokee Nation renounced its land in Georgia in exchange
for land west of the Mississippi river, in a new ‘Indian Territory’ (now
Oklahoma).48 The ratification of this treaty led to the darkest period in
Cherokee history, referred to as The Trail of Tears—the forced removal in
1838–39 of fifteen to seventeen thousand of Cherokees, about four thou-
sand of whom died before reaching their new land.49
29
Figure 47 Drawing by Worcester of the characters to be amended; new character shapes in brackets.
A B C
— — —
– The most significant change was the 180° rotation of the character Ꮩ,
V
now appearing as a V, rather than , in order to avoid confusion of this
character with Ꭺ. Worcester explicitly instructed the use of ‘the small
capital roman v’.52
– The 1835 shape of Ꮥ was very different from its previous form, which
was closely modeled on Sequoyah’s design. It now appeared in a form
which very closely matched Worcester’s transcription (and Hicks’ to
a lesser extent) of the character in his letter of December 1825 [Figure
49]. This new shape distinguished it more clearly from Ꭶ, beside which
it frequently occurs.
– The 1827 shape of the character Ꮯ was replaced by the Latin small
capital letter C from the Latin small pica font, also as instructed by
Worcester53 to avoid confusion with the character Ꮆ. This modification
represents a significant change to the character’s shape, especially
considering that the characters Ꮆ, Ꮯ, Ꮳ, and Ꮹ are all quite similar in
appearance and are only made distinct from each other by small struc-
tural elements. [Figure 50]
– A hook with a ball terminal was added to the ascender of the character Ᏺ
to better differentiate it from Ꮒ.
– The character Ꮅ was rounded off to reduce its confusion with Ꮲ.
– The counter of Ꮢ was opened up further than had been requested by
Lowry in 1827, and its leg made more vertical to bring it further away
from Ꭱ.
– For the modification of character Ꮦ, Worcester indicated what was
needed thus: ‘a little enlargement of the short perpendicular left hand
stroke, [which has been] found too liable to be broken off, thus trans-
forming the character into the other one Ꮣ’. Those parts of the charac- Figure 49 Ꮥ drawn by Hicks (left), and
by Worcester (right). Extracted from
ter indicated by Worcester differed slightly in shape in the final punch figure 34.
from Worcester’s instructions; additionally the spiral tail was closed
into a loop.
– Worcester did not specify the reasons for his request to modify the
character Ꮾ; in its new shape, it lost its serif-like terminal stroke and
was made more similar to a Latin numeral 6.
– The last character, Ꮌ, was recut in a pattern which differs fairly signifi-
Figure 50
cantly from Worcester’s design.
31
Figure 51 Syllabic characters of the second Cherokee font.
authorities, a new Indian Press54 was set up temporarily at a missionary
station called Union. Worcester and John Wheeler undertook to resume
the printing activities which had been suspended in 1834. The printing
press having been seized, they arranged to have a new one sent from
Boston,55 and with it the new revised version of the Cherokee font.
Worcester appeared not to have been overly satisfied with the new char-
acters. In a letter 56 from Dwight, in the Indian Territory, dated 2 August
1835, he writes:
‘I think the artist who made the alterations in the Cherokee type, which
I requested last year, cannot be the same with the one who formed the
original matrices, nor equally skilful in imitation..’
‘[It] will not answer at all. It is so unlike the pattern which I sent, that I am
led to suppose the pattern must have been defaced, before it came into the
artist’s [punchcutter’s] hands... Be so kind as to have a new punch made,
and type cast, and forwarded with our next supplies. I must attempt a new
pattern...’
The character was promptly recut and added to the font; it appeared in its
corrected shape in the 1835 specimen sheet. This and all the other charac-
ters amended in the 1834 recut of the font became the standard shapes for
the syllables they represented in the syllabary.
In the letter of 17 July 1835, Worcester also mentioned the wish to have
a second Cherokee font, ‘cast on a long-primer body’.57 Such a font was
never created. It would not be until the late 1850s that a second Cherokee
font was created, and no documentation of its creation process seems to
have been preserved.
This second font [Figure 51] differs in typographic style from the first
one. The first font was in the then contemporary, and ubiquitous, Scotch
Roman style. The second font was instead modeled on an earlier style, the
Didone. Figure 52 Character Ꮌ. Left: 1828 shape.
Typographic style aside, most of the characters of the second font Middle: 1834 shape. Right: Second font.
retain essentially the same structure as those of the 1834 amended version
of the first font. The only character to differ noticeably is Ꮌ; its shape in
the second font is closer to the shape it had in the first font before the
modifications of 1834 [Figure 52]. More significantly, the decision was ap-
parently taken not to include the character Ꮐ in the second font, problably 54. Walker. Native American writing
because it was almost never used. systems, p.147
55. Bass. Cherokee Messenger, p.187
Despite having been cut in a different style and with the character differ- 56. ABCFM 18.3.1 v.7 23(5)
ences highlighted above, it seems that the second font was intended to 57. ABCFM 18.3.1 v.7 234
33
Figure 53 The second Cherokee font (eg. last character on the right) cast on the same body, small pica,
as the first font.
35
APPEN DICES
Appendix 1 Numeric system created by Sequoyah, in his own hand.
38
A B C D E F G H A B C D E F G H A B C D E F G H
— — — — — — — — — — — — — — — — — — — — — — — —
39
Appendix 3 Two of the small pica fonts in the 1834 Baker and Greele specimen book.
40
Appendix 4 Syllabic characters of the 1828 version of the first Cherokee font.
Appendix 5 Definitive version of the first Cherokee font with the characters amended in 1834.
41
Appendix 6 Type specimen of 1828.
42
Appendix 7 Type specimen of 1835.
Ꭱ 4 Ꮲ 2 Ꮁ 2 Ꮸ 5 Ꮫ 13 Ꮳ 5 Ꮨ 3
Ꭰ 27 Ꭳ 6 Ꭺ 7 Ꮢ 12 Ꭻ 1 Ꭵ 3 Ꮮ 2
Ꮃ 8 Ꮇ 3 Ꮷ 5 Ꮒ 19 Ꮶ 2 Ꮕ 20 Ꮏ 4
Ꮵ 13 Ꮄ 14 Ꮍ 1 Ꭶ 18 Ꮙ 7 Ꮦ 3 Ꮚ 1
Ꮐ 9 Ꭽ 3 Ꮞ 4 Ꮩ 9 Ꮔ 4 Ꮉ 2 Ꮬ 1
Ꮽ 2 Ꮼ 3 Ꮠ 2 Ꭸ 10 Ꮎ 12 Ꮡ 2 Ꮊ 1
Ꮺ 4 Ꮰ 2 Ꮯ 3 Ꮣ 17 Ꮆ 4 Ꮱ 2 Ꮛ 1
Ꮅ 14 Ꮤ 5 Ꮘ 2 Ꭼ 11 Ᏻ 8 Ꭾ 3
Ꮑ 12 Ᏼ 4 Ꮗ 2 Ꮻ 7 Ꮴ 3 Ꮀ 2
Ꮌ 1 Ꮈ 11 Ꮜ 4 Ꭲ 20 Ꮧ 24 Ꮋ 1
Ꭹ 21 Ꭿ 16 Ꮖ 5 Ꭴ 21 Ꮾ 1 Ꮭ 2
Ᏹ 10 Ꮝ 31 Ꮓ 15 Ᏸ 5 Ꮪ 5 Ꮿ 5
Ꮟ 7 Ᏺ 3 Ꭷ 4 Ꮂ 6 Ꮥ 7 Ꮹ 6
43
V I S UA L S O U R C E S
Figure 2 Nefire, Eli, Scott, David and Meredith, Howard. Cherokee Teaching
Grammar, p.20. Oklahoma City and Park Hill: Notsi Press, 2001
Figure 3 Gilcrease Museum, catalogue number 4926.448
Figure 6 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.5, item 234
Figure 7 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.9, item 67
Figure 8 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.7 v.5, item 290
Figure 9 Gilcrease Museum, catalogue number 3526.343
Figure 11 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.7 v.5, item 234
Figure 12 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.7 v.5, item 121
Figure 20 Pickering, John. A Grammar of the Cherokee Language. Boston, 1825
Figure 21 Ibid
Figure 22 Ibid
Figure 23 Ibid
Figure 24 Ibid
Figure 25 Gilcrease Museum, catalogue number 4026.312
Figure 26 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.5, item 229
Figure 27 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.7, item 235
Figure 38 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.7, item 21(5)
Figure 40 Boudinott, Elias and Worcester, Samuel. Cherokee Hymns.
New Echota: Mission press, 1828
Figure 41 University of Tulsa, McFarlin Library. Schleppey ephemera, 6.1
Figure 45 Missionary Herald, volume 23 number 12, December 1827
Figure 46 Cherokee Phoenix, volume 1, number 9, 17 April 1828
Figure 47 Harvard University, Lamont library, archives of the American Board
of Commissioners for Foreign Missions, 18.3.1 v.7, item 234
Figure 53 University of Tulsa, McFarlin Library. Schleppey ephemera, 6.1
Figure 54 American Baptist Society. Cherokee Hymns. Philadelphia:
American Baptist Publication Society, 1866
Ehle, John. Trail of Tears: The Rise and Fall of the Cherokee Nation. New York:
Doubleday, 1988
Foster, Geo. E. Story of the Cherokee Bible. Second Edition. Ithaca: Democrat Press,
1899
Kilkpatrick, Jack Frederick and Kilkpatrick, Anna Gritts. New Echota Letters:
Contributions of Samuel A. Worcester to the Cherokee Phoenix. Dallas: Southern
Methodist University Press, 1968
New England Type and Stereotype Foundry. Specimen of Printing Types from
the New England Type Foundry. Boston, 1834
Walker, Willard and Sarbaugh, James. “The Early History of the Cherokee Syllabary.”
In Ethnohistory, volume 40, number 1, Winter 1993
Worcester, Samuel A. Cherokee Phoenix, volume 1, number 1, 21 February 1828
OT H E R WO R K S C O N S U LT ED
Bass, Althea. “The Cherokee Press”. In The Colophon: A Book Collectors Quarterly,
part 13
Chamberlin, A.N. Cherokee Pictorial Book with Catechism and Hymns. Tahlequah:
Mission Press, 1888
Dwight Presbyterian Mission. Dwight, a brief history of old Dwight Cherokee Mission
1820-1953. Tulsa: Swight Presbyterian Mission, Inc., 1954
Foster, Geo. Se-Quo-Yah, the American Cadmus and Modern Moses. Milford 1885
Greenleaf, Benjamin. Cherokee Almanac: 1860. Park Hill: Mission Press, 1860
Hider, Sam. Writing Cherokee: A Syllabary Writing Practice Book. Various Indian
Peoples Publishing Co., 1995
Holmes, Ruth Bradley and Smith, Betty Sharp. Beginning Cherokee. Norman:
University of Oklahoma Press, 1977
Kilkpatrick, Jack Frederick and Kilkpatrick, Anna Gritts. The Shadow of Sequoyah.
Norman: University of Oklahoma Press, 1965
Moulton, Gary, E. John Ross, Cherokee chief. Athens, The University of Georgian Press
Nofire, Eli; Scott, David and Meredith, Howard. Cherokee Teaching Grammar. Second
Edition. Oklahoma City: Notsi Press for the Cherokee National Historical Society, 2001
Starr, Emmet. History of the Cherokee Indians and their Legends and Folk Lore.
New York: Kraus Reprint Co., 1969
United States. War Dept. Report of the Secretary of War. On Indian affairs.
Washington, 1826
Worcester, Samuel A. Cherokee Almanac: For the Year of our Lord 1836. Union:
Mission Press, 1836