 | Notes on Recent Work in
Descriptive Bibliography |  |
TYPOGRAPHY AND LAYOUT
I published "The Identification of Type Faces in Bibliographical
Description" in
Papers of the Bibliographical Society of America
, 60 (1966),
185–202, and reprinted it (with a
"Postscript") in Journal of Typographic
Research, 1
(1967), 427–447; it was reprinted in a French translation in
Arts et techniques graphiques, 86 (septembre-octobre
1972), 41–55. My sug-
gestions for classifying or naming a type
design (on various levels of speci-
ficity) and for measuring type faces (not type
bodies) were followed by
an example of how one might report features of layout such
as the size
of the type-page and the typography of running titles and headings.
The
idea of measuring faces rather than bodies has been advocated even in
a
professional printing journal: Eugene M. Ettenberg, writing in
Inland
Printer/American Lithographer for January
1969 (162.4) on the possibility of
switching to metric measurements,
said, "it is to be hoped that a newer
method for designating type size, should it
come, would not repeat the
mistake with which we have lived so long, that of making
the type size the
size of body. Type should truly be graded by its optical size
rather than
by its metal size" (p. 48).
In supplementing my essay half a century later, I wish to call attention
first to
Harry Carter's
A View of Early Typography up to about 1600
(1969;
reprinted in 2002 with an introduction by James Mosley)-not only be-
cause Carter is always worth
reading but also because he takes Bowers
as a starting
point. His aim is to show how, given present knowledge of
typographic history, one
can respond to Bowers's doubts about how far
a
bibliographer can go in identifying a type face. At the outset, he says that
a
face is "sufficiently identified" by the name of the punch-cutter (that
is, the
design), the body size, and the "style" as described by standard
adjectives. The
second chapter opens with this passage (which includes a
comment on measuring faces
that could be extended to later periods):
Why should it be so difficult to do what Professor Professor Bowers thought would be de-
sirable, if it were possible: to particularize
the type used in a book? If you measure
it, and find that 20 lines set in it take
up, say, 85 mm., you restrict it to a class of a
particular body—a property of a
type-founder's mould. It remains to describe the
face, which might be cast on a
variety of bodies. I had rather name typefaces for
size by the conventional body
that would best fit them, Pica, Long Primer, Minion,
and such, than by numbers,
qualifying these terms if necessary by adding "large"
or "small". Some time in the
early part or the middle of the sixteenth century these
names acquired fixed
meanings. Until it becomes appropriate to use them it is safest
to measure the face
of a fount, which you can do if you have a powerful magnifying-
glass and a fine
scale and measure from the top of b to the bottom of p or the extent
of an Italic
f. This, called the gauge of the face, cannot vary. (p. 23)
Carter proceeds to discuss the value of naming the designer-cutter (which
sums up
"all manner of information as to place and time, circumstances
and relationships"),
and he notes the inadequacy of the usual terms of
classification. "It is evident,"
he says (and bibliographers should take
note), "that in considering the face of a
fount of type we are in a world of
art, styles, difficulty of saying what styles,
inherited forms, human hands"
(p. 24). But although he concludes that this "humble
art" is not "suscep-
tible of scientific treatment," he clearly believes that
rigorous historical
description is possible.
The next book that should be mentioned, because of its prominence,
is Philip Gaskell's
A New Introduction to Bibliography (1972,
1974), though
unfortunately its section called "Type Sizes; and
Description" (pp. 12-
16) makes recommendations that raise some questions. To begin
with,
Gaskell says that a twenty-line measurement (from the top of an
ascender
in one line to the same point in the twenty-first line below) gives
one
(after division by twenty) the "apparent body-size," though he adds, "It
is
important to make sure that the lines measured are set solid." He says
that if the
vertical gap between the bottom of a descender in one line
and the top of an
ascender in the next line is half a millimeter or less,
"either leaded or printed from a fount cast on an oversize body." How,
then, has the body size been made "apparent"? This is precisely the dif-
ficulty that caused me to recommend not referring to the body size but
focusing instead on what is actually visible, and I continue to believe that
my suggestion is sound (giving the vertical size and the x-height of the
face combined with a multiple-line measurement). The specific number
of lines to be measured is a less important matter, and Gaskell's choice of
twenty is traditional; but he should at least have indicated that a twenty—
line measurement is not the only responsible possibility, since the greater
convenience of ten (which I pointed out) was recognized as long ago
as McKerrow's time (p. 306 of his 1927 Introduction), and later by Allan
Stevenson in his 1961 volume of the Hunt catalogue (p. clxxxii). Another
problem with Gaskell's discussion is the form of report he stipulates. Af-
ter the vertical dimension of the face is measured, he says, "the x-height
and the capital height are measured in millimetres and the result is pre—
sented in the form: '[face height X 20] X [x-height] : [capital height]'";
his example is "Body 82. Face 80 X 1.7 : 2.5." Although this system is
"adapted" from one "introduced" by H. D. L. Vervliet in The Type Speci-
men of the Vatican Press 1628 (1967), it is not in any sense an established
standard. He should have made clear that he is simply recommending this
form but that other plans are plausible; it is regrettable that his prescrip—
tive assertion ("the result is presented in the form …") will mislead the
very people that such an introduction is meant to instruct.
Gaskell's treatment does, however, make two points that usefully sup-
plement my
discussion. First, he notes how measurements are complicated
by paper shrinkage and
ink seepage. His guideline is that shrinkage (as the
dampened paper dried after
printing) reduced the type-face dimensions
"by about 1 per cent, and occasionally by
as much as 2½ percent." One
must certainly keep this point in mind; all one can do
is to take multiple
measurements in different copies and caution the reader about
the prob-
lem in an introduction. As for the ink spread that occurs when the
type
sinks into the paper, one must be sure to measure (using
magnification)
only the indented impression, not the total ink smear. This point was
ef-
fectively made in 1988 by Adrian Weiss in explaining why
photographic
reproductions are not useful for making type measurements
("Reproduc-
tions of Early Dramatic Texts as a Source of Bibliographical
Evidence,"
Text, 4: 237–268). (Cf. Joseph A. Dane's discussion of reproductions and
shrinkage in relation to
type measurement in The Myth of Print
Culture
[2003], pp. 75–82.) The other helpful point in Gaskell's
discussion is that
the height of a capital should be given as well as the total
height (ascender
to descender) and the x-height. David L. Vander
Meulen, in his Dunciad
ments for revealing the proportions of a design, and he suggests simply
giving the three figures (measured to the nearest third of a millimeter)
with virgules separating them (e.g., "face 4.0/2.7/1.7").
There are two other matters that I wish to add to what I wrote
in
1966. First, the indication of the number of lines on a typical page
is
meaningful only for books of prose or long narrative poems. For books
of
short poems, where few (if any) pages have text extending down to the point where the
last line would occur—and where even those pages may
have stanza breaks of more than
a single line space—the bibliographer
will have to depart from the formulaic
statement I suggested and explain
precisely what the situation is (indicating, for
example, the amount of
space between stanzas, altering the multi-line measurement to
something
smaller than ten lines if all the stanzas in the book are shorter than
that,
and noting whether each poem begins on a new page).
The second matter is that there may be times, especially in the hand-
press period,
when it would not be objectionable to give the type-body
size as well as the
type-face size. When the multi-line measurement comes
close to what it would be at a
specific time for a particular standard type
(pica, long primer, etc.) if set solid
(always allowing for paper shrinkage),
one may be justified in concluding that this
type was the one used. Gaskell
provides a table (p. 15) giving the names of the nine
most common body sizes in the hand-press period and the twenty-line measurements of
each
at varying times. (Carter gives a similar list of twelve sizes on p. 127;
and
J. D. Fleeman gives a list of thirteen sizes,
important for not being based
on multi-line measurements but rather on such
combinations as "ly", in
his Samuel Johnson bibliography
[2000], p. xxxvii.) A much more com-
prehensive table was published
by John Richardson, Jr., in "Correlated
Type Sizes and
Names for the Fifteenth through Twentieth Century"
(Studies in
Bibliography, 43 [1990], 251–272). This fourteen-page table
re-
cords twenty-line measurements in millimeters, inches, and points
(pica,
fournier, and didot), along with the type names in seven countries.
Al-
though he is careful to measure the type impressions under magnification
and
recognizes the problem of paper shrinkage, one may still feel that
some of his
measurements are unrealistically precise.
Indeed, this table has been severely criticized by James
Mosley (in
"Type Bodies Compared," Journal of the
Printing Historical Society, n.s., 23
[2015], 49–58) for
"indiscriminately" bringing together measurements
from "sources that are unevenly
reliable" (p. 56); he regards it as a "useful
reminder of the problems
involved"—problems that to some extent affect
the other published tables as well.
Mosley's main point is that size names
varied in the sizes they signified in
different geographical locations and
the various sizes (measured to hundredths of a millimeter) that certain
names could refer to in ten specimens from about 1585 to 1768. Although
one can use the earlier tables for giving a general sense of relative sizes,
Mosley has made it clear that one must be aware of local traditions at
different times. Furthermore, anyone concerned with measuring type and
naming the body sizes in pre-eighteenth-century books should read two
detailed reviews of W. Craig Ferguson's Pica Roman Type in Elizabethan
England (1989): one by Adrian Weiss in Papers of the Bibliographical Society
of America , 83 (1989), 539–546; the other by John A. Lane in The Library,
6th ser., 14 (1992), 357–365. Ferguson's book, which is intended to help
date books and identify their printers, uses a brief checklist of features for
identifying types, but they are not adequate for determining the specific
type stock owned by a given printer (see the comments on font analysis
below).
In my account of levels of specificity for identifying type faces, the
highest one
involves recourse to specimen books, and there is one bibli-
ography that can be
singled out for its excellence in performing this task
for post-Renaissance books.
David Gilson's
Jane Austen
(1982)
includes
type descriptions supplied by Nicolas Barker in
the following form (this
one for Northanger Abbey and
Persuasion, 1818):
Printing: type: Vols.1 and 2 were printed by Roworth, Vols.3 and
4 by Davison. In
Vol.1 the 'Biographical notice' is set in Caslon Small Pica, the
'Small Pica No.2' from
the Caslon and Catherwood specimen in Stower's Printer's grammar
1808, with figures
from 'Small Pica No.1'; the 'Advertisement' is in a
Caslon Pica roman, identical with
the 'Pica No.1' in Caslon and Catherwood's
1819
Specimen, apart from some letters
which are of a different
cut, perhaps slightly earlier. The text of Vols.1 and 2 is set in
Caslon Pica roman,
the first in the Stower Caslon and Catherwood specimen. The
text of Vols.3 and 4 is
set in a Caslon Pica roman, between the second pica roman
shown in the Stower Caslon
and Catherwood specimen, and 'Pica No.4' in Caslon
and Catherwood's
1821
Specimen. … (p. 81)
Barker also contributed "A Note on the Typographical Identifications"
(pp. xi-xiv),
which is worth reading for its comments on the difficulties
of using typefounders'
specimens and the particular problems associated
with the 1780–1830
period. He notes that the work requires "an eye ac-
customed to the minute details
that distinguish the cut of one letter from
its equivalent in another fount." This
point illustrates why Gilson was
wise to turn to a specialist—and suggests that
other bibliographers might
do the same.
For classifying type faces into broad categories, my suggestion of the
DIN system
(the initials standing for Deutsches Institut für Normung
since 1975),
as presented by James Mosley in 1960, remains
appropriate
in my view—especially the terms "Renaissance," "Baroque," and "Neo-
his New Introduction (p. 16). Robert Bringhurst, known for his thoughtful
discussion of typographic matters, has provided one of the best statements
of the reasons to use such historically allusive terms. In his often reprinted
The Elements of Typographic Style (1992), after mentioning research on "sci-
entific descriptions," he says:
But letterforms are not only objects of science. They also belong to the realm
of
art, and they participate in its history. They have changed over time just as
music,
painting and architecture have changed, and the same historical
terms—Renais-
sance, Baroque, Neoclassical, Romantic, and so on—are useful in each
of these
fields. … Typography never occurs in isolation. Good typography demands
not
only a knowledge of type itself, but an understanding of the relationship
between
letterforms and the other things that humans make and do. (p. 111)
He then presents a historical classification using these headings: Re-
naissance,
Mannerist, Baroque, Rococo, Neoclassical, Romantic, Real-
ist, Geometrical
modernism, Lyrical modernism, and Postmodern
(pp. 111–123).
There will never be an end, however, to the construction of clas-
sification
systems, and many other discussions of the subject have been
published since 1966.
Most of Alexander Lawson's
Printing Types: An Intro-
duction (1971,
1990), for example, is devoted to classification. After a
brief
summary of four systems (Vox, Association Typographique Internation-
ale,
British Standards Institution [BSI], and DIN), he offers "An Attempt
to Formulate a
Rational System" (pp. 45–119); although it is useful for
its numerous examples of
designs and its historical discussion of each, the
system itself uses the
traditional "old style," "transitional," and "modern,"
which are widely thought to
be inadequate. Bringhurst himself offered a
"scientific" approach in "On the
Classification of Letterforms" (Serif, 1
[Fall
1994], 30–39); it amounts to an extensive analogy with
biological
classification (using such terms as phylum, class,
order, family, genus, and
species) and seems unlikely
to become widely employed. (His discussions
of some individual period styles
followed in the next several issues of
Serif.) Lesser
examples are J. Ben Lieberman's
Types of Typefaces and How
to Recognize Them
[1967] and Gordon Atkins's
The Classification of Printing
Types [1975].
Among the helpful treatments of earlier systems are Walter
Tracy's "Type Design Classification," Visible
Language, 5 (1971), 59–66
(on Vox, BSI, and DIN); John Dreyfus's "The Typographical Importance
of Maximilien Vox," Matrix, 17
(1997), 1–11; and Craig Eliason's "A
His-
tory of the 'Humanist' Type Classification," Printing
History, n.s., 18 (July
2015), 3–26. A criticism of the DIN
system is Gerrit Noordzij's "Broken
Scripts and the
Classification of Typefaces," Journal of Typographic
Re-
search, 4 (1970), 213–240.
In connection with naming specific type designs, my 1966 essay
noted
several listings of specimen books and several anthologies of type
faces;
some of the most important works in both categories have been
published
since that time. As for lists of specimens, there are Maurice
Annenberg's
Type Foundries of America
and Their Catalogs (1975, 1994) and James Mos-
ley's
British Type Specimens before 1831: A Handlist
(1984). Mosley also edited
facsimiles of the 1766 Caslon
specimen (1983) and the 1796 Stephenson
specimen
(1990), and John A. Lane edited facsimiles of
the 1768 and 1773
Enschedé specimens (1993).
In the category of anthologies of faces, the
two outstanding ones are Rookledge's International Type-Finder by Christo-
pher Perfect and Gordon Rookledge
(1983) and A Manual of Comparative
Typography: The
PANOSE System by Benjamin Bauermeister (1988). Both
are very
helpful, with their cross-reference charts and "earmark" tables,
but the PANOSE
system is more complicated to use and the book displays
many fewer faces. Another
worthy work is Albert Kapr's
The Art of Lette-
ring (trans. 1983), which
shows a large number of "typefaces of the present"
(and includes a brief comparative
discussion of classification schemes). A
few other examples are James Sutton and Alan Bartram's
An Atlas of Type-
forms (1968,
1988) and Typefaces for Books (1990),
Mac McGrew's
American
Metal Typefaces of the Twentieth Century
(1986, 1993), Centennial's
Typeidenti-
fier (1986, with its "Typographical Sleuthing"
sections), Lawrence W. Wal-
lis's
Modern Encyclopedia of Typefaces, 1960–1990
(1990), Michael Mitchell
and Susan Wightman's
Book Typography: A Designer's Manual (2005),
and
Alan Bartram's
Typeforms: A History (2007).
Descriptive bibliographers need a vocabulary for talking about type
designs (in
addition to classifying and naming them), and my 1966 com-
ments (which
deal primarily with weight and width) should now be
supplemented with Philip Gaskell's excellent "A Nomenclature for
the
Letter-Forms of Roman Type" (The Library, 5th ser., 29
[1974], 42–51).
Other glossaries are Type Evidence:
A Thesaurus for Use in Rare Book and
Special Collections Cataloging
(Association of College and Research Librar-
ies, 1990); Robert Bringhurst's "Sorts & Characters" and "Glossary
of
Terms" in his The Elements of Typographic Style
(1992), pp. 214–240; and
Theodore
Rosendorf's
The Typographic Desk Reference (2009,
2016). (See
also S. J. M. Watson, "Three
Notes on Nomenclature," Printing Historical
Society Bulletin,
46 [Winter 1998–99], 9–10.) The nomenclature and clas-
sification of
printers' ornaments are taken up by Roger Burford Mason
in "A Note on Type Ornaments, Borders and Flowers" (Albion, 20
[Sum-
mer 1983], 5–9), and by Jim Mitchell in
"The Taxonomy of Printers'
Ornaments" (Bibliographical Society of
Australia and New Zealand
Bulletin, 9
[1985], 45–60).
Near the end of my essay, I mentioned the possibility of identifying the
individual
font owned by a particular printer—that is, the specific col-
idiosyncrasies, such as damaged and wrong-font types. Although Charl-
ton Hinman in 1963 did some work of this kind (in The Printing and Proof-
Reading of the First Folio of Shakespeare), it was Adrian Weiss who showed
in detail how to pursue such analysis and use it to identify printers—
in several brilliant articles: "Font Analysis as a Bibliographical Method:
The Elizabethan Play-Quarto Printers and Compositors," Studies in Bib-
liography, 43 (1990), 95–164; "Bibliographical Methods for Identifying
Unknown Printers in Elizabethan/Jacobean Books," Studies in Bibliogra-
phy, 44 (1991), 183–228; and "Shared Printing, Printer's Copy, and the
Text(s) of Gascoigne's A Hundreth Sundrie Flowres," Studies in Bibliography,
45 (1992), 71–104.
Another sophisticated approach to type analysis, not mentioned in
my
1966 essay (because it had not yet been undertaken), involves
computer
assistance. Nicolas Barker, in his chapter of The Book Encompassed ("Typo-
graphic Studies"), ed. Peter Davison (1992), quotes from two
1982 articles
by the digital-type designer Charles
A. Bigelow as a basis for suggesting
the adaptation of computer
type-designing techniques to the historical
analysis of type faces (pp. 96–98). In
2003, two projects employing digital
enlargements of type faces were
the subjects of published reports. Kay
Amert explained a
system for superimposing two images, one made up
of horizontal and the other of
vertical screened lines (or of two tints) in
order to reveal minute differences in
type-face shapes ("Digital Com-
parison of Letterforms," Printing
History, 46: 21–35). The other report,
by Blaise Agüera y Arcos, described
the work he and Paul Needham
had done with computer
enlargements of Gutenberg's type faces, reveal-
ing that variations were present
among the types for any given letter
and suggesting that Gutenberg's early types
were made in nonreusable
moulds ("Temporary Matrices and Elemental Punches in
Gutenberg's
DK Type," in Incunabula and Their Readers: Printing,
Selling and Using Books
in the Fifteenth Century, ed. Kristian Jensen [2003], pp. 1–12). There are
also digital
atlases of type faces in progress (see, for example, Heiner
Klocke, "The Software Architecture of the Hebrew Type Digital
Image
Atlas," in Hebrew Typography in German-Speaking Regions: An
Interim Report,
ed. Klocke and Itari Joseph Tamari [2001], pp. 63–70). But at present
most
collections of type designs on the internet are commercially oriented
and require
searching by the name of the design, though "Identifont" and
"Fonts.com" do allow
for identifying certain digital fonts by answering
several questions about their
appearance.
Although the analysis of ink, beyond naming its color, is not likely to
be
undertaken for most descriptive bibliographies, the composition of ink
can be
revealed by cyclotron analysis in the same operation that discloses
the makeup of
paper (as mentioned above under "Paper"). An extensive
my Introduction to Bibliography (2002 revision), part 6, pp. 195–224—which
includes, in sections A-E (pp. 195–204), additional publications related
to the identification and classification of type faces. See also the New
Cambridge Bibliography of English Literature (1969–74), in which the relevant
sections are by Nicolas Barker, Terry Belanger, Graham Pollard, James
Mosley, and Peter Davison. In preparing oneself to describe type designs
and layout, one must always keep in mind that these are art forms.
| Notes on Recent Work in
Descriptive Bibliography | |