University of Virginia Library

Search this document 

Font Analysis as a Bibliographical Method: The Elizabethan Play-Quarto Printers and Compositors
by
Adrian Weiss [*]

In "Edward Allde as a Typical Trade Printer", R. B. McKerrow distinguished two classes of early printers, the printer-publishers who published books on their own as well as printing for others, and trade printers who printed almost exclusively for publishers and produced "a very large proportion of the play-quartos, the smaller volumes of verse, the prose-pamphlets . . . many of whom, it may be said, give a great deal of trouble by their casualness in the matter of imprints and dates." Perhaps McKerrow inspired the few subsequent studies of printers by commenting that "a comparatively small amount of work devoted to such minor people as these would help us very greatly in identifying the house of origin of many books which bear no printer's name"[1]; however, the inspiration was short lived as bibliographers turned increasingly to studies of specific texts based upon now-familiar methods of analysis. Nearly a half-century later, Peter Blayney reiterated the need for


96

Page 96
attention to the matter of printer identifications not only in books lacking an imprint, but also in books assigned to one printer and shared with one or more other printers, a practice sufficiently common so as to merit classification as "prevalent".[2] The categorical necessity of detecting and distinguishing two or more printers' work in a book hardly requires comment: it is a prerequisite to the analysis of compositors and presswork. Moreover, the identity of the printer(s) of a book link it to the personnel and production methods of a particular shop(s) and how they could corrupt a text.

In general, printer identifications in both categories of books have relied upon the recognition of ornaments and decorated upper-case initials, a form of evidence with limited reliability and value because of the common practice of lending among printers and because of the large number of books in which neither appear. A practical limitation in shorter shared books, particularly play-quartos, arises from the fact that printer's devices usually are restricted to the preliminaries and first letter of a text; later shared sections simply continue setting the text without an opportunity to insert ornamental stock.[3] In contrast, evidence from a printer's type fonts exhibits no such limitations and, together with identifiable types, can be considered de facto proof of his identity.[4] Hence, such evidence can be used to identify a printer's work in assigned and shared books, and distinguish sections of shared books or books printed by a printer with two or more fonts. Furthermore, it can provide a more accurate focus for compositorial and presswork analysis since, if two fonts appear in a book, at least two sets of type-cases were used. In turn, this fact implies the possibility of two compositors setting simultaneously or alternately and the consequent need for casting-off copy, factors that are considered crucial in influencing accuracy in the textual transmission process. Unfortunately, the subject of font analysis and its uses have received very little attention, despite its potential importance in bibliographical research.[5] It seems that so fundamental a physical fact such as the type fonts used in a book could hardly be overlooked by scholars habituated to working with vast amounts of minute details. But that, in fact, has happened in many cases where the presence of two or more fonts in a book has been overlooked and bibliographical analysis has proceeded as if the book were printed by a single printer in one font.[6] It seems fair to say that the discipline as a whole has suffered from the lack of a practical method of analysing fonts for the purposes of distinguishing sections of shared books and of identifying printers. One hopes that this present introduction to a method of font analysis and its use in bibliographical study will enable textual scholars to assess more accurately the impact of various aspects of the printing process upon early texts. Font analysis, although simple in nature, requires attention to the enormous amount of detail that characterizes a particular font. My discussion of font differentiae perhaps draws attention to more detail than is necessary, but I think that too much is the best option in this case despite the potential tedium for readers and I apologize in advance. Likewise, I believe that this introduction would be incomplete without an explanation of the logic that underlies


97

Page 97
the use of font analysis as a bibliographical method and comments about the kinds of problems that can be resolved by font analysis.

I

On the simplest level, font analysis distinguishes between two fonts in dissimilar faces with no greater difficulty than obtains, by analogy, in recognizing that a modern face such as Courier is different from Prestige. In general, typefaces can be distinguished upon the basis of differences in size, either apparent or real, and stylistic differences in the design of the letters. The width and height of the body of the types respectively determine the horizontal spacing between letters and the vertical spacing between lines, although measurements of the latter using the standard method are sometimes less than precise because of paper shrinkage across four centuries.[7] Width is generally classified in relative terms such as "condensed" or "narrow", "medium", "wide", and "expanded". Most letters (or sorts) of the S-face are set on a wider body than in the Y-face, especially in regard to 'a r' and the thin letters 'i j l t'. The effect of this body-size on overall appearance can be seen by comparing the spacing in combinations such as "are", "and", and "illusion". The type seems spread out in the S-face leaving more "white space" within words in a line of set type. The use of wide and expanded 'k x z', and a bold left-stem on the 'v y' adds to this effect.[8] In contrast, the thin letters of the Y-face are set on a narrower body, and, along with the medium 'k v x y z', produce an appearance of more densely packed type, with quite prominent white spaces between words, but little within words.

The general design of the S-face (see footnote 12 for the naming of these fonts as S and Y) further contributes to its squarish, open appearance. The vertical elements of the letters are thicker than the horizontal curved portions so that the former print more boldly than the latter. In technical terms, the "weight" or relative density of blackness of the parts of a letter establishes the "contrast" between thick and thin strokes, and the orientation of thick to thin defines the "stress": thus the S-face exhibits "vertical stress" or vertical "shading". The stress is not necessarily uniform in all letters, but dictated by shape. The diagonal letters (A M N S V W X Y Z v w x y z) exhibit oblique stress with thin vertical or right-leaning and bold left-leaning (relative to the baseline) strokes. In the lower-case letters, the bold inking of shoulders (a h m n) and bowls (b d p q) at the x-line makes the typeface appear higher and wider than it actually is. On the other hand, the Y-face is relatively unstressed so that all portions of a letter print with nearly equal boldness, except for the light shading that characterizes the shoulders of the tall letters, ligatures and the curved portions of several lower-case letters. The Y-face's lighter weight and contrast is obvious to the naked eye even when heavily inked. Viewed at high magnification (30X) in original books, the metal elements that form Y-face letters (i.e., the surface of the type which makes contact with the paper) seem rounded on the edges, whereas S-face letters exhibit elements with usually sharp squared edges.


98

Page 98

An analysis of the specific differences in the structure of the letters in the two faces can clarify the relation between the stylistic design of specific letters and the overall appearance of a typeface. Font analysis is facilitated by the fact that letters can be classified in groups on the basis of shape. A major difference in design can be seen in the group consisting of the 'f' and the long-(or tall-) 'ſ' and their associated ligatures. In the S-face, the uniformly thin shoulder stroke is a rather flat arc which abruptly curves just below the h-line into the bold vertical stem. It terminates at the right with a quite small "pear". The shoulder frequently appears shaded because of the accumulation of ink along its bottom edge and around the pear; lightly inked letters and the shoulders of some ligatures reveal the actual weight of the shoulder. In many of these letters, damage has nudged the right half of the shoulder upward so that it terminates horizontally. The shoulders of the 'ſſ ff ffi' ligatures are similarly formed. Both shoulders are of equal height and the left joins the right slightly below the h-line, rather than being vertically staggered as in the Y-face. The 'ſh ſl ſt' ligatures clarify this design characteristic. A bit of the ascender of the 'h' and the 'l' in the former extend above the shoulder junction, while the 't' of the 'ſt' ligature reaches almost to the h-line before joining the shoulder. Damage frequently flattens the shoulders into a single continuous arc and nudges the right termination upward.

In contrast, the tall letters of the Y-face are shaded. The shoulder of the 'f' and tall-'ſ' commences with a graceful curve just above the cross at the x-line and thins as the stroke moves through a diagonal to the horizontal, then broadens in the downward stroke that completes the shoulder. It terminates noticeably below the h-line with a large pear. The letters in the ligatures are vertically staggered: the left 'ſ' or 'f' is a bit shorter than the right letter. The shading of the downward stroke that completes the shoulder is obvious in the ſh, ſl and ſt ligatures, where the 'h l t' are shortened to accommodate the low junction with the shoulder. The junction commonly gathers ink and accentuates the shaded structure of the shoulder. The difference between the two faces in regard to shoulder structure is quite obvious at high magnification where the actual impression of the face elements can be detected without the obfuscation caused by the inking-in of the curves and joints. The overall effect of the shoulder structure in the ligatures and tall 'ſ f' is that the horizontal space between lines in the S-face is uncluttered by the bold inking produced by the shaded, staggered shoulders that characterise the Y-face. Finally, the 'ct' ligature provides an easily detected discriminant. The "open" 'ct' ligature is typical of the S-face: the body of the 'c' is slanted left and seems improperly formed since neither end of the 'c' "closes up" with the 't' and a considerable amount of white space obtains between the two letters. In contrast, the Y-face uses an elegantly formed 'ct' ligature in which the body of the 'c' is fully formed and closes up with the cross and base of the 't'; the letters are joined by a graceful link whose orientation and size balances the 'c'.

The two faces are further differentiated in the lower-case by the formation of the shoulders, bowls and top-serifs. In the S-face, the counter, or enclosed white space, in the round (or "oval" to be precise) letters (o d q)


99

Page 99
appears to lean left as if the axis were rotated counterclockwise. In the 'o', for example, the thin portion of the letter occurs at 11 and 5 o'clock, rather than at 12 and 6 o'clock. The ascender of the 'b' and descender of the 'p' replace the part of the oval to the left of the thin part; the converse is true of the 'd q'. Thus the junction of ascender and bowl forms an oblique angle just below the x-line, a proportioning (less than half) that accentuates the apparent height of the ascender and makes the bowl seem smaller. The stylistic theme seen in the oval letters appears uniformly in other lower-case letters. If attention is limited to the portions of the 'b p a m n h' abutting the x-line (by covering the lower portion of the letters with a card, for example), it can be noted that the letters are nearly identical. The shoulder of the 'm n h' is formed by a thin straight line at an oblique angle that rises to the x-line, then curves abruptly downward with an angular rather than a smoothly curving stroke. The right shoulder frequently prints as an angle rather than a curve because of heavy inking. This characteristic is accentuated in two letters. The shoulder of a common variant 'n' is wider (+0.1mm) than the right shoulder of the 'm'. In undamaged samples of the 'a', the low junction of the stem (or back) and the nearly horizontal top-stroke of the bowl is considerably below the x-line and leaves the "squarish" shoulder "hanging" over a prominent amount of white space. The formation of top-serifs is identical in the 'p q m n': by analogy with hand-writing, the downward stroke that forms the stem begins on the x-line with the broad tip of the pen set at the characteristic oblique angle and bends to the vertical almost immediately. The top of these letters frequently prints as a horizontal zig-zag because of the agreement of the coincident oblique angle of the top-serif and shoulder (or bowl-top of the 'p q'). However, the 'u' violates this stylistic uniformity with its horizontal, bold, left top-serifs, although it agrees with the base-serifs of the 'a d' which are inversions of the top-serif stroke of the 'm n p'. The top-serif on the tall letters (b d h l k) is a left-pointing triangle, while the bottom serif extends on both sides of the descender of the 'p q'. The overall effect of the oblique "fineline" at the x-line in S-face lower-case letters is to "square up" the face at the x-line, with the left counter of round letters complimenting the angular shoulders of the 'm n h'.

Lower-case letters are unstressed in the Y-face. The 'o' is formed by a round stroke of uniform thickness. The bowls of the 'b d p q' exhibit barely noticeable vertical stress because the portions of the bowl which join the vertical stroke are thinned slightly at 12 and 6 o'clock. In contrast to the distinct oblique junction angle of the S-face, Y-face bowl and shoulder junctions are nearly right angles at the x-line. In the 'b', for example, the round line to the left of the axis is bent upward toward the horizontal to join the ascender almost on the x-line at half the height of the ascender. At the bottom of the letter, the round line of the bowl touches the baseline and curves upward slightly to join with the ascender, which is thus displaced above the baseline. Evenly proportioned round letters result from this design concept, which is inverted in the 'p q': the baseline junction is nearly horizontal, while the top junction forms a slight oblique angle. The shoulder stroke of the


100

Page 100
'm n h' begins nearly horizontal and immediately broadens from a thin tip at the junction with the vertical stems into a symmetrical, round curve. The width of the shoulders of the 'h n' are identical to the right shoulder of the 'm'. The graceful rounding of the Y-face is most obvious in variants of the 'a', which is well-proportioned compared to the awkward 'a' of the S-face. Its shoulder, back and base-serif are formed by a single gracefully flowing stroke, and the letter is "closed up" by the higher junction of bowl and back. Serifs of Y-face letters are frequently worn away or indistinct.

The two faces can be further differentiated by lower-case diagonal letters (k v w x y z) and the 'g'. The S-face diagonal letters are commonly of expanded width which is accentuated by pronounced serifs, the right-angle crossing of the diagonals in the 'x', and the long arms of the 'z'. Pronounced serifs and heavy oblique stress of the left stem creates an impression of extra width in the 'y v', although they are of medium width. The "squat" appearance of the 'k' results from the low junction of the link of the arm-leg at ¼-height on the ascender, the right angle enclosed by the arm-leg in the common variants, and the pronounced arm and leg serifs. In contrast, medium-width diagonal letters are found in the Y-face in the 'v w x y' sorts, although a condensed 'y' variant is common. The left stem of the 'y v' exhibit minimal stress and a smaller enclosed angle than in the S-face. Several 'k' variants occur, but the junction of the arm-leg link at about ⅓-height on the ascender is always higher than in the S-face 'k' and produces a more balanced proportioning. Finally, the loop of the 'g' is noticeably larger than the bowl in the S-face; the loop-stroke begins at the base line, while the link pushes the bowl to the x-line. In Eld's version of the S-face, the left edge of both bowl and loop are vertically aligned, creating a left-slant; in another common variant, the bowl and loop are centered for an erect appearance. The bowl of the 'g' in the Y-face, in contrast, is approximately the same width as the loop, and the letter is vertically aligned.

Stylistic differences are generally more obvious in the capitals, particularly with respect to several awkward, poorly proportioned letters in the S-face which contrast with the more balanced symmetry of the Y-face capitals. The gross, excessively wide and awkward 'M' of the S-face is characterised by a single left top-serif and slanted legs. The Y-face uses a symmetrical 'M' with double shoulder serifs that are integral to the 'V' stroke that connects the erect legs. The bowl stroke of round S-face capitals such as 'P B D R' begins considerably to the left of the ascender on the characteristic oblique angle and complements the accentuated shoulder- and base-serifs of other capitals, while in the Y-face, the bowl stroke is horizontal and barely overextends the ascender. The S-face uses an obtrusively large 'P' whose bowl stroke terminates far below the x-line and does not close up with the ascender; the letter clashes aesthetically with the 'B' and especially the 'R', with its aenemic bowl that terminates at the x-line and its stressed tail that extends awkwardly beyond the bowl. In the Y-face, the 'B P R' are of similar proportions. The tall rectangular 'E F' are noticeably narrow for the S-face, particularly in regard


101

Page 101
to the 'E', where the foot barely extends beyond the medial cross and arm. The cross of the 'E' appears to be high on the ascender. The two letters are square in the Y-face, with longer horizontal members and distinctive serifs. Overall, S-face capitals overwhelm the lower-case letters on the printed page; perhaps the capitals were cut by a different punchcutter who failed to integrate them stylistically with the lower-case. Conversely, S-face punctuation sorts appear small for the face. The ascender of the exclamation point and the back of the query are bisected by the x-line and do not rise to the h-line. The comma of the semi-colon rides above the baseline, with the dot at or above the x-line, and the dots of the colon are positioned between base- and x-line. The comma is positioned normally on the base-line. In contrast, Y-face punctuation marks conform to the proportions and design of the letters. The Y-face query is a gracefully formed, full-height mark, as is the exclamation point, and the other marks are positioned correctly for the height of the face.

A few hours examination of the two faces in originals or high quality facsimiles of any of the books noted in later discussion will familiarize the reader with the essentials of typeface design and structure. The considerable differences between the S-face and the Y-face render them easily distinguishable even to the novice after such an exercise. It seems a fortunate coincidence that in many cases of shared printing, versions of the two faces alternate in sections of a book. In such cases, a comparative analysis need only progress through a few groups of letters before the "gross features" illustrated above distinguish the different fonts, providing sufficient cause to suggest that further bibliographical analysis proceed upon the assumption that two printers were involved. However, that is merely the beginning step in the application of font analysis. A book printed in more than two fonts almost inevitably requires distinguishing two same-face fonts, followed by an attempt at identifying the fonts in other books for the purpose of sorting out the printers. That process is facilitated by a comprehensive font composite that graphically records the distinguishing traits of a given font.[9]

The identification process itself requires that the font in the target book be compared to extremely similar, same-face fonts in other books. Similarly, in many instances of sharing that involved more than two printers, two or more versions of the same typeface are encountered in different sections of a book. In both situations, the bibliographer then is faced with a more complex task of analysis, since distinguishing between fonts in the same face is a much more difficult proposition because the general stylistic traits are identical except for infrequent variants. Assuming that the two same-face fonts and their respective sections of the book have been successfully distinguished, an attempt to identify the printers is further complicated by the fact that, quite understandably, the preliminary candidates will be selected primarily because they used the typeface in their known work. Hence, the general kind of stylistic analysis described above must be supplemented by a more refined and detailed focus upon potential discriminants within the context of nearly identical fonts. As a general rule, discriminants can be detected in this situation


102

Page 102
because each font has a unique history of usage, abuse and transformation that in all probability will have produced both obvious and subtle variations.

II

The refined evidence required for distinguishing same-face fonts can be divided into internal and external categories depending on whether a potential discriminant is imported from a dissimilar face, or whether it is a variation within the stylistic parameters of the face. In both categories, a discriminant is two-dimensional, and consists of the qualitative stylistic difference combined with the quantitative proportion of differing letters in a sort. External discriminants are easily detected because of the clash in the stylistic features noted in earlier discussion, and are imported into a font through processes of fouling, replenishment, and large-scale mixing of two fonts. The proportion of wrong-face letters in a sort usually indicates which process is involved: a high proportion of wrong-face letters in a given sort suggests a high probability of replenishment, and conversely, a small, or better yet, almost infinitesimal proportion suggests that fouling is responsible. Mixing introduces significant proportions of wrong-face or stylistically different letters into most sorts of a font. It is impossible to state fixed proportions or rules that apply to all fonts.

The distinction between replenishment and fouling is based upon the assumption that replenishment is responsible for the presence of abnormal (wrong-face or miscast) letters in significant quantities in a population which retains a varying proportion of normal letters in the depleted sorts. The converse is taken to indicate fouling since the overwhelming majority of letters in a sort consists of normal letters. The distinction can have serious practical implications: the former come from outside the shop and merely provide an additional discriminant in a single font; however, the latter are from within the shop, and indicate that the font belongs to a printer with two or more fonts. In turn, this permits the narrowing of the field of candidates to such printers and provides an additional criteria to guide the search for the printer of a section of a shared book. The correct candidate will use two fonts that either could produce or actually do exhibit an appropriate degree of mutual cross-contamination.[10] The replenishment process produced four categories of external wrong-face or miscast capitals, ligatures and lower-case letters. Although many miscast letters are in the correct-face and hence internal discriminants in a technical sense, in most instances they present such a striking contrast to properly formed letters that they warrant classification as external discriminants. As a general rule, potential discriminants must be checked against the same letters in other printers' fonts to verify that they are abnormal for the typeface. The on-going process of piecemeal replenishment usually will have produced quantitative differences in the proportions of replenished abnormal and original letters in same-face fonts.

Wrong-face capitals are the most easily detected external discriminants and typically are produced by replenishment of a Y-font with S-face capitals.


103

Page 103
The latter are of a slightly larger design scale but are cast on roughly the same pica body size as a Y-font and thus do not distort baseline evenness.[11] Capitals exhibiting radical stylistic differences, as noted above in regard to the symmetrical Y-face and the oversized awkward S-face 'M', are difficult to overlook even when separated by pages, and other wrong-face capitals with minor stylistic differences become obvious when juxtaposed with their normal counterparts. Juxtaposition with normal tall letters and ligatures likewise calls attention to differences in height. The difficulty of locating samples of original or replenished upper-case letters varies with the kind of text: prose texts often require a cover-to-cover search to produce a few samples while verse texts usually produce a plethora of samples of a replenished upper-case letter plus remnants of the original. Thomas Purfoot's Y-font in The Fawne Q1-Q2 (STC17483-4, A-D) provides a clear example of extensive replenishment with S-face capitals and the proportions which emerge. In this alternating verse/prose text, capitals tend to be concentrated in verse pages. The awkward S-face 'M' appears in Q2 seven times before and three times immediately after the original Y-face 'M' which is first seen at B2:33 and then at C4:38 for two appearances in contrast to 26 appearances of the replenished 'M'. In higher frequency sorts, a page or two may provide an adequate number of samples. Seventeen replenished S-face 'A', for example, appear in D2v, with one conveniently juxtaposed at D3:11 to an original 'A' at D3:10. Or, the actual count may be low but revealing, as on B1v of Q1, where two replenished 'P' (B1v:1,21) appear amidst four originals (B1v:7,15,21,34), and on C1, where two replenished 'P' (C1:23) and three originals (C1:24,31) are juxtaposed; and the S-face 'S' at (Q1) B1:12,28 vs a normal 'S' at B1:20. A difficult situation occurs when the sample population is so small in one sort that replenishment must be inferred from the contextual replenishment that is obvious in other sorts. For example, an original 'R' appears at (Q1 and Q2) B1v:32, and replenished letters at (Q1) B2v:9, C2v:23, 27, and in Q2 at B2v: 8,22, C3v:22. When no sample of the original letter can be found to provide a basis for comparison, a close examination may reveal a basis for inferring replenishment from wrong-size criteria. For example, the 'N' sort in Purfoot-Y1 appears to have been totally replenished; it is taller than the Y-face 'G' (2.8mm vs 2.75mm) and equal in height to medium S-face letters such as the 'A B D F R' and it usually rides low on the baseline.[12] However, a survey of other fonts reveals that this capital 'N' is unique in that it occurs as the normal letter in both S- and Y-fonts. All other S- and Y-face capitals differ in design and size.

In some instances, proportions such as those noted in the replenishment of Purfoot-Y1 with wrong-face capitals may be deceptive. As a general rule, once the possibility of replenishment because of wrong-face capitals is entertained, the lower-case letters and ligatures must be checked to verify the correct-face and then be confirmed in another book. Nicholas Okes's S-font (1607), for example, was purchased with Guyot capitals, but the lower-case letters and ligatures are S-face except for the medium 'ſſi' and 'ffi' ligatures that somehow slipped in.[13] Similarly, the different proportions caused by fouling as opposed


104

Page 104
to replenishment can create the impression that the latter is responsible; the cross-contamination of Eld-S and Eld-Y1 illustrates this problem. The odd method of setting the speech prefixes in Sejanus (STC14782, 1605) with a pica capital followed by small capitals (as on B3v:5,6,7,8,9,21) put a severe strain on the S-font 'S A T' sorts and occasioned relatively high-frequency appearances of Y-face letters in these sorts. Without checking the font in another book, the question of whether these wrong-face capitals intruded by replenishment or by fouling remains unresolved. The compositor(s) seems to have attempted to distribute correctly, since the proportion of Y1-capitals seems to decrease noticeably in later gatherings of Sejanus, but it must be borne in mind that the demands of the text on the affected sorts can vary significantly among gatherings. Hence, the decrease can be more apparent than real. The presence of the large proportion of a Y-face 'S' in Eld-S in Sejanus is an example. This 'S' is about 0.2mm shorter (2.65mm vs 2.85mm) than the correct-face letter with a diagonal stroke that is closer to the horizontal. The letter is found throughout the book, but primarily in speech prefixes. A check of later books reveals that, in fact, fouling was the source of the Y-face 'S'. A second potential discriminant in Eld-S is seen in the shorter asymmetrical Guyot 'S' in which the bottom counter is noticeably larger than the top and the diagonal bends tightly into the shoulder (see Low Countries, Fig. 203, p. 268). Its rate of appearance of about 3-4 per gathering in Sejanus (B1v:27, B3:24, B4:28, B4v:19) suggests fouling, and indeed, that is confirmed by its apparent failure to appear in later Eld-S books (Fearful News STC21511, 1606; A Brand Taken From the Fire STC5704, new STC5693.5, 1606) although these prose texts required fewer capitals and provide a marginally adequate sample. The appearance of this Guyot 'S' in Eld-Y1 (The Malcontent Q1 STC17479 [1604], B1v:24, C2:21, C2v:24) in mid-1604 and later in Sejanus suggests that Eld distributed a small batch of these letters into Eld-Y1 which fouled Eld-S along with the normal Y-face 'S' during the setting of Sejanus. The Guyot 'S' could have easily been purged from Eld-S because of its radically different style; however, it remained resident in Eld-Y1 where it is equally obvious. In contrast, there is clear evidence of replenishment in the 'A' sort. In addition to the foul-case Y-face 'A' (Sej, B2v:2,3,26), the sort includes a relatively new batch of S-face 'A' that print boldly with a sharply defined offset-right peak at the apex (Sej, B2v:2,11,21,27,30, B3:22,26,27); the older letters have a worn, rounded apex (B2v:7,8,13,26).

In the absence of obvious stylistic differences, variations in size provide the primary clue to the presence of wrong-face capitals. The general rule about cross-checking potential wrong-face discriminants against other fonts is especially relevant here, since early pica typefaces exhibit a normal range of variation of height in the capitals and ligatures which could be mistaken as potentially significant.[14] For example, a survey at high magnification of the capitals in Eld-Y1 could lead to the false impression that large-scale replenishment in the capitals produced a varying range in height of about 0.3mm, which, if unique to Eld-Y1, would certainly be a powerful font discriminant ('K' = 2.4mm, 'G' = 2.75mm). A similar but smaller variation in


105

Page 105
height can be seen in the capitals and ligatures in S-fonts amounting to about 0.15mm. S-face capitals and ligatures are taller than their Y-face and Guyot counterparts. In all instances, these variations are integral to the sets of punches from which the typefaces derived. The precise measurements of height variations in originals are somewhat affected by printing effects and four centuries of paper deterioration, but can be confirmed as integral to a typeface in a unique specimen of the Claude Garamond pica seen in Stanley Morison's John Fell, the University Press and the 'Fell' Types (p. 133). The specimen is an actual impression (not a reproduction) produced by types newly cast in the Garamond pica matrices in the Fell collection and composed by hand (and so for the rest of the book). As such, it provides a most accurate record of an early typeface unaffected by damage or deterioration.[15] Variations in size were also introduced during the copying of an original letter by a particular punchcutter hired to replace damaged or worn punches. Given the striking similarity in size variations of capitals and ligatures among many pica fonts of the period, it seems likely that they all derived from just a few sets of punches. Nonetheless, height variations in capitals and ligatures can provide useful discriminants. For example, the S-face 'I' in Eld-Y1 exhibits greater height and exaggerated serifs, but it is often indistinguishable with the naked eye from heavily inked normal letters. Its presence in nearly equal proportion to the normal 'I' is probably attributable to replenishment, even though fouling is a possibility because of the heavy strain on the sort in the highly reflective Montaigne's Essayes (STC18041, 1603). However, it is also present in two other 1603 books (Nero STC12551, and A Treatise . . . Antichrist STC7120) and may have been introduced into the font previously, since the printing sequence of the three books is unclear. The S-face 'I' remains resident in the Eld-Y1 throughout its lifetime. Similarly, Eld-S was replenished with Y-face 'I' early on. Later replenishment with S-face 'I' has the effect of diminishing the proportion of the Y-face 'I' in a growing population of correct-face letters. This combination in the 'I' sort provides a discriminant between Eld-S and other S-fonts such as John Windet's and Richard Field's, which use a single size of correct-face 'I'.

Miscast capitals and ligatures occur both as original abnormalities in a font and as products of replenishment. Antecedent to their appearance in a font is the history of the set of punches or sets of matrices from which they derived. Due to the nature of the type trade in the 16th century, a somewhat limited source of new type forced a homogeneity among fonts which sometimes levels potential discriminants from miscasting. Thus, a miscast letter is not necessarily unique to a single printer's font, since many printers were probably acquiring type from the same few founders during the period. The most obvious evidence of this consists of the small 'W', which is commonly the only version of the letter found in most pica fonts, and several miscast ligatures which are found in these fonts from the 1550's onward. However, given the number of possible combinations and proportions of normal and miscast letters across the upper- and lower-cases and ligatures, composites of same-face fonts usually reveal significant discriminants. "Miscast" is used here


106

Page 106
generically to denote several kinds of deviation from the normal that are attributable to mistakes at various stages in the type production process.[16] Technically, "miscasting" denotes a flaw in a type caused during the casting stage by the incomplete flow of molten metal into the impression of the letter in the matrix. An example of a possible miscasting occurs in Eld-S in the defective variant F2 which lacks the serif on the cross (Sej, B1:13, B1v: 14,31, B2:5, B2v:32 vs properly cast F1 at B1:29). However, a misstruck matrix could produce a similar defect: if the punch was not perfectly perpendicular to the face of the blank matrix at the instant of impact, a shallower impression would result on the periphery of the letter opposite to the direction of the punch's slant.[17] The commonness of the defective F2 suggests a misstruck matrix rather than type-caster error, since it appears also in Creede-3 (Romeo & Juliet Q2 STC22323 [1599], B3:9,23, C2:11 vs normal F1 at B2v:5,10, B3:24) and Creede-4 (The Regiment of the Church STC1827 [1606], H5:10); and in William White's mixed font (hereafter designated "-M") where most are defective (Love's Labor Lost STC22294 [1598], A2:1, A3:31,34, A4:26 vs normal F1 at A3v:32, A4:34). In contrast, the 'F' sort in (John) Windet-S and (Richard) Field-S is normal. The existence of both the normal and the defective letter suggests that at least two matrices fed the 'F' sort in these S-fonts. Other kinds of flaws can be attributed to earlier stages of type production. The cutting of a new punch could produce a difference in the width of the metal elements that form the face of the letter so that it prints bolder or finer than normal, as in the fine-printing variants E2 (LLL, B2:20,21,22) and S2 (LLL, A3:16,21, A3v:2,21,36) that appear in White's (White-M), John Danter's (Danter-M), Thomas Creede's (Creede-2, Creede-3), and other mixed fonts. Similarly, the justification of the strike by filing down its face could produce a difference in the height-to-paper relative to the other types in a font with the result that the abnormal type protrudes beyond the normal letters after pounding and lockup in the chase and prints boldly. Numerous examples of such bold-printing lower-case letters occur in the fonts surveyed here. Danter-M was replenished with several B-face lower-case letters which are not only obviously too large, but print quite boldly because of the difference in their height-to-paper (see below). Conversely, the impression of a letter in one strike might be shallower than in the strikes of the other letters; the impression could be deepened only by further striking with the same punch. However, if the set of strikes was purchased and then justified elsewhere (a quite common practice), the letters produced from the strike would print lightly because of the reduced height-to-paper. Furthermore, the variations in horizontal and vertical orientation of letters deriving from the same matrix suggest that a matrix could shift position in the mold during the opening and closing process for jettisoning a newly cast type.[18]

The most common and useful kind of abnormality consists of the misalignment of a letter in the vertical or horizontal planes so that it either leans or is positioned improperly at the baseline. For example, a strike could be taken with the punch so misaligned vertically that no amount of justifying the strike by filing its long sides could produce a matrix with a correctly


107

Page 107
aligned letter. Letters cast in the matrix would inevitably lean. A replacement strike from the original punch could be improperly justified at the ends by a new workman to produce a slight displacement from the baseline as established in the original set of matrices. Letters cast in the replacement matrix inevitably would ride high or low at the baseline when mixed in with letters cast in the original matrix.[19] However, this kind of evidence must be approached with caution. Since moveable types are free to shift in relation to other letters, especially when set without leading (a very thin strip of lead set between lines), numerous samples should be examined before concluding that a letter is improperly positioned in the horizontal plane. At times, the displacement is so exaggerated and pervasive that there can be no doubt. For example, the 'G' found in Eld-S in Sejanus and in other S-fonts is so low (0.25mm) that the baseline cuts through the white space inside the letter ("counter"). This low-rider 'G' appears foul-case in Eld-Y1 as a two dimensional discriminant: both wrong-face and miscast. The miscast E3 in Creede-3 (R&J Q2, A4:19, B1:4, B2v:18,21, C2:35) rides low (0.25mm) and leans right to the point that the adjacent letter sits on the tip of the bottom serif and is easily distinguished from the normal erect letters (B2:37, B2v:37, C1:2, D4: 14, D4v:14,22). The same miscasting appears in White-M (LLL, A3v:28) along with the fine-printing erect E2 noted above (B2:20,21,22). In contrast, all of the 'E' sort is erect in Eld-S and Windet-S. Creede-4 has about equal proportions of a normal erect E1 and an oversized E4 (3.05mm high) (Regiment STC1827, B1v:29) which positions correctly on the base-line. Variant D2 which leans left is quite common to S-fonts but occurs in differing proportions. It is the only version in Eld-S (Sej, B3:4,14,25) and Creede-4, seems to be the only S-face 'D' in White-M (LLL, A4:3, B1:29,36, B1v:10,11), and seems slightly in the majority in Creede-3 (R&J Q2, leaning: A3v:19,21, B1v: 20, D3:16; normal D1: A3v:34, B1v:29, C3v:8, D3v:3). Normal D1 is exclusive in (Simon) Stafford-EF. A fairly uncommon miscast T2, in which the thin cross slants downward to the right, seems in the majority in White-M (LLL, A2:10,12,13,24,25), while the sort in Creede-3 is evenly distributed between the normal T1 (R&J Q2, A3v:1,2,3, A4v:5,6) and slanted cross T2 (A4v:14, 20,29); it is infrequent in (Valentine) Simmes-S (Richard III Q1 STC22314 [1597], E4v:19,26), but dominates Danter-M (Titus Andronicus STC22328 [1594], C1:22) along with the foul-case 68mm 'T' (C1:16), and does not appear in Eld-S, Windet-S, Field-S or Creede-4.

The divergence in same-face fonts caused by acquisition and replenishment from different sources at different times is suggested by the combinations which occur. Eld's low-rider 'G' appears along with low-rider 'B C F K L M P R T', but correctly aligned 'I', in Creede-4 (Antichrist, STC7120, [2]H—M; Monsieur d'Olive STC4983 [1606], A, B(i)). Some of these letters appear in Eld-S, but are in closer alignment with normal letters, suggesting that the two original fonts were purchased from separate foundries, one of which used several misjustified replacement matrices to produce Creede-4. Eld probably replenished from this foundry, but the misjustified matrices were, by chance, more closely aligned to the matrices used by the first founder.


108

Page 108
On the other hand, the 'G' is normal in White-M (LLL) and Creede-3 (R&J Q2). Windet-S and Field-S, in contrast, use all non-deviant capitals, suggesting a single uniformly justified set of matrices at one foundry. It is frequently impossible to determine whether a letter is miscast although repeated examination creates an impression of a distinct abnormality. A small portion of White's 'C' sort rides low (0.2mm) (LLL, A3v:26, A2v:29, A2:8) while most are normally positioned (the Creede-3 sort is similar), but the effect is not so pronounced as to imply miscasting, since the baseline is normally distorted in this mixed font. Yet the fact that some print this way is a potential discriminant, since all of Eld's are aligned correctly. Moreover, the perceived difference may simply result from replenishment, since new clean-printing letters are bound to look different than old worn versions; the resulting layers of newness also are a potential discriminant.[20]

In addition to misalignment in the vertical and horizontal planes, the two letters forming a poorly cut ligature can be improperly oriented with respect to each other. These are especially prevalent in S-fonts, where variants occur in which one letter leans and produces a narrower gap between the letters at the baseline, or one letter is above the baseline. The skewed letters of ligatures make clear the difficulty of cutting a punch with perfectly aligned letters.[21] S-fonts seem to vary in the number of types available in their ligature sorts. Eld-S, Simmes-S, Field-S and Creede-4 almost exclusively set an 'ſſ' or 'ſſi', while Windet-S frequently resorts to alternate settings with combinations of 's', tall-ſ, 'ſi', and 'i'. However, the 'ſſi', 'ſ ſi' and 'ſſl' ligatures occur at low-frequency so that locating one in a text is difficult at times. But the miscast 'ſt' with the leaning 't' is so pervasive in S-fonts that the discriminating factor usually is the proportion between new, bold-printing (ſt2) and older-appearing ligatures (ſt1). In Sejanus, two old letters can be seen at C1:11,12 among a dozen new samples. Damage usually consists of a flattening or distortion of the curve of the shoulder of the 'ſ' and can be seen in the "new" letters at C1:3,30. In contrast, five new letters (LLL, A4v:18,30, B1:4, 10,34) are found among nine damaged in White-M, and the two are about equal in Simmes-S (Mal Q3, C2:7,8). A crisper, deeper impression in the matrix(ces) probably accounts for the difference in weight and distinctness of the two 'ſt', as must be the case with the 'ſſ' ligature. Heavy inking on the inside of the left 'ſ' seems responsible for a bold apparent variant ſſ2 with the left 'f' curving inward at the baseline (White-M, Titus, B4:12). The left shoulder often prints as an oblique straight-line. The consistent inking effect suggests a variant punch although there is no measurable difference. Some erect, normal appearing ſſ1 can be found in most S-fonts (Eld-S, Sej, D1:28; Field-S, Arte of Poesie STC20519 [1589], D4:25,26,35) along with the apparent variant (Sej, C3v:31; Arte, D4:27). A single punch doubtlessly produced all these 'ſſ' ligatures. The 'ſh' appears in S-fonts in a normal version ſh1 along with two common miscastings. In ſh2, the 'ſ' and ascender of the 'h' lean right in varying degrees, while the right stem of the 'h' is erect; the foot of the 'ſ' is above the baseline (0.15mm) while the feet of the 'h' are generally level at the baseline; the body of the type frequently seems to lean right. In ſh3, the


109

Page 109
'ſ' and right stem of the 'h' are erect, the ascender of the 'h' slants to the right, and all three feet are even on the baseline.[22] The frequent juxtaposition of the leaning ligatures, particularly the 'ſh', with round letters (o e) in combination with inking of the base-serifs often creates an optical illusion that the left and the right foot ride respectively above and below the base line, although at high magnification this effect can be discounted. It is worth noting that Braddock-Y1 was replenished with a miscasting similar to ſh2, quite possibly the same (MND Q1, D1v:26), which perhaps was part of the purchase that introduced the S-face 'ſt' noted below.

The letters of variant ff1 are correctly aligned (spaced 1.15mm), and the long right shoulder terminates horizontally in a small pear and extends 1.15-1.2mm beyond the right 'f'; the right shoulder of miscut variant ff2 curves into a sharp tip and extends 1.05mm beyond the ascender; the broadly curved left shoulder separates the misaligned letters 1.2-1.25mm (at 0.4mm above the x-line) which then converge to a spacing of 1.1-1.15mm at the baseline; the left 'f' is slightly higher in most. Misjustified matrices contribute low-riding miscastings of both variants (ff1: Danter-M, Coblers Prophesies STC 25781 [1594], C1:7, C4v:32; LLL, B2v:11, C2:27, C3:32. ff2: LLL, B2v:11, C2:27, C3:32) as well as correctly aligned ligatures (ff1: Sej, C3:20, C3v:4; Creede-1, Selimus STC 12310a [1594], D1:32 "off"). ff2: Sej, C1v:22; Simmes-S, R3, B3:35, D2v:31; Creede-2, Selimus, D2:32 'effeminate"). Both variants that appear in the 'fl' sort are miscut with the letters converging toward the baseline. The two are easily distinguished by height (fl1: 2.75-2.85mm, see Sej, B4:24, C4:17. fl2: 2.9-3.05mm, see R3, E3v:35, E4:6; Read-S, Cynthia's Revels, STC14733 [1601], B3v:6,14,30,35); the misalignment is less pronounced in fl1 (above x-line: 1.2mm, baseline: 1.15mm) than in fl2 (1.25mm, 1.1-1.15mm). A third variant with correctly aligned letters appears in Creede-1 and White-M although it may not belong to the S-face set of punches (fl3: 2.7-2.8mm high, 1.15-1.2mm wide; see Selimus, C4v:23 "flight", C4v:24 "flies"; LLL, G3:7). The right-leaning 'i' of miscast variant fi2 (2.85-2.9mm high) produces a width of 1.2mm at the upslanting cross and 1.05 mm at the baseline; the 'i' often appears to ride high and the ligature frequently leans right (Danter-M, Coblers, C3:5; White-M, LLL, A4:2, Fools, D1v:20).

An 'ſt' with right-leaning 't' (ſt1) is common in Y-fonts, although the lean is not so pronounced as in the S-face ligature since the portion of the 't' below the cross is closer to the vertical. Short-Y contains a large proportion of the ligature in which the deviation is so slight that the 't' appears erect (R3, H2:26,29,33), although the shoulder over-extends the 't' so that the top-half of the 't' leans right (ſt1b). Eld-Y1 includes a small proportion of this version (Mal Q2, C4:20,21, C4v:7). A second version in which both letters lean right and the 't' rides high (ſt2) is also common to Y-fonts, appearing in low proportion in Short-Y (R3, H2:13,37) and Eld-Y1, but in the majority in Braddock-Y1 (MND Q1, D3:1,15,27,31); the latter was replenished with a significant number of S-face 'ſt' which is quite obvious because of size (MND Q1, D3v:6,21, F3v:28,29). The 'ſt' sort in Braddock-Y2 is split evenly between the leaning 't' (ſt2) (Antichrist STC7120, [1]B3v:1,11,18) and partially erect


110

Page 110
't' (ſt1b) ([1]B3v:27,32), both of which equal the 'G' in height. Neither the Garamond nor Guyot 'ſt' with the tall 't' and right-angle shoulder junction seem to appear in Y-fonts. Common miscastings occur in two other ligatures. A miscast ſi2 with right-leaning 'ſ' (or left-leaning 'i') is the majority variant in Eld-Y1 (Mal Q1-2, E2:5 "pleasingly" vs normal ſi1 E2:4 "sinne"), in the minority in Purfoot-Y, almost exclusive in Field-Y (Observations in the Arte of English Poesie STC4543 [1602], A7:8 "sillables" vs ſi1 A7:7 "consisting") but less frequent in Braddock-Y1 (MND, D1v:17) and -Y2, and rare in Short-Y. A miscast fi2 with a high-riding 'i' and the cross accordingly slanted upward likewise is almost exclusive in Field-Y (Observations, B1:6,8), of about equal proportions in Eld-Y1 (Mal Q1-2, E3:5) with the normal version (Mal Q1-2, E2:3,25), and uncommon in Braddock-Y1 and Short-Y.

Although samples of most lower-case letters are available in large quantities in a text, their value as discriminants is somewhat undercut by their size and penchant for sustaining damage and distortion through extended usage. Generally, wrong-face upper-case letters are obvious to the naked eye because of stylistic differences, but identifying miscast and wrong-size lowercase pica letters is difficult if not impossible except at high magnification, since the measurable differences often amount to less than 0.1mm. Inking effects easily obscure this magnitude of variation, especially when new types ink more boldly. Read-S provides examples of replenishment with obvious Y-face letters in several sorts. Graceful Y-face 'a' variants distinguished by the shoulder, which extends to the left of the bowl, appear in varying proportions, accounting on some pages for about 30% of the total (Revels, B3v), and in lower proportions on others such as I2, where six (I2:4,20,23,24,26,31) are found among 72 older S-face 'a', reflecting the random recurrence of low-level replenished letters in page-length textual units. The Y-face 'k' variants and squat S-face k2 (junction height: 0.75mm) appear in roughly equal proportions, as on B3v where the ratio is 5:6, and both are in good condition, suggesting proximate replenishment without concern for stylistic integrity. Some pages show a few older S-face k1 (junction height: 0.65mm) (E1v:9,14) with worn serifs. The ratio of Y- and S-face 'p' is about 1:2 (C2v, 6:11). The fact that wrong-face letters occur in just a few lower-case sorts is a useful discriminant between Read-S and mixed fonts such as Danter-M and White-M, which exhibit variants in most sorts. The frequent skewing of type also is a pivotal factor in lessening the value of lower-case letters as discriminants and increases the need to cross-check against other fonts in order to establish whether a given letter is actually abnormal. A survey of many pages of Revels such as B3v, for example, reveals that a vertical shifting of type during the locking-up of the forme skewed many lines and as a result moved many types above and below the distorted baseline. Judgments about specific types are thus rendered difficult. In many instances, examination of such letters at high magnification in originals reveals that they actually are positioned correctly on the baseline, and the illusion of horizontal displacement is caused by inking effects. Read-S was replenished with miscast and oversized letters in several sorts. The 't' sort is about equally distributed between normal, worn


111

Page 111
letters (t1) that lean slightly to the right (height: 1.8-1.95mm depending on wear) and a new oversized, usually erect t3 (height: 2.3-2.45mm) which positions correctly and is distinguished by an accentuated hooked terminal at the baseline (see Revels, B4v:16 "that" and B4v:22 "but the" for juxtaposed samples). The new type inks more boldly, emphasizing the size difference and spreading the ink-smear below the base-line, although the foot of the letter is actually positioned correctly. The same t3 appears in White-M, where it actually does ride low (0.2-0.25mm) and leans right (LLL, A2:7 "then", A2: 18,19). In contrast, the 't' sort of Creede-3 includes a normal, erect t1 (R&J Q2, A4:4 "it") and a normal-size replenished 't' with a severe right-lean (t2) that, even with heavy inking, is correctly positioned on the base-line (A4:4, "part"). This letter seems slightly taller than the normal because the top has not yet been worn down.

In general, discriminants consisting of miscast and wrong-size lower-case letters are limited to a few sorts in pica fonts, partly because of the limited sources of new types, and certainly because of the difficulty of detecting abnormal letters with any certainty. High-frequency letters such as 'c e l r s' seem useless as discriminants because of the absence of inherent distinguishing features. The cross of the S-face 'e', for example, appears to rotate through a range of about 8 degrees above and below the horizontal, suggesting a potential discriminant. However, the simultaneous appearances of 'e' with varying degrees of rotation in fonts such as Creede-1 and Creede-2 (Selimus STC 12310a) indicates a misstrike, misjustification, matrix-wobble during casting, or common damage.[23] Variations in inking, paper-stretch, and wear frequently obscure the stylistic differences in actual variants and abnormal letters and can be resolved only at high magnification in originals. For example, S-fonts commonly have three variants in the 'n' and 'u' sorts which differ in newness, height, width and weight (boldness). The height of vertical letter elements, particularly serifed ascenders and stems, varies according to the amount of wear and the kind of damage a letter has experienced so that dimensions must be expressed as a range of values (as in the following examples). Variant u1 (1.05mm wide, 1.7-1.8mm high) and the tall u2 (1.1mm wide, 1.9-2.0mm high) frequently appear identical because of these factors. The u2 at Revels B3v:22 inked lightly just to the edge of the top-serifs with no excess ink to blot upward as the wetted paper stretched around the top-serif during impression; in contrast, the short u1 at B3v:23 ("would") inked heavily and exhibits the stretch effect. Both types exhibit the common distortion caused by beating the type during lock-up: the left serif is depressed into the type because of the absence of support and the stem is bent inward slightly (see u2, Fools, F1:16, vs new u2, E2v:17) while the right serif is pushed upward rather than depressed because of the support provided to the malletblow by the left stem. As a result, the top of the letter slants upward, with left-stem heights of 1.7mm (u1) and 1.8mm (u2) and right-stem heights of 1.8mm (u1) and 1.95mm (u2). The printed images of the two 'u' variants (B3v:22,23) are nearly identical in height (1.85/1.9mm vs 1.8/1.95mm) and indistinguishable with the naked eye. Variant u3, which is exclusive in Eld-S


112

Page 112
(Sej, B3:2,3), is slightly narrower (0.95-1.0mm) and of normal height (1.7mm), but when encountered in the context of other variants and the uneven inking and printing characteristics of a mixed font such as White-M (LLL, A3:4 "but", 24 "but"), it is easily confused with a lightly inked worn u1 or the tall u4 (0.95mm wide, 1.9mm high) (LLL, A3:2 "studie"). Similarly, inking and paper-stretch effects frequently confuse medium variant n1 (1.1mm wide, 1.65-1.75mm high) and bold, wide n2 (1.15mm wide, 1.7-1.8mm high), but since both commonly appear together, their value as discriminants is lessened. However, both can be confused with two miscastings that are valuable. Bold variant n3 is wider at the baseline (1.15mm vs 1.05mm) because the left stem leans (Revels, B3v:2 "and"; LLL, A3:15 "blinded"), while the left stem of n4 bows outward to 1.15mm at the center (Sej, B3:4 "mine", 11 "unto") to appear wider although the separation of stems is 1.1mm at the baseline.

The 'w' especially is prone to creating an illusion of extra height and low positioning, since the paper stretches along the pointed apexes and down into the type-shoulder during impression. About 30% of Read's 'w' sort is oversized (height: 1.9mm) and usually rides low (w2) (Revels, C2v:14,15,17), although a far greater portion merely seems oversized. The same is true of White-M, Creede-3 and other fonts in which the w2 variant occurs. A condensed "tight" 'w' (width: 1.55mm) (Revels, B1v:16, B3v:26, C1:18,32, I3:26) that prints with distinct apexes and lacks a center serif is difficult to decipher and illustrates one kind of problem associated with pica lower-case letters. Since no equivalent variant appears in contemporary S-fonts, the likely conclusion is that it should be dismissed as undefinable. However, it is a genuine variant that appears in an S-font used by Henry Bynneman some thirty years earlier (The Enimie of Idlenesse STC11476 [1568], prefatory letter, [unsigned verso]: 8,10,12,13,22). By 1579, the condensed 'w' had been replaced by the normal S-face letter. However, it remains in John Allde's S-font as late as 1580 (A Warning for the Wise STC5259, C4:4 etc.). The newer majority w2 in White-M always rides low (LLL, A2:13) and appears along with an alternate-face 'w' (A2:23) as well as low-frequency remnants of the Guyot short second-stem 'w' (A2:14, A3:10). Creede-3 uses a distinctive w3 (introduced via Creede-2) in which the outer stems print finely and seem to bow outward, with serifs inside the stems (R&J Q2, A4v:2,13); it appears oversized in some instances (A4v:16 "swoong"). This w3 occurs in White-M and Stafford-EF as well as in S-fonts and is integral to the EF-face. In contrast, Eld-S and Windet-S lack the 'w' variants found in the above fonts. Overall, the comparison of these letters in several S-fonts is necessary to distinguish miscast letters from normal letters whose appearance is modified by inking and impression.

It seems that useful miscastings are even more limited in Y-fonts. Two variants of 't' were introduced into Eld-Y1, including a leaning low-rider t2 (Essayes STC18041, Yy1v:3 "it" "their"; Antichrist STC7120, [2]A2v:2,3,4) and an erect t1 which positions correctly (Yy1v:9 "wittily"). By mid-1604, the proportion of the leaning version diminished (Mal Q1, B4v:14 "witty" and B4v:16 "vnutterable", where it is juxtaposed to quite normal letters) with


113

Page 113
further replenishment by the normal t1. Braddock-Y1 (1598) and -Y2 (1603) show a roughly equal distribution between erect t1 and a leaning t2, both of which position correctly (MND Q1, D1:7,8,9; Antichrist STC7120, [1]B2:5); Purfoot-Y1 is similar. Short-Y includes a low proportion of t1 (R3, H1:4,13, 17) and a very few low-riding t2 like Eld's that generally print more boldly (R3, H1:15,32, H1v:12,23,37). The 'w' sort includes four variants in Y-fonts. w2 is taller than normal (Short-Y, R3, H2:14,15,16). The left stem of w3 is taller than the right so that the top of the letter appears to slant downward to the right (Eld-Y1, Mal Q2, C1:1,12,22). This variant is probably attributable to a misstrike with the punch leaning to the left and producing an impression with uneven lateral depth. It is difficult to explain the w4 variant in which the right 'v' is quite short (1.4mm vs a normal 1.7mm) (Eld-Y1, Fools, B4v:15,32). A severely damaged punch seems the only possibility.

Finally, rather high concentrations of wrong-face and wrong-size punctuation (especially in Y-fonts) present a confusing situation. The distinction between replenishment and fouling is academic here since all printers had complementary fonts as a source of foul-case punctuation marks in black letter, italic, and larger faces. However, the appearance of several sizes of comma in Braddock-Y1 and Short-Y suggests deliberate replenishment, at least in regard to the size normally found in 96mm fonts; the 108mm variants are so grossly oversized as to suggest fouling. Commas are especially difficult to work with due to paper stretch and the effects produced by medial vs terminal justification. As a result, the 96mm and 108mm commas frequently cannot be reliably distinguished from each other, although they are usually distinguishable from heavily-inked normal commas.[24]

III

Despite the overall uniformity of a typeface, stylistic variations in some minor aspects of the design of a letter, the width of a letter, and serif formation occurred which provide a category of internal discriminants for distinguishing same-face fonts. These divide into two groups, depending upon whether the stylistic variation is obvious to the naked eye, or whether examination and measurement at high magnification is necessary for discrimination.[25] The latter is generally true of Y-face variants and S-face 'k' variants. In instances such as the S-face 'n' and 'u' noted earlier, a difference can be sensed because of the heavy inking of these letters, but both widths randomly appear bold, or as in the case of the 't', the actual horizontal positioning is confused by heavy inking of the underside of the foot. When the 'n u' appear amidst other wrongface bold variants as in Danter-M and Creede-3, the difficulty of sorting them out with the naked eye is aggravated.[26]

In the S-face, the vertical and oblique stress along with accentuated serifs render the diagonal letters 'x y z' especially susceptible to illusions. Variations in inking and damage to the stems and serifs of the two 'y' variants often suggest stylistic differences, but the variants are identical except that the right diagonal of "tiny" y2 ("Le Be" specimens 12, 13) is shorter (2.55mm vs 2.7mm). Both exhibit an enclosed angle of 45 degrees, a heavily stressed left


114

Page 114
stem and a quite fine right diagonal terminated in an offset-left ball, with top-serifs extending 0.25mm on both sides of the diagonals. y1 usually rides correctly, although miscasting sometimes causes a left-lean so that the right top-serif extends above the x-line, while y2 seems always to ride high (as much as 0.4mm: Creede-1,2, Sel, D1:11,12, D1v:4; Simmes-S, R3, C4v:2). S-fonts use both wide x1 and expanded x2 variants that are frequently indistinguishable without magnification because of wear, damage and inking of the accentuated serifs of the expanded variant. The left top- and base-serifs of many expanded x2 in Field-S fail to ink on the outside, making them appear narrower (Arte, E3:22, "extraordinary" vs "extol-"), although the diagonals cross at a right angle. A faulty strike from the x2 punch produced this defective x3, which is quite common and seen in Creede-3 (R&J Q2, B1v:31) and other S-fonts. However, both the wide x1 and x2 appear in Simmes-S (Hamlet Q1 STC22275, E1:27,34; Mal Q1, F3:4 "extinct", F3:7 "fixt") and are differentiated by the diagonal crossing angle (x2,3: 90 deg., x1: 72 deg.). Two common variants of 'z' occur. Wide z1 (1.6-1.65mm wide across tips of serifs, diagonal: 2.4mm) usually prints fine (LLL, D3:19; Sej, C4v:13; Coblers, C2:13; Sel, D3v:7); expanded z2 (1.75mm, diagonal: 2.55mm) is squarish and bolder (LLL, A3:14; Sej, D2:14; Sel, D3v:15; R3, E4:35). Oppositely inked z1 and z2 are difficult to distinguish except at high magnification.

Other useful obvious S-face variants occur in the 'b f g h i k' sorts. Two apparent 'b' variants (bold: b2) probably can be credited to a difference in the depth of impression in the matrices which supplied them, since both exhibit the same acute enclosed angle of the bowl-top junction with the ascender and length of oblique fineline (0.7mm), and the same bowl-width (1.3mm) and separation between bowl-junctions on the ascender (1.3mm). The junction-angle of taller (3.1mm vs 2.9mm), wider (1.35mm) variant b3 is roughly the same, but the oblique fineline is longer (0.9mm) and the bowl junctions are separated by 1.45mm; the bottom of the bowl joins the ascender almost horizontally at the baseline, whereas it curves upward to form the junction in b1-2 (d'Olive, E1v:11 "best" vs "be"; Whore, D3:2 "ru b1 b3 ers"). Both d1-d2 are tall letters (2.9-3.05mm high) with a marked left counter and acute bowl-junction angles but differ in the separation of junctions on the ascender (d1: 1.15mm; d2: 1.25mm); the bowl-stroke of the narrower d1 (1.15-1.175mm) forms an inverted pear-shaped counter that often seems awkward; in d2, the stroke thickens as the bowl widens toward the baseline (top: 1.15mm vs 1.2mm). In both, the ascender broadens to a slanted termination sans top-serif (d1: LLL, A3v:10,11,13,14; Whore, A3v:5 "dinner", C3v:15 "heard", d2: LLL, A3v:33 "made", B4:14 "helde"; Whore, A2v:2 "behold", 9 "ashamde"). The thicker d2 almost invariably prints bold; oppositely inked d1-2 are usually indistinguishable (see LLL, B4:21 "d2 i d1"). New clean-printing samples of the tall (2.9-2.95mm) f1 variant exhibit a slightly stressed ascender which thins as it curves smoothly high above the cross (0.9mm) into the shoulder that broadens and terminates horizontally in a small pear (Sel, D1:15, D2: 4,5). The right half of the base-serif frequently bends downward in good letters (LLL, A2v:15, A3:9; R3 A4:4), a result either of damage or a misstrike.


115

Page 115
Flattening damage to the shoulder, downward bending of the pear, and heavy inking of the ascender and underside of the shoulder create a variety of appearances. However, a separate punch is probably responsible for variant f2 which is similar to f1 except that the bolder ascender bends to an oblique line which curves over at the h-line and downward into a larger pear (LLL, B4:5 vs B4:6). The uncommon shorter (2.65-2.75mm) f3 variant with a bold ascender and flat shorter shoulder (1.15-1.2mm) terminated in a slight pear dominates Eld-S (Sej, B1v:31, B2:22; Sel, H1:13,16); it probably is imported from a smaller face, also appearing in Eld-Y1 (Fools, B4:28). The narrower EF-face 'f' appears in mixed fonts (LLL, B2:34 "confesse") and also in S-fonts that have been extensively replenished (Simmes-S, R3, A4:10 "if"; Read-S, Revels, D4v:4). Two 'g' variants occur. g1 leans left because the body, link and loop are vertically aligned at the left, and exhibits a left counter and extra-long ear (Sej, B3:5; LLL, A3:7 "lyght"). The left lean is less pronounced in some letters, probably because of matrix-wobble. Creede-3 uses roughly equal portions of g1 (R&J Q2, A4v:3 "go",16,18) and the vertically aligned g2 with a smaller bowl (R&J Q2, A4v:6 "judgement") (see later discussion). Three 'h' variants occur. The smaller h1 (1.075mm wide, 2.9mm high) is common, while the larger h2 (1.125-1.15mm wide, 3.0mm high) is less frequent; the ascender usually seems to lean right although the vertical elements are parallel and, as with b2 and d2, h2 invariably prints bold because of the thick face elements (Simmes-S, R3, F4v:4; Read-S, Revels, B4v:4; White-M, LLL, B4:14 "helde", 15 "with", Fools, C1:4 "thriftie", I3v:25 "this"). The ascender (2.85-2.9mm) of miscast h3 leans right (x-line width: 1.05mm, baseline: 1:15mm), usually prints bold (d'Olive, E2v:9 "with", 10; Fools, I2v: 11 "have", E4v:24 "her"; Whore, A2v:17 "had") and is easily confused with the EF-face miscast (normal) 'h'. Two 'i' variants occur. In the common version i1, the dot is vertically aligned with the letter and is positioned at the tips of triangular top-serifs. In the fairly rare i2, the dot is to the left of the letter and below the h-line. Simmes-S and Field-S use the left-dotted i2 exclusively, while Creede-3 divides between the original i2 and the center-dot i1 from Creede-2 (R&J Q2, A4:21 "enemies"; A4:18 "villaine"); Creede-4 uses the common i1. Finally, three "squat" 'k' variants appear in S-fonts, but these are difficult to distinguish except at high magnification. The common k1 and k2 are differentiated by the height of the junction of the arm-leg link with the ascender (k1: 0.65mm; k2: 0.75mm); the arm and leg, oriented 45 degrees above and below horizontal, form a right angle and are tipped with accentuated serifs which, in k1, usually are worn to nubs and produce a narrower appearance. The shorter arm (0.65mm) of variant k3 is at a higher angle (52 deg.) and extends 0.9mm beyond the ascender (vs 1.0mm in k1-k2); the low (0.65mm) junction of the arm-leg link on the ascender is at an oblique angle rather than perpendicular (Simmes-S, R3, E1v:10, E2v:13; Field-S, Arte, D4v:1, E3v:3; Creede-4, Antichrist, [2]H1:35, H2:30). The low junction permits distinguishing k3 from the EF-face oblique-junction k2 (0.65mm vs 0.85mm).

Unlike the S-face, the Y-face exhibits considerable susceptibility to damage


116

Page 116
and wear which tend to obscure the stylistic features of letters. Routine battering and the beating of the type into place in the chase during locking-up of the forme wreaked havoc with lower-case letters, especially complex forms such as 'a k y' and some ligatures. The consequent changes in shape and inking caused by battering often create the appearance of variants when, in fact, the suspected variants probably derive from the same punch or matrix. This is also true of miscast letters which lean or displace from the baseline. The general principle that the form of a letter can be inferred from the combination of traits that recur consistently in most samples must, therefore, be applied with extreme caution when working from reproductions.[27] The diversity in Y-face letter appearance caused by battering and inking is no where more apparent than in the 'a'. Any page of text in old type seems to exhibit a plethora of variants, but these usually can be reduced to four or five at most by analysis of the shapes and orientation of portions of the letter.[28] The bowl's diagonal top-stroke in variants a1-2 is at a high angle (28 degrees above horizontal) and joins directly to the back with an enclosed angle that varies because of the differing back structures of the two (see list of samples in Appendix following 'k' variants). In a1, the shoulder stroke curves directly into the straight back which is usually perpendicular to the baseline, producing an enclosed junction angle of 62 degrees, although the right edge of the back gathers ink above the junction and frequently appears curved to the naked eye. In contrast, the shoulder stroke of a2 continues to curve slightly inward to the center of the bowl, then bends to the right and abruptly reverses direction at the baseline to form a "fishing hook" serif which terminates vertically. The enclosed junction angle is difficult to measure, since the straight top-stroke joins a curving back. The back of a1 curves broadly at the baseline to form a long (0.45-0.5mm) hook that points upward at about 35 degrees; it symmetrically balances the curve at the top of the back, but is frequently jammed inward to approximate the a2 hook or is blunted. Unlike the Fell and Le Be 'a', the oblique left portion of the shoulder strokes of both a1-2 are unstressed and thin imperceptibly (if at all); both are terminated with a pear (larger and longer in a2, but often blunted) that extends slightly beyond the edge of the bowl in new letters (0.05mm); damage to the terminations produces a longer, flatter appearance or a tightly curved shoulder that terminates short of the bowl's edge. Infrequent samples of both (15%) print a medial bend because of ink accumulation although the shoulder is a continuous curve. In both variants, the fine oblique top-stroke drops directly to form the oval left edge of the bowl and usually inks as a sharp angle; this contrasts with the Fell and Le Be bowls which are formed by a curving top-stroke that broadens into the bowl's oval edge. Finally, the bowl of a1 is slightly higher (0.65 vs 0.6mm) as measured between bottom and top junctions on the back; conversely, the peak of a2's shoulder is slightly higher above the top junction (0.75 vs 0.7mm). These dimensions remain remarkably constant despite the condition of the letters. The top-stroke of variants a3-4 is at a lower angle (22 deg.) and the final 0.1mm bends to form a perpendicular junction with the straight back. The dimensions of a3 are identical to a2 (bowl-height of 0.6mm,

117

Page 117
0.75mm top junction to shoulder-peak), but the base-serif is broad as in a1. The straight back is usually vertical but appears to lean slightly right in a small number of samples. a4 appears slightly taller because of a lower bowl (0.55mm, 0.8mm top junction to shoulder-peak), and the "fishing hook" base-serif is rather tall (0.55mm). The straight top-stroke curves into the oval of the bowl to produce a rounded appearance, but not nearly as much as in the Fell 'a'.

Given the susceptibility of the 'y' to damage, a wide range of appearances occurs, but just two punches supplied the 'y' sort. Variant y1 exhibits a wider apex angle (42 deg.) than y2 (38 deg.) The lightly stressed left diagonal accentuates its apparent width. The right diagonal is slightly longer than in y2 (2.7mm vs 2.65mm) and terminates at the bottom with a round ball, while the termination of y2 is oval and offset to the left of the diagonal. Flat top-serifs extend 0.25mm on both sides of the diagonals of y1; the outside of the left top-serif frequently is bent upward, and both serifs experience damage and wear which reduces the outside portion (y1: Essays, Yy1v:15 "your" [2]; MND, B1v:14 "euery"; Poetaster STC14781 [1602], B2:7 "day", 8; LLL, A3v:9 "say", 11 "day"). The top-serifs of y2 slant downward and extend 0.3mm on the insides of the diagonals; in new samples, the left top-serif has a very slight nub outside the diagonal which is quickly eradicated (y2: Essays, Yy1v:11 "by"; MND, B1v:12 "choyce", 13 "boyes"; Poetaster, B2:12 "thy", 13 "my"; LLL, A2v:2, A3:7 "whyle"). Both variants appear together in most Y-fonts, although y2 seems exclusive in Short-Y. Miscasting seems responsible for apparent variants of both in which the letter is rotated counter-clockwise on the apex. Damage to y2 produces an inward-bowing right diagonal both above and below the baseline.

The variety of appearances in the 'k', unlike most other sorts, actually indicates a large number of variant letters, all uniform in style with minor but consistent differences in structure and dimensions. Vulnerability to normal bending damage is limited to the area of the letter below the link which connects the arm-leg to the ascender. While other portions of the letter have relatively broad foundations on the type-body, the counter enclosed by the leg, link, and foot of the ascender eliminated internal support so that a "knee" bend frequently occurs in the leg, and the foot bends inward. Neither element ever bends outward. In contrast, bending damage is rare in the fineline arm which seems most vulnerable; the exposed arm-serif, however, normally experiences wear, bending, and blunting, often being reduced to a dot in old letters and sometimes bent in-line with the arm.

Three different size ranges occur. The height of the correct-size 'k' ranges from 2.75-2.85mm, depending upon the condition of triangular top-serif, which is rounded, blunted, or worn off in older letters. Anomalous undersized variants probably were cut for smaller faces (about 70mm bare 20-line height) and in most cases ride slightly high and usually lean slightly right. The sophistication required of a punchcutter in controlling a letter's printed appearance while working within nearly identical size parameters is evident in as much as these smaller 'k' always seem "tiny" despite the fact that the junction


118

Page 118
height (0.85mm), the length of the link (0.4-0.45mm), and the extension of arm and leg (0.85-0.9mm) to the right of the ascender duplicate the dimensions of correct-size letters. The new "tiny" 'k' at LLL A2v:13, for example, exhibits a normal height of 2.85mm, a junction height of 0.9mm, a medium length (0.5mm) link, and a quite short (0.35mm) high-angle (52 deg.) arm topped by a massive 0.95mm crescent. Yet the letter looks tiny amidst S-face letters in White-M, as does another 'k' in Read-S (Revels, C3v:29 "lookt"). The effect is similar in Y-fonts: the small 2.5-2.65mm 'k' variants with 0.8-0.9mm arm-leg extensions look tiny in Eld-Y1 (Mal Q2, B2v:7 "looke", C2:31 "harke"; Q3, H1v:15, H2:21; Fools, G1v:14 "knocking", G2v:23 "knowing"), in later White-M (Fools, C2v:25 "take", C4:16 "thinke", I1v:17 "drunke"), and Allde-Y1 (Granados Deuotions STC 16902 [1598], p. 23:17). The variety of undersized 'k' provide useful discriminants. For example, an anomalous 'k' with a low-angle arm (42 deg.) and long leg (0.95mm) which extends a full 1.0mm occurs in late Eld-Y1 (Fools, H1v:31 "locke") but apparently not in other Y-fonts. However, caution is required in assuming that a potential 'k' discriminant is undersized. Old, very worn normal 'k' variants with reduced height-to-paper and element width, such as those encountered in late White-M in Fools (CD, I), generally look like a 2.5-2.65mm letter even in originals, but measure 2.7-2.8mm at high magnification.

At the opposite end of the scale, 2.95mm 'k' variants with correspondingly high (0.95-1.05mm) junctions appear in Y-fonts. The Braddock-Y1 tall (2.95mm) variant k7 is a scaled-up version of normal elegant k1 differentiated by a higher junction (0.95-1.0mm), longer link (0.55mm), shorter (0.55mm) high-angle (48 deg.) arm, and long leg (1.0mm) which extends beyond the arm (0.9mm vs 1.0mm). The letter as often rides low and leans right (MND, C4v:8 "makes", D2:11 "talke") as not (D1:16 "quake", D2:1 "looke") and appears in Read-S (Revels, I3v:3 "thinke" etc.). The apex of the triangular top-serif of the tall variants is commonly sheared or blunted and prints with a flat top. The intermediate stage of the process is evident in the letter at MND C4v:24 ("skill"), where blunting has flattened the apex of the triangle and pushed the apex metal into a small spur (0.1mm) on the right of the ascender. This peculiar form of damage is uncommon in other 'k' variants where the apex of the top-serif is usually rounded by wear. The effect of newness on apparent size is obvious when new "elegant" k1 are visually compared without magnification to the tall variant k7. A new k1 in Braddock-Y2 is directly beneath a 'p' (Poetaster, D4v:7/8), a setting which totally eliminates paper-stretch effects and permits an accurate measurement of the height of 2.8mm. Other samples measure up to 2.9mm (B2:3 "workes", B3v:4 "broken"). A folio prose setting such as Essays STC18041 apparently reduces overall paper-stretch since the new k1 in Eld-Y1 measures 2.75-2.8mm (Yy1v:19 "speake", 27 "talke" etc.). Furthermore, the new ascender element in both Braddock-Y2 and Eld-Y1 is wide with distinct edges not yet rounded by beating during lock-up. The same is true of other new Y-face 'k' variants noted in the Appendix.

A few simple variants appear in Y-fonts. Short's i3, l2, and t2 are cast on a narrow body and are left-justified so that the crammed spacing is obvious in


119

Page 119
various combinations of the letters (R3 Q1, H3v:8 "will", H3v:13 "it", H4v:4 "vnwillingnes", H4v:10 "enuie", "selfe", H4v:33 "shortlie", H2:24 "left"). The narrow 'i' dots at the tip of the triangular top-serifs of tall letters. Variants include a normally spaced i1 with high dot and i2 that dots slightly beneath the top-serif of tall letters (Fawne, D1v:15 "vile"). The difference is apparent with crisp new letters. The left-justified t2 appears also in Field-Y where the unsuccessful kerning with the 'f' is quite obvious (Observations, A8:16 "after"). Most Y-fonts use the normally justified letters. In general, the Y-face seems to include relatively few internal variants, suggesting a closely guarded source of matrices from the seminal set of punches. On the other hand, it seems probable that the punchcutter responsible for the S-face cut two sizes in the same style, given the striking correspondence of stylistic features in the normal b1, d1, h1 and the slightly larger but otherwise identical b3, d2 and h2; copying by a second punchcutter seems impossible in this instance.

IV

The process of mixing produced a class of unique fonts characterized by significant proportions of letters in two or more typefaces across both cases. The simplest form occurred within a shop when the printer mixed together the cases containing two fonts to produce a third. The rationale behind this kind of mixing is obscure; however, the fact that at least a few sorts in mixed fonts contain only letters of one face suggests a cost-efficient approach to salvaging the investment in the original font which, inevitably, would have been replenished in many sorts during previous use. By melting down only the scrapheap of discarded types and perhaps the severely depleted sorts toward a new font, a possibly significant cost-savings on the metal required for the new font could be realized, the investment in the salvaged sorts would continue to produce, and the addition of the new font would eliminate the need to replenish in the near future. A reluctance on the part of some printers to acquire new fonts is suggested by what we know in this respect. One extant record of font management reveals what seems to be an enormous amount of discarded type (40%) on the scrapheap awaiting melt-down.[29] Given the amount of type needed to initially fill two sets of cases, it would be possible to make-do with a quite depleted font by setting from both sets of cases at once rather than alternating from them or by transferring all type into a single set of cases.[30] However, this cost-savings rationale is purely speculative, and there seems to be no consistency among printers. Thomas Creede progressed through four fonts in 16 years (1593-1609); significantly, mixed font Creede-3 endured for half this period (1595-1603), suggesting perhaps that mixing enhanced the longevity of the font. Creede-4 served from 1603-1609, slightly less. Eld-Y1 was replaced by Eld-S after about 2½ years, and Eld-S by Eld-Y2 in less than two years.[31] Then again, Simmes-S was in use from the beginning of Simmes' career in 1595 to about 1606,[32] a remarkable but unfortunate fact since it should have been scrapped long before.

The in-shop kind of mixing is illustrated by the creation of Creede-3 by the mixing of Creede-1 and new Creede-2. The significance of W. Craig Ferguson's


120

Page 120
discovery of the initial mixing process as it progressed in a single book cannot be overstated,[33] for it provides crucial evidence of the independent existence of the hybrid set of matrices that produced Creede-2. Creede-1 is an S-font with normal variant letters throughout, including the squat k1, left-leaning g1, left-dot i2, expanded x2,3, and the miscast ſh3 and fl1,2, and erect fl3, ligatures. However, Creede-2 is in itself a mixed font, and therein lies another tale. Overall, S-face capitals, ligatures, and the majority of lowercase letters form the base of Creede-2, but stylistically uniform, slightly smaller letters fill the remaining sorts: a 'b' with a right-angle bowl junction at the x-line, a shorter oval 'd' and medium oval 'p'; erect S-face g2; center-dot i1, two variants (medium and wide) of 'k' characterised by the nearly vertical arm (74 deg.) and high junction (0.85mm) on the ascender, w3 with bowed outer stems, high-riding "tiny" y2, the EF-face high-junction 'ſh' with right-leaning ascender, and a very few smaller 'ſt'. Once mixed, Creede-3 exhibits both S-face and alternate-face letters in these sorts; later replenishment with wrong-sized, bold letters and miscastings complicated the relatively simple two-face composition of Creede-3, thereby obscuring its origin in two sets of matrices: a uniform S-face set (Creede-1), and a hybrid set with S-face matrices and the Creede-2 alternate-face subset (Creede-2). Depletion of some sorts in Creede-1 seems the probable cause of the mixing; for example, the Creede-1 k1 accounts for only about 15% of the sort in Creede-3. A second example of direct evidence of a hybrid set of matrices occurs in Nicholas Okes's mixed font purchased June, 1607 (noted earlier). Original Guyot capitals (Low Countries, Fig. 203, p. 268) accompany S-face ligatures including the common miscastings (excepting? the 'ffi' and 'ſſi') and the S-face lower-case letters, including both the left-leaning g1 and the erect g2 as well as other variants. Similarly, Adam Islip's mixed font (Every Man out of his Humour STC14767, 1600) uses Guyot capitals exclusively, while the lower-case combines S-face letters with most of the EF-face subset of letters and ligatures. Similar mixtures are seen in other fonts.

These three examples provide insight into one kind of mixing that in all probability occurred at the type foundries which supplied mixed versions of these fonts to Creede, Okes, Islip, Allde, Stafford and other printers. For whatever reason, the founder(s) combined partial sets of matrices from two sets of seminal punches (S-face and alternate-face) to produce Creede-2. Significantly, the S-face erect g2 is integral to Creede-2. The slightly smaller bowl and vertical counter of this variant absolutely indicate a duplicate punch rather than a misaligned or a misjustified strike from the normal S-face left-leaning g1 punch. The presence of both variants in clean S-fonts is further evidence of duplicate punches in the seminal set.[34]

Once Creede-2 and Creede-3 are compared with other mixed fonts, the most intriguing aspect of font history emerges: how and where in the type production and distribution processes particular combinations of dissimilar-face letters became mixed in groups, creating mixed fonts with extraordinary similarities and yet striking differences in composition. The differences probably can be explained by analogy with Creede-3 where replenishment after


121

Page 121
mixing contributed variants found in neither Creede-1 nor Creede-2. Other mixed fonts appear in the advanced state of Creede-3 so that replenishment has somewhat obscured the composition of the original mixed font as purchased or of the two fonts that were mixed. As a result, the distinction between mixing and replenishment can sometimes be resolved only with reservation, since piecemeal replenishment can plausibly produce what appears to be a mixed font. That possibility is hinted at in Danter-M, which apparently was replenished at one time with a large subset of crudely cut, poorly shaped B-face letters (a b c e h m n o p r t u) that leap from the page because of their sheer bulk.[35] Danter obviously was not overly concerned about the appearance of his work. The minor proportions of this subset clearly indicate replenishment which, if repeated several times, could tip the balance in favor of mixing given the quite large number of affected sorts. However, even if that occurred, replenishment could be inferred from the general principle that a genuine mixed font contains letters from just one typeface in some sorts, assuming that replenishment has not occurred, as in the subset of Creede-2. Although the B-face subset in Danter-M is quite extensive, it duplicates pairs of letters in the sorts that represent the pre-replenishment state of the font. On the other hand, White-M contains duplicate S- and Y-face letters in all sorts, a strong indication of in-shop mixing. Similarly, the Y-face subset in Read-S is limited to a half-dozen sorts, all containing S-face letters as well, an obvious indication of replenishment.

The fact that mixed fonts are composed of the S-face and at least three other typefaces underscores the difficulty of sorting them out relative to their sources in identifiable seminal sets of punches. Except for a few instances of replacement or duplicate punches, it seems clear that the composition of seminal sets of punches endured unmodified. Hence, the stability of certain combinations of alternate-face letters suggests an ongoing process at various foundries of mixing strikes or matrices from several sets of punches. Strictly speaking, "typeface" denotes a stylistically uniform set of seminal punches and thus must be re-defined to include such stable combinations of letters as seen in Creede-2, Stafford-EF or Windet-F, where the "C2-face", "EF-face" or "F-face" label denotes a hybrid set of matrices that supplied two or more printers with a particular combination of stylistically heterogeneous letters from alternate typefaces. Ultimately, it may be impossible to relate a particular subset of alternate-face letters to a distinct set of punches and define the seminal "typeface" in a traditional sense. One problem is the difficulty of locating a specimen in which the subset appears along with the remaining letters from the seminal set of punches. A composite, as a whole, defines not only the stylistic traits of individual letters, but also the combination of letters which comprise the typeface; the design of a particular letter is clarified by juxtaposition with other letters (i.e., the cross-strokes of the 'e f t' provide reference points for judging the height of shoulder-ascender junctions). The certainty that a given letter belongs to a typeface grows in proportion to the number of letters in which the design and reference points coincide. The work of sorting out mixed fonts lies in the future and should shed valuable light on


122

Page 122
the production and distribution network that supplied type to Elizabethan printers. Nonetheless, the varying compositions of mixed fonts provide a plethora of discriminants for distinguishing them in books of the period.

V

The process of font analysis aims at detecting discriminants which differentiate two or more fonts. Since White-M and Stafford-EF combine extensive S-face bases with an alternate-face (Y-face, EF-face subset), they contain far more discriminants than typical same-face fonts, a situation which, in combination with their worn condition and the poor printing quality in Fools, makes them ideal for illustrating the problems encountered in the typical search process which employs reproductions for the initial stage of comparative analysis. Font analysis, as a method, benefits from a comparison of perceived stylistic variants in a font with similar variants in other fonts in order to establish that they are genuine and historically demonstrable rather than transient effects of wear, damage, inking and printing. Ideally, variants could be verified by reference to an extant founder's specimen of a typeface so that any transformations in composition could be easily detected.[36] In practice, the appearance of a typeface in other books usually is the only kind of specimen available for comparative analysis, although these can reflect a different state of the typeface occasioned by practical circumstances.[37] Although neither White-M nor Stafford-EF correspond to any extant specimen, a preliminary examination of the two fonts in Fools CD, I, and EF reveals capitals, lowercase letters, and ligatures that obviously combine an S-font with smaller alternate-faces quite similar to the Y-face, suggesting that specimens of the two faces in other books (or other sections of the book) be used as a reference point during analysis. Similarly, Guyot-style letters ('G', short second-stem 'w', 'g k p q z fi') call for reference to the Guyot pica specimen in Low Countries (Fig. 203, p. 268) or books by Henry Bynneman or George Bishop. Previous study of fonts in other books can also suggest reference specimens; in this instance, similar fonts had been noted in Whore EF (Simmes as primary printer, others unidentified), Monsieur d'Olive EF (Creede, sharing unnoted), The Seven Deadly Sins (STC6522, 1606, Edward Allde), Every Man in his Humour (STC 14766, 1601, Simon Stafford) and Lymbo-Mastix (STC19295, 1607, George Eld).

The analytical process progresses through three stages that vary in scope and precision: (1) an initial seriatim survey of the book for discriminants consisting of obvious wrong-face and stylistically variant letters and punctuation; (2) a more complete and precise survey of each possible section of a book aimed at detecting subtle stylistic variants and devising font composite(s); and (3) a comparison of variants in the two fonts to isolate both stylistic and quantitative discriminants. The analysis must proceed in an orderly fashion from gathering to gathering, first focusing upon the capitals, then ligatures, punctuation and lower-case groups. The seriatim order is essential because the divisions of labor in shared books follow no logical pattern; moreover,


123

Page 123
time pressure is a practical factor that invites such haste and must be resisted in font analysis.[38] The assumption that a shared book was printed in halves is valid in The Fawne Q1-2; however, Windet-S and Windet-F alternate in an anomalous pattern within his section (E-I). Likewise, the "two gatherings per printer" assumption is frequently valid in shared books. However, if either assumption led to a hasty survey of Pericles (STC22335, 1609, White and Creede), for example, by jumping from A to C, the presence of Creede-5 in B could be overlooked. Similarly, Fools was divided in an illogical order, with Eld doing AB, GH, K, printer #2 doing CD and then I between Eld's H and K, and printer #3 doing EF. d'Olive is an even more devious case. The shift from Creede-4 on A4v to a Y-font on B1 is obvious and presents a temptation to skip to B4v and C1, where an obvious shift from the Y-font to a mixed S-font occurs; however, yielding to the temptation would occasion the oversight that inner-B is also printed in an S-font (Creede-4 again) and then the B-outer Y-font reappears in G-H3v. Admittedly, this instance of one printer perfecting a sheet begun in another shop is almost unique but emphasizes the fact that books were printed forme by forme. As an illustration of the process, the results of the first and third stages of analysis of Fools are summarized here. While the possibility of distinct fonts in CD, I, and EF is hinted at in the first stage, the evidence that emerges at this stage usually is too general and includes potential discriminants that eventually are discarded as insignificant in the final stage of comparative analysis. In general, it can be assumed that replenishment introduces quantities of wrong-face letters that can be detected in the initial survey, while low-density foul-case letters will usually be overlooked.

First Stage

Capitals. Uniform Y-face capitals appear in Eld's section (AB, GH, K) except for a few foul-case S-face capitals. The presence of both S-face and alternate-face capitals in C-F, I, approximates the pattern of extensive replenishment in Purfoot-Y, and at first glance suggests a single font. The single major discriminant occurs in the oversized (height: 2.9mm, 96mm letter), awkward, right-leaning 'W' in EF, but its cogency is undercut by four appearances (F1:16, F4:12,32, F4v:20) of the normal 'W' found in CD, I, which raises the possibility of two sets of cross-contaminated cases rather than two independent fonts. Similarly, the few alternate-face symmetrical 'M' cut across the possible CD, I, and EF boundaries (C3v:18, E1v:19, E4:21, E4v:14, I2:32). However, two anomalies are limited to CD, I: three appearances of a Guyot 'G' (C2v:14, D2v:21, I3:18) and the 76mm 'Y' in D and I (D2:34, D4v:15,16,17, I1:21, I2v:25, I3v:2), which could be interpreted as transient fouling (see later discussion) of one of two cases. Alternate-face letters 'E K R V Z' are lacking in CD, I, and 'O S V Z' do not appear in EF, but these omissions must be viewed cautiously as supplemental evidence, since non-appearance does not prove non-existence. Both S-face and alternate 'I T' occur in roughly equal proportions throughout. In general, the evidence of capitals clearly indicates


124

Page 124
a distinction between AB, GH, K, and C-F, I, but is ambiguous regarding a further split in the latter section. The fact that two different styles of alternate-face capitals appear in CD, I, and EF cannot be determined from a reproduction.

Punctuation. Fouling with italic query and black letter colon, query and period can be dismissed as insignificant, since these occur throughout the text; similarly, S-face query occurs throughout C-F, I. Eld's section uses the Y-face query consistently. An obvious discriminant occurs in the exclusive appearances of Y-face query in CD, I, and the large Guyot query in EF.

Ligatures. Uniform Y-face ligatures occur in Eld's section. The evidence is ambiguous in C-F, I, since S-face and alternate-face variants occur in the ligatures. The S-face ligatures are obvious, but given the printing quality of Fools, it is improbable that the double variants (excluding S-face) in several ligatures would be detected at this stage, since examination in an original is necessary.

Lower-case. Again, Eld's section uniformly uses Y-face letters fouled by about a dozen S-face letters. The evidence is ambiguous with respect to further subdividing C-F, I, since most obvious variants seem to appear throughout. Except for some S-face g1 (CD, I) and g2 (EF), differentiating other variants requires examination in an original, including the bold variants of 'b d f h', condensed and wider variants of 'k p u v x y', the oversized 't,' the squarish w4 and the bowed outer-stems w3, the Y-face (CD, I) and Guyot (EF) 'g', and S-face k1 (CD, I) and poorly-printing or damaged EF-face k1-k2 (EF). Overall, the lower-case evidence as seen in a reproduction could easily suggest a single font in C-F, I.

Second and Third Stages

The first stage of analysis, in short, clearly indicates a distinct Y-font in Eld's section, but yields ambiguous evidence as to whether one or two fonts appear in C-F, I. The few major differences seen in CD, I, and EF suggest that the second stage of analysis proceed with an awareness of a possible division at the D4v-E1 boundary. The comparison at high magnification of numerous samples of each letter within the sequence of gatherings focuses upon stylistic structure and minute variations in size in order to complete the font composites and isolate variants. In this instance, the analysis of E1 yields new variants not seen in CD, suggesting that a second font composite be devised. The division of labor in Fools, with I separated from CD by EF, automatically requires moving to the third stage of analysis which consists of a comparison of the font in I with the refined evidence accumulated in the two composites. In other cases, the font composite from a section of one book usually is compared with a font in another book. The third stage of analysis of Fools focuses on the lower-case and ligatures. A summary of the results is given in the Appendix.

Overall, the comparative analysis of CD, I, and EF confirmes their separate identity and the assignment of their respective sections to two sharing


125

Page 125
printers. In addition to distinguishing the divisions of labor in Fools, the analytical process produces font composites of CD, I, and EF which are the basis for a search of other books aimed at establishing the identities of the two printers. Once a similar font is located in another book, a font composite is used to determine whether the fonts are identical in the same manner that the CD and EF composites identify the font in I of Fools.[39] In this instance, the font in EF of Whore and The Converted Courtesan (STC6501, 6501a, 1604) exhibits obvious similarities to EF, including the awkward 'W', large Guyot '?' and other discriminants. Analysis of the two fonts shows a precise stylistic correspondence in all particulars as well as some shifts in proportions attributable to time. Some discriminants appear in greater proportions such as the condensed 'fi' and 'ſſ' ligatures in Whore. The only major difference occurs in the appearance of eleven transient foul-case italic 'I' in Whore E3v that were purged before Fools. While the same font is in both books, the unfortunate fact remains at this point (as often happens) that the font appears in shared sections of Whore, Courtesan and Fools by one unidentified printer. Nonetheless, the identification of Fools EF in Whore EF and Eld-Y1 in G-K along with previously identified Simmes-S in AB of Whore establishes the division of labor among the sharing printers and clarifies previous problems about the printing of the two editions, leaving only the identity of the printers unresolved.[40] The search process for unidentified printers is unpredictable, but once suitable candidates are located, the font composite usually easily decides the matter. Whore CD prints in an S-font characterized by low-riding capitals in many sorts and a resident cluster of foul-case italic capitals, traits which identify it as Creede-4 (see below). Similarly, the exclusive discriminants of EF link it to Simon Stafford. A comparison of Fools CD, I, to EF of d'Olive yields positive verification of the identity of this font, but again it belongs to an unidentified printer. The proportions of S-face g1 and Y-face 'g' remain the same, although the S-face k1 seems to decrease slightly; the variants in the ligatures are the same; and anomalous letters reappear such as the few Guyot 'G' and short second-stem 'w', the condensed 'ff' (F1v:34), and the Y-face 'ct' (Fools, D1v:23, D3:11; d'Olive, E1v:5). Identification of a few damaged types completes the process. Since the font composite successfully distinguished Fools CD, I, from EF and identified d'Olive EF, the eventual identification of the printer is a virtual certainty, given a dogged persistence and a good dose of luck. In this instance, White-M's seven years of replenishment and deterioration between LLL and Fools, compounded by the poor printing quality of Fools, buried some critical evidence in the blotted images of the reproductions which were laboriously (6 weeks) examined before consulting originals. The unique 'th' ligature of White-M appeared clearly enough in the reproductions of LLL; the equally unique 'ch' ligature, however, was totally overlooked; neither were hinted at in Fools and d'Olive. Once originals were consulted, the 'th' was easily confirmed, the 'ch' discovered (LLL, A2:10, A2v:38, A4v:3, B1v:24), and then the remnant 'th' recognised in Fools (C2:4 "either", 7 "mirth") and d'Olive (E1:14,15); the few

126

Page 126
'ch' of LLL seem to have been exhausted during the intervening years. Finally, the reproductions were inadequate for confirming that the alternate-face in White-M actually is a Y-face.

VI

Relatively few same-face fonts exhibit the plethora of discriminants seen in mixed fonts such as White-M and Stafford-EF. Generally, font analysis must resort to the kind of tedious search that is necessary for detecting low-frequency foul-case letters. Foul-case pica roman letters constitute a category of discriminants between same-face fonts and are, quite simply, distinguished from the normal letters by the wrong-face and size differentiae discussed earlier. It can be generally assumed that the source of such letters is another pica roman font previously or currently used in the shop; hence, fouling of an S-font by Y-face letters usually implies the possibility of a two-font shop, although this inference must be verified by a survey of books by the printer. However, anomalies occur where the source of a foul-case pica roman letter cannot be explained in this manner; for example, Simmes-S contains a very few Y-face letters although he never used a Y-font. These include Y-face 'k' (Mal Q1, F4v:30; Q2, F1v:33), 'M' (Q2, G3:24; Q3, C1:29, G4:26) and 'P' (Q3, G4:11); further appearances, if any, have been overlooked.[41] Similarly, Felix Kingston's Y-font contains a few Guyot 'G' (The Works STC12312 [1599], C3v:34, C4:1, C5v:11, D2:17 etc.), and White-M uses the unique 'th' and 'ch' ligatures. The value of these letters as discriminants depends upon their rarity (either in a quantitative or stylistic sense) or longevity in the context of other same-face fonts. Windet-S is fouled with the short second-stem 'w' and B-face 'a', both fairly rare variants which also appear in White-M and Danter-M but not as a result of fouling. Both Eld-S and Windet-S show a few wrong-face medium 'z' which do not correspond in either instance to a second font in their shops and do not appear in the other S-fonts discussed here. Letters from the small capitals and shorter roman fonts (68mm, 76mm) constitute another common class of foul-case roman letters. In addition, the damaged 'ſſ' ligature distributed into the 'ſl' ligature sort and foul-case punctuation provide two kinds of valuable discriminants. It seems that compositors were little concerned with stylistic integrity in the latter area, and promiscuously set punctuation from wrong-size roman, italic and black letter sorts when shortages developed in the roman text-font. Hence, quite a few combinations resulted from mixtures of: S- and Y-face roman, black letter, and short and tall italic variants of the query and exclamation point; large Guyot query; S- and Y-face, black letter and italic (two degrees of lean) colon and semicolon; black letter period; and three useful sizes of comma. As a general rule, it seems logical to assume that foul-case italic and black letter punctuation marks are by definition a result of fouling rather than replenishment.

The complementary italic font in a shop was a major contributor of foul-case letters in the strict sense (i.e., an italic letter set in a roman text), but as a practical matter, the concept should be extended to include the use of italic for emphasis in italicised titles and names, speech prefixes and


127

Page 127
stage directions.[42] Both kinds of appearances of italic letters serve the same identification function in font analysis. Generally, foul-case italic letters serve as simple wrong-face discriminants. However, italic typefaces also exhibit some major stylistic variations that can contribute complex discriminants, that is, stylistically variant (Guyot vs Granjon), wrong-face (italic) letters. An analysis of a printer's complementary pica italic font is necessary to identify such complex discriminants for use in a composite. It seems that many printers replenished the upper-case and ligatures of their italic fonts without regard for stylistic integrity and the process produced a rich lore of potential discriminants. These italic stylistic variants are usually easily detected in the upper-case and ligatures for the same reasons noted in regard to pica roman fonts. Simmes's italic font uses capitals in two typefaces and two sizes. For example, his italic 'C' appears both in a normal-height version positioned on the baseline, and a tall version that equals the height of the typeface (i.e., extends from top of ascenders to bottom of descenders), and the italic 'S' and other capitals come in normal and small sizes.[43] Since the italic was used primarily as an emphasis font, it is frequently necessary to survey several books by a printer in order to accumulate a sample of italic letters sufficient for determining whether or not he used a particular stylistic variant. Given a typical dramatic text with the italic font used only for emphasis and speech prefixes, it is unreasonable to reach such a conclusion exclusively from the letters in a single text.[44] For example, between Q2 and Q3 of The Malcontent, Eld purged the italic 'M' with the swash terminating in a curl and replenished the sort with a simple swash 'M', and also purged the italic 'w' with the stem-strokes extending above the x-line and replenished with the variant whose stems begin at the x-line.[45] Unless books before and after Q3 were surveyed, it would seem that Eld's italic appears either in Q1-2 or in Q3, with another printer represented by the variant italic 'M' and 'w'. In some fortunate instances, the extent of the survey required for examining an adequate sample is lessened by the appearance of long italic passages in dedications, prologues and epilogues (see Sejanus, ¶3-A1v:6; All Fools, A3-A3v:14; Romish Spider, A3-A4v), in the peculiar reversal of text and emphasis fonts in Romeo & Juliet Q2 (B4-4v) and Volpone STC14783 [1607], and in rare instances, in an entire text (Vertumnus STC12555 [1607], [2]B-H). As the Eld example implies, a composite of italic letters based upon an extensive survey of a printer's books can be valuable for dating a particular book.

As noted earlier, the distinction between fouling and replenishment is quantitative, with the former implying a second font in a shop. The distinction is also important as regards the function of foul-case evidence in font identification, and beyond that, in the analysis of compositors, method of setting and distribution, standing type and sequence of printing, since the two processes have a different effect upon the manner in which variant letters reappear. A wrong-face population created by replenishment in significant proportions remains stable throughout the lifetime of the font, since it results from a deliberate choice to introduce wrong-face letters, and unless the standards about font integrity were reversed, the letters would not be purged.


128

Page 128
The average ratio of normal and wrong-face letters remains stable across a sequence of gatherings and texts; both variants of a letter will reappear consistently, although the ratios on a given page, pages or gatherings may vary because of the varying demands on a sort by a text; or the ratios may vary in a sequence of texts because of a shift in proportions produced by subsequent replenishment. In general, replenishment does not produce the kind of evidence needed in applications other than font discrimination unless, of course, just one of two cases were replenished with a variant letter. Low-level replenishment represents a special case that will be discussed below. In contrast, the very low density of foul-case letters precludes any consistency in their appearance, although in practice they usually do reappear across a sequence of gatherings or texts. However, their recurrence is both random and unpredictable in the context of predefined units such as the page, gathering, or even two texts. The patterns of recurrence will emerge in the following discussion. In general, a few foul-case letters in a sort will recur with no more regularity than identified types, which often skip gatherings and texts between appearances. But a fundamental difference between foul-case types and identified types must be recognized. The non-specific nature of foul-case evidence is a problem in itself in as much as the very concept of "evidence" is inseparable from the notion of specifically identifiable facts. Ideally, reappearing foul-case letters could be identified and their recurrence thus proven, and this is possible in some instances. For example, Braddock-Y1 is fouled with S-face 'A' (among other letters) that occurs 28 times (an average of three per gathering) in MND Q1. S-face type A1 accounts for six of these appearances (B4v:2, C2:21, D3v:4, F1:22, G4v:26, H1v:13). Three damaged 'ſſ' in the 'ſl' sort recur as well: types ſſ7 (B3:18, C3:11), ſſ8 (D4v:16, F2v:4), and ſſ9 (D4v:6, E1:18, H3:20). Damaged type ſſ2 recurs at least twice in the 'ſl' sort in Short-Y in R3 (L4:4, M3:27). Nonetheless, the use of foul-case evidence is complicated by the fact that most foul-case types are unidentifiable due to the lack of unique damage, and hence must be treated as representatives of a class of letter, that is, as members of a sort (i.e., "three appearances of S-face 'A'" as opposed to "type A1 appears in three locations").

However, the obstacles presented by the nature of foul-case evidence can be circumvented by a method of analysis based upon two concepts that render foul-case types valuable despite their non-specific identities both in font identification and more specific applications related to compositorial and presswork analysis. First, the critical distinction between transient and resident fouling must be recognized in dealing with this kind of evidence. Transient fouling, which endures for a few gatherings but not beyond a text, is of little value in font identification, but can be useful in presswork analysis, since it signifies compositorial behavior in response to conditions such as shortages in a sort. Consequently, short-term transient fouling can provide strong circumstantial evidence of the sequence of printing. The fact that only A-C of MND Q1 are fouled with black letter colons suggests that the book was printed sequentially beginning with A, and not B-H then A, as has been suggested.[46] In contrast, resident fouling endures across several texts (and frequently


129

Page 129
throughout the lifetime of a font) and is of primary importance in font identification and presswork analysis. Due to the random recurrence patterns exhibited by foul-case letters, it is necessary to establish whether fouling is transient or resident before applying foul-case evidence to bibliographical analysis. Otherwise, fundamental errors in interpretation can result. Treating the recurrences of small capitals 'I' and 'A' throughout MND Q1 as repeated responses to newly developed shortages in these sorts (transient fouling) totally distorts the picture of presswork that emerges, if, in fact, these foul-case letters are resident in the font, i.e., already in the respective sort compartments before setting began. Verification of resident fouling requires surveying the font through the target text, and then other approximately contemporary texts. "Contemporary" need not be interpreted too strictly: resident fouling endures in Eld-Y1, Eld-S, Creede-4, Windet-S, Purfoot-Y, Short-Y and other fonts during a significant portion of their lifetimes. The necessity of surveying several books in distinguishing resident from transient fouling is illustrated by the dramatic drop in foul-case 68mm roman 'Y' in Short's section (H-M3v) of R3 Q1 (1597). Eight appearances occur in H, but only one in I-L and two in M, although textual pressure on the 'Y' sort shows no similar decrease, a pattern that suggests transient fouling. However, the 68mm 'Y' fouling is resident in Short-Y at least from 1595-99 (3 Henry VI STC21006 [1595]; i Henry IV Q1 STC22280 [1598]; and The English Secretary STC6404 [1599]).

Second, an understanding of the manner in which fouling occurs is essential to the interpretation of foul-case evidence. The non-identity of recurrent foul-case letters is compensated for by the fact that resident fouling of a font almost inevitably occurs in two or more sorts and thus produces a cluster of fouled sorts that reappears in sequences of gatherings or texts. Treating a font's foul-case letters as a "cluster" is simply an application of the statistical concept of a randomly selected sample population that is to be measured against a predefined distribution of variables in order to determine the probabilities that the sample conforms. In this instance, the variables consist of the cluster of foul-case letters in the populations of the affected sorts as a whole. A comparison of the total populations of affected sorts in two fonts reveals the order of statistical probability that the fonts are the same or not. Since resident fouling usually exhibits a fairly consistent cluster of affected sorts, it is possible to rely upon a quantification consisting of the density of the members of the cluster in a sample population whose limits are set in response to criteria suggested by the text or texts under examination. The quantification need not be overly rigorous when attempting to distinguish two same-face fonts. Each font can be treated as a whole without respect to the distinction between two sets of cases. Hence, a definition of a cluster in units of member density per gathering is entirely adequate; in practice, the cluster compositions alone will frequently distinguish two same-face fonts without reference to member density. Hence, a simple application of this approach sans statistical analysis can show that the cluster remained consistently fouled through the sample texts, thereby providing strong circumstantial evidence of the identity of a font.


130

Page 130

The random recurrence patterns exhibited by low-density fouling are determined by the dynamic relationships among at least four factors: the continuous process of fouling and purging as affected (1) by the ease of purging and (2) by compositorial skills and intention; (3) the variation in the demand upon fouled sorts by sections of a text or by different texts; and (4) the distribution of foul-case letters deeper into sort compartments or into a different case, thereby delaying later reappearances.[47] Although the first two factors seem too obvious to belabor, the standard method of interpreting foul-case evidence in compositorial analysis suggests the need for a closer examination of the interaction of the two as they affect the permanence of fouling in a font. First, the ease with which fouling types could be detected during distribution seems important. The fact that compositors distributed on the basis of a visual recognition of each type seems well established since Charlton Hinman first made the point, and is confirmed by the pervasiveness of one kind of fouling in roman pica fonts. The damaged 'ſſ' ligature which lacked the right shoulder approximated the appearance of the 'ſl' ligature and was commonly distributed into that sort. These damaged ligatures are obvious in proof because the right letter retains the spur which distinguishes it from a straight-edge 'l', but the spur would be virtually invisible during distribution. Due to the damage, it is possible to establish many of these foul-case ligatures as separate types, such as three in Windet-S (Sophonisba, B2:34, D2:1, F4:37) and others as recurrent types, as in Windet-S (Fawne Q1, E1:29, I2v:19), Eld-Y1 (Fools, A4v:14, H4v:2), Braddock-Y1 and Short-Y (noted above). Eld-Y1 presents an interesting instance (unless my eyes deceive me) of visual confusion during distribution: ſſ4 first appears in the 'ſl' sort (Mal Q1, C2:33, E3v:7) and then the 'ſt' sort (Mal Q3, H3:20). On the other hand, foul-case italic capitals, for example, would be obvious on the stone and easily purged, but their permanence varies among fonts. Four appearances of italic 'S' occur in Simmes's section of The Malcontent Q1 (F2v:23, F3: 10,24, H4v:1), but none in the same sections of text in Q2, nor in Simmes's section of Q3 (B-G). Obviously, textual demands on the 'S' sort were exactly the same in Q1, Q2, and F of Q3; hence the non-reappearance of italic 'S' fouling suggests routine purging of the sort at the natural break-point between editions. Similarly, the five variants of the query (S-face, Y-face, italic, black letter, large Guyot), the S-face and italic exclamation point, and the black letter colon and period could probably be recognised during distribution and purged, but on the whole they seem to remain resident once introduced into most fonts. But unless a compositor distributed against proofs, which seems extremely unlikely, the clash presented by oversized capitals such as the S-face 'G S E' in a Y-font (usually quite noticeable in proof) and by less obvious smaller capitals such as Y-face 'G S' in an S-font (fairly difficult to detect in proof) would go unnoticed during distribution and remain resident in a font. Thus the smaller Y-face 'S A T' which recur in Eld-S in sequential gatherings of Sejanus also are found in later texts, even when a prose text, such as The Romish Spider (STC5704, new STC5693.7 [1606]), requires relatively few capitals per gathering.[48] Windet-F 'A D G H M O T' fouling in


131

Page 131
Windet-S is seen in Fawne Q2 and Sophonisba. As noted of Eld-Y1 below, even the larger S-face letters seem to remain in Y-fonts. Small capitals presumably could be purged if a compositor desired, but small 'Y' fouling is resident in Short-Y from 1595-1599; the residence of small capital 'T' in Danter-M is probably attributable to a lack of concern about fouling in this mixed font. Finally, the stylistic clash created by wrong-face lower-case roman letters could hardly be detected without proofs whether or not a compositor was intent upon sorting them out. Thus the Y-face 'g k p' in Sejanus still reside in Eld-S in Spider despite seven(?) intervening texts which provided ample opportunity for purging.[49] Likewise, the Windet-F 'g w' reside in Windet-S at least from 1604-06. The S-face expanded x2 (and x3) seems the only likely lower-case letter that could be distinguished without proofs; nonetheless, it remains in Eld-Y1 after its introduction in Mal Q3.

Not surprisingly, the compositor is the second factor responsible for the difference between resident and transient fouling. Compositors seem to vary considerably in their purging skills and attitudes. Although the intention and skill of a compositor cannot be taken as the sole factor influencing the permanence of fouling in a font, the contrasting treatment of transient vs resident fouling reveals the importance of compositorial intention.[50] That is implicit in the frequent instances where foul-case letters in previously fouled sorts are tolerated while new foul-case letters are purged. It seems that the newly fouled area of text or the kind of newly introduced foul-case letters are fresh in mind and a deliberate decision made to purge. Heavy textual pressure on the query sort in R3 K4 produced fouling with the black letter query, followed by a complete purging. The previous pages of K required six italic, four Y-face and one black letter queries, but in K4 the one italic and four Y-face queries were supplemented with 13 black letter queries. However, the ratio of black letter to Y-face and italic queries remains in LM at the previous levels, suggesting that the K4 foul-case marks were immediately purged, but the resident black letter queries earlier in the gathering left undisturbed. Since this compositor could purge 13 marks from K4, he obviously could recognize and purge the remaining black letter queries if he intended to. Furthermore, the density of black letter query fouling in Short-Y changes radically across three texts, confirming the purging ability of Short's compositors. In 3H6 (1595), black letter query appears in proportions similar to that seen in 1H4 (1598) and English Secretary (1599).[51] The persistence of resident fouling with italic capitals in many fonts suggests a decision to tolerate certain foul-case letters while purging others. Italic 'T' fouling in Short-Y varies in a manner similar to the black letter query, appearing at moderate density in 3H6 (1-5 per 8 pages or gathering) and H of R3. One compositor then fouled his halves of I(16 'T'), K(11) and L(10) extensively with italic 'T'. Both cases were incompletely purged, so that no foul-case 'T' appears in 1H4 until I3v(1), K2v(2) and K3(1), and once in Secretary (1599) (Ll4v:29). Sort-pressure is an unlikely explanation for these appearances.[52] The types were probably buried in the sort compartment when the 'T' was purged as part of the replenishment of the sort after R3 with S-face 'T' in about equal proportions, a factor that


132

Page 132
probably accounts for the much-delayed recurrence.[53] But the italic 'S' in R3 (L1:21, L2:4) does not seem to reappear. In the meantime, other foul-case italic capitals are tolerated through the four texts. Fairly heavy italic 'I' fouling occurs in all four. Other low-density italic capitals 'E F K P' recur sporadically, skipping gatherings and texts in the typical manner. Next, the italic 'B C L S' remain in Purfoot-Y1 after Dutch Courtesan STC17475 and Romish Spider to reappear in Fawne while the low-density 'swash-B F G swash-P V' do not; the transient heavy fouling in DC A-E with 'I' (76) and 'T' (27) was purged after DC. [54] One Windet-S case (Windet-S1, see below) is fouled with 'S' in both Fawne Q2 (H4v:6) and Sophonisba (Bv1:24, C2:29,38, D4:37, E4v:31,33, F1v:6,25), but the italic 'A B I L T' cluster in Windet-S2 does not recur in the latter. The difference probably is attributable to the attitudes of the S1 and S2 compositors. However, the difficulty of purging even obvious foul-case letters despite the compositor's intention is revealed in instances where these letters are eventually purged, but not entirely until two or more distributions. Fouling of MND Q1 with black letter colon begins with one in A2, then B is heavily fouled with 20, most of which were immediately purged, leaving two in C1 to be purged at the next distribution; only the roman and italic colon occur in the remainder of the text. The large number of other foul-case letters in Braddock-Y1 were simultaneously ignored. Likewise, Simmes-S was fouled with italic 'I' in A of Mal Q3 which were eventually purged by the distribution of F (A3-4v: 12, B: 5, C: 5, D: 4, E: 4, F: 3, G: 0). Compositorial intention to purge local large-scale fouling is evident in both instances. The kind of oversight implicit in the sequence of distributions of Simmes-S suggests the probable manner in which low-density obvious foul-case letters became resident in a font. It seems likely that fouling of portions of a text by one of two alternating compositors of different competence would increase the possibility of foul-case italic letters remaining in a font, especially if the second compositor distributed the fouled sections without being alerted to the need for purging them.

The varying dynamics of the continuous process of fouling-purging due to differing textual demands upon vulnerable sorts, ease of purging, and compositorial intention are exemplified in Creede-4, where heavy italic fouling with 'I' and moderate fouling with 'S T' were tolerated as resident while other italic letters were routinely introduced and then incompletely purged. In an early book, the text was fouled extensively with italic 'A swash-A' (26) and 'Guyot A' (1) in Creede's section of Antichrist STC7120 (1603, [2]H-M, see [2]H2v,3,4v,6v,8,8v), and was then purged, since it does not reappear in later gatherings or books. The resident cluster begins to develop in I (two C), then in K with 'G' (23), 'swash-G' (2), 'I' (23), and 'P' (2); the 'I' remains in L (9), 'swash-P' emerges (1) and 'P' (6) increases; only the 'P' reappears in M (10). Compositorial intention explains the localization of fouling, since sort-pressure remains constant ("God", "Priest") throughout. By Honest Whore CD (1604), the cluster is in the font. High-frequency 'I' appears 46 times in C alone, while 'S T' occur at moderate levels in C(9, 5) and D(7, 7). The cluster is completed by low-density italic 'C F G swash-G P W' and Y-face 'M'


133

Page 133
that recur in the typical manner. The 68mm roman 'C' (C4:8[2], D1:5,11) and 'I' (C4:16) do not seem to reappear as part of the cluster, although this may be an oversight. Textual demands in high-frequency words such as "Bishop", "Church", "Doctor", "Epistle" and "God" in Woeful Cry of Rome STC1833 (1605) pulled italic 'B swash-B E' and two Y-face 'D' (B3:34,35) into the text and markedly increased the frequency of appearance of resident low-density cluster members (excepting italic 'W' and Y-face 'M') through additional fouling. An attempt to purge at some time following Woeful Cry is implied by the state of the font in Regiment of the Church STC1827 (1606). Regiment begins with italic 'I' fouling to B4 where the first four italic 'F' ("Father") intrude; only nine 'S' have been set, all roman. Additional fouling that persists throughout Creede's section of the book obtains by D with the full cluster, except that the italic 'swash-B swash-G E' and Y-face 'D' do not appear by this point. It is possible that only one case was completely purged after Woeful Cry and used to set most of Regiment B, but that seems less likely than an incomplete purging of both cases. In any event, both were fouled by the completion of Regiment. Creede's section (A, B(i)) of d'Olive (1606) exhibits the same italic cluster and Y-face 'M'. The large size of the fouling cluster and the progressive fouling in Regiment suggest the sequence in which the two books were printed. d'Olive was not entered in the Stationer's Register and the date of printing has not been determined beyond the imprint date of 1606. Regiment was entered 12 August 1605 but bears an imprint of 1606. The state of Creede-4 with respect to fouling at the beginning of each book seems significant. In Regiment, the text is clean except for 'I' (the first on B1v after five S-face 'I') until the four italic 'F' in B4. Pressure on the sorts in the cluster varies up to and including B4. S-face cluster members appear as follows: B(5), C(9), D(5), G(9), F(10), P(6), S(9), T(22). d'Olive is fouled by the end of the second page (A2v) with italic 'B I P S' before sort-pressure has a chance to develop. This is clear from the order in which the roman and italic capitals were set in their respective sorts. The italic 'I' at A2:6[text] is the first of the sort to be set, the 'S' at A2:8 is the second, the 'P' at A2v:6 is the first, the 'B' at A2v:29 the second, and the 'G' at A4v:29 the second. In short, Creede-4 began d'Olive with the cluster in residence. It seems certain that Regiment was printed first, and quite likely was followed closely by d'Olive. The increased recurrence rate of some members in comparison to Whore suggests that, despite the effort, each subsequent attempt to purge the font left a few more foul-case letters in residence.[55]

The importance of the compositor as a key variable cannot be overestimated, judging from the relation between Windet-S and Windet-F. The alternation of Windet-F with the two cases of Windet-S in Fawne Q2 provides confirmation of proper distribution of substituted roman capitals. One compositor set text from the Windet-F cases and from one of two Windet-S cases (Windet-S2) distinguished by fouling with Windet-F. "Dulcinea" required extensive substitutions of roman 'D' in the Windet-S2 sections of G through E of Q2.[56] For whatever reason, the compositor substituted from both the Windet-S2 and Windet-F cases. Hence, the route for possible mis-distribution


134

Page 134
was: Windet-S2 and Windet-F to speech prefixes; speech prefixes to the Windet-S2 case, into which the roman type on these pages was being distributed. Windet-S2 contained a few resident Windet-F 'D' that reappear in the text at G2(2), G3v(2), F3(1), E2v(1). The Windet-F 'D' substitutions include G2(3), G3v(2), E2v(4). If these had been distributed into the Windet-S2 case, we would expect the rate of reappearance of Windet-F 'D' to increase in the latter gatherings (F-E). Moreover, the Windet-F 'D' fouling appears in Windet-S2 sections of Sophonisba at the same low rate (C2v:19, C4:33, D2:11, G3: 15). Presuming that Sophonisba was printed after Fawne Q2, the substitutions in Fawne Q2 were transient and correctly redistributed.[57] At the same time, the resident cluster of Windet-F capitals in Windet-S2 (A D G H M O T) is tolerated and recurs in Sophonisba at about the same low rate. Given the virtual impossibility of purging fouled lower-case letters, it is not surprising that the Windet-F cluster (a b g k w x y z) remains resident in Windet-S2 in both books. Purging foul-case ligatures is equally difficult, yet this compositor exhibits the ability to sort out the S-face 'ſh' and condensed Windet-F 'ſh' ligatures. Overall, Windet-S2 sections of G are heavily fouled with about 68 Windet-F letters, including two condensed 'ſh' (G3v:36). In resetting G4v from the Windet-S2 case, this compositor substituted ten condensed 'ſh' (G4v: 23-40) and then purged them. Only three appearances occur in later S2 sections of Fawne Q2 (E1:27,33, F3:39), and in heavily fouled S2 sections of Sophonisba (E3:35, E4:1, F2:1).

In general, compositorial intention must be considered a pivotal factor in determining the permanence of fouling in a font. However, the process of purging foul-case letters was subject to a deliberate choice on the part of the compositor, but that decision was not inevitable, since foul-case letters were routinely tolerated and produced resident clusters. Moreover, the effectiveness of the process was limited by the class of letters that were involved. It seems clear that deliberate large-scale fouling of a small section of text created favorable mnemonic conditions for immediate purging if the compositor so desired, although even obvious foul-case letters presented some difficulties in this situation. If a highly skilled compositor was intent upon purging foul-case letters, the kind of letters made a significant difference: purging some classes of letters was well-nigh impossible, regardless of skill or intention. Hence, the relation between the ease of purging, and compositorial skill and intention, has serious implications in font analysis and identification. Fouled lower-case letters can be trusted implicitly as font discriminants, roman capitals slightly less, and italic capitals only if they can be established as resident in the roman font over a sequence of texts.

Given the difficulties of purging even obvious letters, resident foul-case roman types thus tend to reappear fairly consistently over a period of time, either in a given sort or two, or in clusters if several sorts are fouled, a fact which enhances their value as font discriminants as well as evidence of presswork despite the inherent ambiguity of their specific identities. This is also true of wrong-face letters introduced through very low-level replenishment. As will be noted below, replenished wrong-face letters can occur in such low


135

Page 135
proportions that they approximate the recurrence behavior of low-density foul-case letters even though they appear more frequently than the latter. Hence, it is expedient at this point to qualify the concept of "foul-case" to include such replenished roman letters. In general, low-density fouling exhibits a random unpredictability with respect to reappearances within a text or across several texts. Thus foul-case evidence must be approached from the perspective of the statistical principles of randomness and probability rather than in simple numerical terms. The rate of appearance in a sequence of gatherings can be described in terms of the minimum-maximum frequency per unit of text, i.e., 1 appearance per three gatherings, 2-5 appearances per gathering of approximately 300 lines and so on. Since recurrence is random, these can occur anywhere within the unit, i.e., page two of the first gathering, four within the first 20 lines and none after, none until two in lines 298-299, etc. It is surprising to find such theoretical extremes regularly occurring in texts.

This random behavior of low-density fouling can present a problem in font identification if the target font is compared to a similar font in only one or two other books. Members of a cluster, or the cluster itself, can skip gatherings and entire texts and thus create a presumption that a different font is in the other book. The recurrence of the S-face 'g k x' cluster in Eld-Y1 illustrates how the principle of randomness translates into non-reappearance of foul-case letters even though they are probably in the font. S-face 'k' first occurs in Mal Q2 (1 appearance), then in Whore (3), Q3 (1), Fools (4), and Eastward Hoe! Q1 STC4971 (4); S-face 'g' first occurs in Mal Q1 (2), not in Q2, then in the later books in increasing numbers as fouling progresses toward the end of the lifetime of Eld-Y1; S-face expanded 'x' occurs first in Mal Q3 (2), skips Fools, then recurs in Eastward (3). Although the 'k' reappears consistently across the sequence of texts, the sporadic absence of the 'g x' portion of the resident cluster could suggest that the fonts are not the same. The problem is amplified within the gatherings of a single text because non-reappearance seems much more frequent, especially with clusters of 2-3 letters. In Whore, for example, the 'k' skips H-K, the 'g' appears in all four (G-K), and the 'x' does not appear.[58] The 'w' of the Windet-F 'k w' cluster in Windet-S1 likewise skips F and I(standing) in both editions of Fawne.[59] Randomness is especially noticeable in regard to the capitals. Textual demands upon upper-case sorts vary considerably, even in play-texts, a fact which can lead to the non-recurrence of a fouling capital even though it is probably in the font. Eld-Y1 is fouled early on with S-face capitals 'D E G H K O P S' that recur fairly consistently in most texts from 1603 onward except for the 'E K P'. S-face 'K' first appears in Antichrist STC7120 (1603) ([2]B5:5) in a text with low demand upon the sort, is not seen in Mal Q1-2, but then emerges in later texts (Whore, G2:7; skips Mal Q3; Fools, B2:27; Eastward Q1, B1v:28, B2:28, E1v:28, F3v:37, I1v:8). The 'E P' are similar. The 'E' appears in Mal Q1 (B1v:14, standing E1:30), skips Q2, then appears in Whore (I1v:33), Mal Q3 (H1v:16), Fools (G3v:6) and Eastward (B4v:27, C2:24, I1v:13); the 'P' appears in Mal Q1 (B2v:16, standing D2:3), skips Q2, Whore, and Q3, then


136

Page 136
appears in Fools (H2:14) and Eastward (C2v:13, I1v:32). It seems valid to assume that the 'E K P' sorts are continuously fouled from 1603-05 for two reasons, despite the fact that they skip texts. First, their eventual reappearance conforms to the behavior expected in the context of randomness. Second, identified foul-case capitals exhibit similar behavior.[60] The S-face type G1 in Eld-Y1, for example, appears in De Vnione STC13951 (March, 1604) (B1:29), skips several books, appears in Mal Q1 (July, 1604) (E1v:9), skips Siege of Ostend (September) and Whore (November), appears in Remaines STC4521 (November) (a1:7, e2v:4), skips two, appears in Mal Q3 (January, 1605?) (H3: 24), then Fools (?, 1605) (sig. G2) and Eastward Q1 (?, 1605) (C1:30, E1v:25, I2:7). The same rationale applies to low-density foul-case italic capitals in Short-Y. Low-density italic 'P' fouling occurs in R3 (L3:5, L3v:19, M1:1), skips 3H6 and reappears in Secretary (P2, O2:26, O2v:25, Q2:26, R3v:20, V2:3), and swash-italic 'P' in R3 (I1v:2), 1H4 (E3:38) and Secretary (F3:23). The frequent roman 'P' substitutions in speech prefixes in 1H4 had no effect on italic fouling in the sort, since the compositor distributed both 'P' correctly. It may cautiously be inferred from the recurrence of swash-italic 'P' in Secretary that he organized distribution according to textual units on 1H4 E3, where seven italic 'P' occurred in speech prefixes prior to the foul-case 'P' at E3:38, but the two types are not identifiable. 'F' is frequent in 3H6 (B2:25 and eleven others), skips R3 and 3H6, then reappears in Secretary (O4v:12, R1:33, Aa2:30); similarly, the 'K' fouling in 3H6 (B3:8, D6v:16) resurfaces in Secretary (F4:12). 'E' fouling in R3 (I3:36, I3v:4, L1:13) skips 1H4 but reappears in Secretary (Rr3v:21); 'B' appears in 1H4 (K1:25) and Secretary (H4v:4). Identified roman capitals in Short-Y recur in similar fashion. B1 appears four times in 3H6 (A3:24, B6:4, C6:18, D6:19) and 1H4 (B2v:17, C3v: 30, E1v:27, K3:14), once in R3 (H4v:19) but apparently not in Secretary. S-face T1 from the post-R3 replenishment occurs thrice in 1H4 (C1v:9, H4:33, I3v:18) and twice in Secretary (Ee4:9, Ff1:15).

Despite the probability that members of a low-density cluster will fail to reappear consistently as a whole, the odds are just as good that they will. Small capitals and low-density remnants of original capitals in fonts replenished with wrong-face capitals usually reappear, perhaps because the sorts were only partially replenished. The recurrence of normal 'W' and symmetrical 'M' of Fools EF in Whore EF and Courtesan EF illustrates the manner in which randomness influences both the numerical density of recurrences and their location within units of text, a tidy two gatherings in this instance. The normal 'W' skips E of Fools to appear four times in F (F1:16, F4:13,32, F4v:20); of the three in E of Whore, E3:21 and E4v:31 are distributed (E3v:24 left standing); a normal 'W' then is reset in Courtesan at E3:33; the 'W' at F4:24 is left standing. The appearances of the symmetrical 'M' are limited to three in E of Fools (E1v:19, E4:21, E4v:14), but four appearances occur in each gathering in Whore (E1v:18, E3:23,26, E3v:32, F1:20, F1v:12,21, F4:33); two in distributed E(o) do not recur in Courtesan. Even though the 'M' at F1:20 was distributed along with the rest of the page, Courtesan sets another symmetrical 'M' in exactly the same location—what are the odds on this happening?


137

Page 137
However astronomical, another instance occurs in reset B in Fawne Q2, where the Y-face 'R' appears again at B1:32, accounting for the only two appearances of the letter in Q1-2. The few anomalous letters of Fools CD, I, reappear in d'Olive EF with random variation: the Guyot 'G' once per gathering in Fools (C2v:14, D2v:21, I3:18) but only once in d'Olive (F3v:34); the Y-face 'M' twice in Fools (C3v:18, I2:32) but twice in one gathering in d'Olive (E1v:1, E2:14, F3v:17); the 76mm 'Y' four times in Fools (D2:34, I1:21, I2v:25, I3v:2) but once in d'Olive (E3v:31). Finally, the small, low-riding 'W' in R3 (H3:38, H4:28) skips two gatherings before recurring (L1v:32, L2v:5).[61]

In short, the behavior of low-density foul-case letters must be approached in terms of a probable range of recurrence rather than in a simple numerical term, and the even probability of random non-recurrence must be borne in mind. In the instance of Eld-Y1, for example, the cluster members consisting of S-face 'D G H O S', Guyot 'S', and black letter query represent stable components of the resident population of their respective sorts. The principle of randomness indicates that, in theory, each or all may skip gatherings. But across several texts, they produce appearances per gathering consisting of usually 1-2 'D', 1-3 'G', 1-2 'H', 1-3 'O', 2-4 S-face 'S', 2-5 Guyot 'S', and about half of the time 1-3 black letter query. Even though individual letters in the cluster do skip gatherings, they all usually appear in texts of three or more gatherings. The same kind of behavior based on the probable range of recurrence can be noted in low-density clusters in Short-Y, Purfoot-Y, Windet-S1 and -S2, Danter-M, and Braddock-Y1.

The principle that the recurrence of low-density fouling is independent of pressure on a sort is implicit in the preceding examples, particularly in regard to reset portions of Whore and Mal Q1-2 ('S'), where sort-pressure is identical to that exerted in the original setting. However, the minute number of foul-case types noted in these examples obscures the randomness of recurrence in short units of text, such as the page or a gathering or two, that is characteristic of independence from sort-pressure. That randomness is more obvious in the recurrence of foul-case and replenished wrong-face letters that occur in higher proportions than low-density foul-case letters. Again, proportions must be viewed in terms of a range into which numbers are distributed. In general, fouling and replenishment of less than roughly 8-10% exhibits a randomness similar to low-density fouling in an experiential context, that is, the wrong-face letters are missing much of the time and it is necessary to search for them. A transition region occurs roughly between levels of 10-20%. As the proportion of a wrong-face letter moves upward through the region, recurrence exhibits an increasingly direct correlation between sort-pressure and frequency of appearance until the wrong-face letter behaves as a dependent variable. For example, Simmes-S was replenished with a Y-face 'g' and condensed 'p' but in different proportions. The random behavior of moderate fouling and low-level replenishment can be seen in recurrence of the Y-face 'g' across four texts. Hamlet Q1 (1603), Mal Q1-2 H, and Mal Q3 A-E are unrelated texts, while FG of Mal Q1-2-3 are essentially identical (give or take a few lines in Q3). The number of appearances of Y-face 'g' vary


138

Page 138
in the sequence of gatherings, but as a variable that is independent of the total number of 'g' in the texts. The frequencies of the normal 'g' therefore are omitted in the tabulation (see Appendix, Table 1). The number of appearances is random, that is, is as likely to be low (1-2) as high (5-8). Moreover, the appearances in FG exhibit no correlation with respect to the three related and one unrelated texts. Although the 'g' consistently recurs in each sequence of gatherings set from Simmes-S, it frequently skips pages and is buried in disproportionate numbers of normal letters (Hamlet B, 3/131). This aspect of random behavior can be illustrated by a comparison of two fonts that are fouled in the same sort, a factor which levels any skewing that might arise due to the variable frequency of letters in an English text. The low-density S-face erect g2 of Fools EF and the Y-face 'g' of Simmes-S occur as shown in Table 2 (see Appendix), which groups appearances in related (Whore-Courtesan E and Mal Q1-3 FG) and unrelated texts (Fools E, Hamlet E) for each font. The randomness of appearance is obvious when the locations are examined in gatherings with similar totals. The two 'g' of Q1 F and Q3 G appear both early and late, the seven of Q2 G appear in only three pages while the eight of Hamlet E in 6 pages, only one 'g' occurs in page 3 of the ten gatherings, and so on. Likewise, the 'g' in related reset texts vary both in totals and locations. Fools EF differs from Simmes-S in regard to the slightly higher frequency of appearance of the 'g'. If both were S-fonts, this would provide a potential quantitative discriminant between the two. Other members of the low-level clusters in each would probably exhibit similar differences which, when combined with that seen in the 'g', would provide strong evidence that the fonts are not the same. The Simmes-S cluster, for example, includes the replenished 'p' and S-face x1. In contrast to the Y-face 'g', the replenished 'p', which accounts for about 22% of the sort in Simmes-S, exhibits a clear dependence upon sort-pressure across the sequence of gatherings in Hamlet Q1 and Mal Q3. Hence, a positive correlation obtains between the frequencies of appearance of the replenished and the S-face 'p'. However, the frequency of appearance within the pages of a gathering is influenced not only by sort-pressure, but also by additional factors such as the order of distribution and the alternation of the cases from which type is set. Thus the appearances according to pages can vary widely and appear random to a degree, although the relative densities of the two letters are constant for the most part (see Appendix, Table 3). Textual demands differ considerably, with seven gatherings of Q3 requiring more 'p' than eight of Hamlet. Although the sort-pressure swings radically among gatherings, the frequency distributions track rather closely. The slightly higher maximum in Q3 (25 in Q3 B, 21 in Hamlet G) is not significant statistically as an indicator of further replenishment between the two texts. Taken as a whole, the low-level 'g' and moderate-level 'p x' cluster could be expected to discriminate Simmes-S from other S-fonts. For example, The Supplication of Certaine Massepriests Q1 STC14432 (1604) was assigned to the English Secret Press by Woodfield (Surreptitious Printing) and the new STC although Simmes shared in several subsequent editions. The S-font composite of Q1 (A2-G2r) exhibits the same

139

Page 139
variants as in Simmes's sections of later editions, including the 'g p x' cluster. The 'g p' distributions included in the Tables leave little doubt that the font is Simmes-S.

Nonetheless, subsequent fouling or replenishment can shift the proportions of wrong-face letters in a sort and possibly cause confusion in comparing two fonts. For example, a moderate-level replenishment of Simmes-S with S-face 'p' after Hamlet Q1 would have this effect. In such a situation, Knowledge of compositorial fouling-purging habits as observed in other texts can support a hypothesis that progressive fouling is responsible for a shift in the number or the proportions of cluster members, as seen earlier in Creede-4. This kind of shift occurs in the progressive fouling of Windet-F with S-face 'w' between Spectres and Fawne. CD of Spectres exhibit ratios (S-face/normal 'w') of 11/187 and 10/213; the ratios in Fawne Q1 G(o) and reset portions of F (1, 2v-3, 3v-4) are 13/151 and 11/141.[62] The fouling behavior of the Windet compositor(s) in Fawne Q1-2 G, reset portions of Q2, and Sophonisba D, E4, F2, and G2v-3 leaves progressive fouling of Windet-F as the only plausible explanation for the increased level of foul-case 'w'. In another instance, progressive fouling in Eld-Y1 after Mal Q1-2 expanded the resident cluster with S-face 'A C L Y' that recur as usual in later texts. The additional letters first appear in Whore ('C': G4v:23; 'L': I1:38, I3:20; 'Y': I4v:15), then Mal Q3 ('A': H4v:12, I2v:39; 'C': H2:22; 'L': I1:14; 'Y': I2:4), and Eastward ('A': E2:12, F3v:16; 'C': C2v:22, F2v:37; 'L': E1:14, F2v:17). Finally, 'R' first appears in Eastward (A2:27, C2v:27, E1:36, F2v:22) toward the end of Eld-Y1's lifetime. Situations like this definitely require a survey of earlier texts to define the initial resident cluster and its rate of recurrence. Overall, the frequency of appearance of the original cluster is fairly consistent until Whore, when an increase can be noted along with the expansion of the cluster. That the two shifts occur together suggests progressive fouling of one font rather than two separate fonts. Eld's compositors were quite careful about fouling, setting texts in both Eld-S and Eld-Y1, and alternating the two fonts in Eastward Hoe! Q1-3, without producing anything more than low-density fouling in a few sorts, and successfully purging most of the foul-case letters from Eld-S after Sejanus. The shift in Eld-Y1 in Whore probably is attributable to the depleted condition of the font rather than compositorial habit. A different kind of shift occurs when large-scale replenishment modifies cluster composition. The resident cluster in Short-Y includes moderate levels of italic 'T' in 3H6 and R3, but the replenishment with S-face 'T' and purging of italic 'T' has a dramatic impact on the cluster. Similarly, introduction of the 'oo' ligature in Short-Y after 3H6 adds an obvious (although not wrong-face) letter to the font that is useful as a cluster member. Short-Y looks quite different as a result. A similar effect occurs in Simmes-S through normal attrition of wrong-face letters introduced by replenishment. The few Y-face 'M' in the 1604 Simmes-S (noted earlier) are remnants of low-level replenishment (4-7 appearances per gathering) around 1597 (see R3, A-G).

Finally, the relatively large number of recurrent identifiable S-face capitals in the Braddock-Y1 cluster from January, 1598, to October, 1600, reveals


140

Page 140
the effect upon cluster composition and recurrence behavior exerted by the relationship between a periodic decision to purge unwanted letters and the size of a sort-population. It should be noted that compositorial intention to purge letters perceived as not belonging in a font applies equally to foul-case and heavily damaged letters. The residence of the identifiable S-face capitals is attributable to two factors: (1) the deliberately replenished wrong-face capitals were automatically tolerated; and (2), the extent of damage was inconsequential as a motivation to purge such letters routinely within or between texts: the very bold, heavily damaged A2 and gross T1-3 are brutally obvious on a small octavo page with 3-7 capitals at most, yet were tolerated in the cluster for two years.[63] The regularity and frequency of recurrence is striking, given the size of three of the sample texts (prose octavos, 3, 11, and 13 sigs.) with their extremely low demand on capitals and the slightly higher demand of one verse octavo (7 sigs.) as compared to the two mostly-verse play-texts (see Appendix). At some point in 1599, a decision to purge removed some damaged S-face capitals with differing impact upon cluster appearance. The cluster is unaffected by the loss of letters (such as A2, A5, A6) in sorts with high populations or the non-recurrence of A3 between Edward II Q2 (STC17483, 1598; F2:37) and MND (October, 1600; E1:13, F3v:19, G1v:13). However, the eventual purging of T1-3 (probably foul-case 96mm letters) dramatically affects the cluster, since these are the only oversized 'T' in the font. Similarly, the short-term transient 96mm 'C' apparently were purged at the same time with significant impact upon the cluster (A Short Forme STC12312, 1599, nine appearances). An earlier decision (early 1599?) to purge the large italic portion of the cluster had even greater impact upon cluster composition (see Appendix). Despite the effort, italic I1 (and possibly I2) escaped detection or were tolerated (I1: Ed.II, A3:13; C1:18, E2v:3, G1:33; Celestial Elegies STC 21225 [1598], C4v:6, D1v:10; MND, A3:13, E4:10 etc.), as was the italic Q. In the final analysis, a survey of a font across several texts or years is the only way to detect the kinds of shifts in cluster composition illustrated in the foregoing examples.[64]

VII

To summarise, although this kind of evidence requires a tedious searching of sample texts, the section of a font composite which indicates the kind and proportion of replenished abnormal letters and resident foul-case letters and punctuation is a potent tool for distinguishing same-face fonts. Mutually exclusive clusters of these letters usually provide adequate evidence of the identities of such fonts.[65] Fortunately, much of this evidence can be found in high quality reproductions since it consists of obvious classes of letters and marks. However, it must be approached with an awareness of the manner in which the ease of purging and compositorial intention is affected by the class of letter. Otherwise, confusion can result about the distinction between transient and resident fouling. Thus the interpretation of obvious foul-case italic capitals and punctuation, black letter punctuation, and small capitals


141

Page 141
must be preceded by a survey of the font aimed at establishing these as either transient or resident. Lower-case letters can be trusted implicitly as resident in a font. Letters introduced by moderate-to-heavy fouling and replenishment appear with such consistency across a sequence of gatherings that residence is usually not an issue. The interpretation of this evidence in font identification must be qualified by an understanding of the relations among the density of cluster members, sort-pressure and random recurrence behavior. Low-density fouling is accompanied by an even probability of non-recurrence in a sequence of gatherings and across texts. Hence, the non-recurrence of low-density letters must be approached with extreme caution when comparing two fonts, since the non-identity of the fonts cannot be inferred from non-recurrence unless this behavior is consistent across several texts. A difference in the proportions of cluster members or an expanded cluster in one of two fonts may be significant in distinguishing fonts, given the dynamics of the fouling, purging and replenishment processes. A part of the cluster will usually remain constant despite the variations in the other members.[66] Furthermore, differences in proportions provide the basis for inferring states of a font and can provide evidence of the sequence of printing if an adequate sample of cluster members is available in a sequence of books. Finally, clusters are especially valuable in initially determining whether portions of a text were printed from standing type in a sequence of editions even though the cluster letters are not identifiable. The first obvious clue to printing from standing type is that the wrong-face and wrong-size letters are always in exactly the same textual positions in both editions. Reproductions are adequate for this initial insight, but only an examination of originals can provide positive proof of printing from standing type.[67]

As implied in the preceding discussion, the recurrence patterns exhibited by resident clusters suggests an obvious relevance of this kind of evidence to compositorial and presswork analysis, but the subject as a whole is beyond the scope of this paper, needs further investigation, and will be addressed later. However, the implications of recurrent patterns of resident fouling should be briefly noted in regard to the common practice of inferring the method and order of setting, and distribution points, from type shortages and substitutions. The simple fact that resident wrong-face types tend to recur in clusters argues against viewing them as repetitive acts of fouling unless the history of the font shows that these appearances result from transient fouling during the setting of the particular text in question. Given the varying demands of a text upon the variety of sorts in a wrong-face cluster, the odds against recurrent shortages that require the same limited set of substitutions in a sequence of gatherings must be astronomical. For example, this is tantamount to claiming that the text of Mal Q1 required substitutions of the S-face 'G O D' part of the resident cluster first on B3, then again on D1. Unless the distinction between transient and resident fouling and the principle of randomness is borne in mind, the locations of appearances of foul-case letters can be tortured into supporting any plausible explanation of the order of


142

Page 142
setting and distribution. As the random recurrence tables discussed above show, no particular setting order is necessary to distribute foul-case letters into the early or late pages of a gathering, nor is a distribution required to yield a normal letter after the first appearance of a foul-case letter. Random recurrence does the job without help from the compositor.

Appendix I

Y-face 'k' Variants

The various dimensions noted for the 'k' are measured as follows. Width is taken between points dead-center in the particular elements of the letter: (1) link: ascender to arm junction; (2) arm-leg alignment: ascender to arm-serif junction and to the dead-center of the leg at the base-serif junction. Length: (1) arm: point at the arm-link junction to the junction at the bottom of the arm-serif; (2) leg: right apex of the enclosed arm-leg angle to the center of the base-serif on the right edge of the leg, adjusted for damage to the base-serif. Height: top edge of the ascender top-serif to the bottom edge of the base-serif, measured on new undamaged letters (otherwise adjusted for damage to the serifs). Arm-angle: measured in degrees above horizontal as established by a perpendicular through the ascender.

The normal 'k' sort includes several variants differentiated by subtle structural differences that cannot be attributed to the usual distorting factors. The reason for the plethora of 'k' variants is that the Y-face is doubtlessly French in origin and consequently lacked the 'k', hence the need for 'k' punches to complete sets of matrices probably supplied from France. The puzzling fact is that the Y-face 'k' is not even a close stylistic cousin of the more awkward, broader, high-angle arm (72-84 deg.) 'k' variants cut by the Dutch and German punchcutters. Paradoxically, it has more in common with italic 'k'. The major stylistic trait of "elegant" k1, for example, is the staggered vertical alignment of arm-leg, where the leg extends beyond the arm (0.95mm vs 0.85mm), a standard italic 'k' trait that is complemented by the high-angle arm (48 degrees) topped by a right offset, wide (0.65mm) crescent serif that reaches upward like an outstretched hand. The right half of the leg-serif usually points downward like the extended foot of a ballerina. Like a Granjon italic 'k', k1 is a flowing lively letter, totally unlike the ponderous static Dutch and German 'k'.

The remaining 'k' variants illustrate the difficulty of cutting such a complex letter and maintaining balanced proportions among the various elements. The arm-leg of "bent-arm" variant k2 align vertically (0.9mm), and the serif usually extends on both sides of the leg. The structure of the high-angle (48-52 deg.) arm is evident only at high magnification: it bends to form perpendicular junctions with both the short link (0.45mm) and the centered bullet serif which frequently inks underside only and prints as a crescent. New samples of "defective serif" k3 show a full-printing, centered bullet arm-serif and full leg-serif; a shallow strike is probably responsible for the defective arm-serif which usually prints only the rear underside, if at all, as if


143

Page 143
the arm stroke jerked upward to the left. The high angle arm (48 deg.) and leg align vertically (0.85-0.9mm); an inking peculiarity gives the leg a "knee" or apparent bend, although the bend is real in many old types; the leg serif is usually missing to the right.

Several low-angle (42 degree arm) variants are common. "Oblique-junction" k4 is unique in that the arm-leg link (0.5mm) forms an oblique rather the usual perpendicular junction with the ascender while the link-leg junction is nearly perpendicular; the low-angle arm is topped by a large (0.75mm) level slab serif offset right, frequently bent up on the right to approximate a thin crescent, or worn down in old letters to a much shorter length. The arm-leg align vertically (0.9mm); the right portion of the leg-serif is frequently missing. The "long-leg" k5 variant is formed with a high ascender junction (0.95-1.05mm) and a short link (0.4mm) in order to accommodate the long leg (0.95-1.0mm vs a normal 0.85-0.9mm). The low-angle arm is shorter than usual (0.5mm vs 0.65mm) and topped by a medium crescent. The ascender base-serif is quite thick, making measurements of the junction height difficult. Later White-M (Fools, CD, I) has a 'k' (k5b) with a similar appearance due to the high link junction (0.95mm) and short arm (0.55mm) but the link is longer (0.55mm) and the leg shorter (0.9mm) than in k5 (Fools, C1v:11 "winke", C3:25 "booke", 36 "Marke", D1:31 "kind", D2:12 "make", I1v:18 "drinke"). In "long-arm" variant k6, the quite long low angle arm (0.75-0.85mm vs 0.55-0.65mm) extends slightly beyond the leg (0.95mm vs 0.85-0.9mm), despite a short link (0.4mm). New letters exhibit a wide (0.7mm), level slab serif which, ironically, usually is worn down and prints as a dot. The letter's appearance is dominated by the arm which hangs in space somewhat like a fishing pole. "Defective-serif" k3 often approximate the appearance of k6 because of printing characteristics, but measurements easily settle the issue if no k6 is nearby for visual comparison (R3, I4:34, M2v:17; Fools, E1:9,33; Essays STC18041, Yy1v:5 "knots").

Samples of Y-face 'k' Variants

Eld-Y1. New K1: Essays STC18041, Yy1v:27 "talke", 37 "bookes" "speaking"; old k1: Fools, B2v:24,25, G2v:38, H4:16; k2: B2v:24,25, H4:27; k3: Essays, Yy1v:5 "knots", Fools, B3v:20; k4: Essays, Yy1:2 "shrinketh", Fools, B3v:12, B4v:19, H3:22; k5: B4:19, G3:7, H2:33; k6: not seen.

Short-Y. New k1: Sinner's Conversion STC22702 [1594], A3v:20, A4v:13 "forsake", B3v:5; old k1: R3, I2:5 & 8 "clocke", 10 & 11 "strike", 12 "Jacke"; k2: not seen; k3: I3v:4-8 etc; k4: not seen; k5 I1v:4 "kill"; k6: not seen.

Braddock-Y1. Old k1: MND, B2v:27, B3:5, C2v:25 "snake", C3:9 "backe", D2:21; k2: D1v:25 "speake"; k3: D1v:28; k4: C2v:22, D3:19; k5: B3:30, C3:33, C4:17; k6: not seen.

Braddock-Y2. New k1: Poetaster, B2:3 "workes", B3v:4 "broken", D4v: 7,8; k2: Antichrist, [1]B1v:20, C1v:34; k3: B2v:11,19; k4: Poetaster, B1v:4 "trickt", B4v:30 "shrunke", Antichrist, [1]B3v:26 "speake", C1:9, C1v:6; k5: Poetaster, B3v:11; k6: Antichrist, [1]C1v:28, F3:20, T2v:20.


144

Page 144

White-M. Old k1: Fools, D4:24, I1v:3, I2:24; k2: C1:27, D4:11, LLL, A3: 23, C3:24; k3: A3:34, B2v:24, C2v:13,14, F1v:21; k4: Fools, I2v:14; k5: C3:5, D4:17; k6: not seen.

Read-S. Old k1: Revels, B3v:12, D4v:3, I3v:11,20, K2v:3; k2: I3v:25 "like-wise"; k3: seen, not noted; k4: I3v:20; k6: D3:8,11.

Stafford-EF. Old k1: Fools, E4:20, F1:22, F3:20,25; k2: E1:11 "Booke", E3:6, E4:29; k3: E1:17, 33, F1:6; k4: E2v:3, 22 "sake", F1v:3, F3:26; k5: E2v:4 "like", E3:27, E3v:9[text]; new k6: Essays STC5775, O1:11 (0.9mm arm); k6: F1:15, F3:29 (0.8mm arm).

Samples of Y-face 'a' Variants

Eld-Y1. a1: Essays, Yy1v:17 "and", 33 "what"; a2: Yy1v:12 "a Dogge"; a3: Yy1v:13 "ayre", 17 "speake"; a4: Yy4v:17 "what", 33 "strange", 35 "make", Fools, H2v:18 "and".

Short-Y. a1: R3 H1v:33 "face" "scars"; a2: I1v:12 "and", 13 "saie".

Braddock-Y1. a1: MND, B3v:21 "of a", 29 "call"; a2: B1v:10 "and", 24 "again"; a3: B1v:32 "al", 34 "what"; a4: B1v:8 "painted", C3:11 "a".

Braddock-Y2. a1: Poetaster, B4v:6 "speakest", 7 "all"; a2: B2:11 "and"; a3: B3v:9 "faces"; a4: B2v:23 "Alas".

White-M. a1: LLL, A2v:14 "was", 15 "make"; a2: A3:4 "all", B4v:20 "fayre"; a4: A3v:37 "have".

Read-S. a1: Revels, D4v:13 "after", I3v:12 "saide"; a2: K3:13 "and"; a3: E1:13 "pleasures", K2v:10 "and".

Appendix II

Heights of Fell Pica Capitals and Ligatures

Measured at 30X from points 0.025mm inside the top and bottom edges of a type, in millimeters. Width of the ligatures is taken between dead-centers of the two vertical elements and given in brackets, i.e., "[1.15]", or "[1, 1]" for three vertical elements.

  • 2.4: X
  • 2.45 I
  • 2.5: A E F H K L M N V W Z
  • 2.55: D P R Y T; ct [1.75]
  • 2.625: O
  • 2.65 C G; ſh [1, 1], ſt [1], ſſi [1.05, 1.1]
  • 2.7: Q S; ſl [1.15], fl [1.1], ff1 [1, 0.95], ffl [0.95, 1.0]
  • 2.75: f1 [1.15]
  • 2.8: ſi [1.1]
  • 2.85: ff [1.1]
  • 2.9 ſſ [1.15]
The 'm' is the standard reference for the body-width of a typeface; the relation of the width of the 'm', 'n' and some lower-case letters: 1.0: 'u'; 1.05: 'n', left half of 'm'; 1.1: 'h', right half of 'm'; 1.2: 'd p'; 1.3: 'b q'; 2.15: 'm'.


145

Page 145

Appendix III

Comparative Analysis of Fools CD, I, and EF

In general, same-face fonts frequently require an extensive analysis of every variant sort. The relative lack of variants and the stylistic uniformity of Y-fonts forces the comparative analysis of two Y-fonts to focus upon subtle differentiae such as the proportions of new/old letters in various sorts, and anomalous and foul-case letters. The opposite is true of S-fonts. Letters from most of the 54 variant punches and 16 miscastings (excluding misjustified low-rider capitals: b2, h3, n3,4 t2, D2, E3, F2, fl2, ff1b, ff2b, ſh2,3, ſi2, ſſ2, ſt2) frequently appear together in normal S-fonts, making necessary an analysis of all composite groups in order eventually to isolate several discriminants that appear in only one of the two texts being compared. The S-face components of White-M and Stafford-EF are easier to distinguish than normal S-fonts, since the EF-face lacks several S-face letters (see list below).

In general, the lengths of the texts affect the certainty of the results. In shared sections of 4 or more quarto gatherings, the inference that a variant is absent from one font is fairly reliable. However, in short sections of one or two gatherings, this inference must be considered tentative, especially in low-frequency sorts (see later discussion of random recurrence patterns), until verified in other texts (White-M, Witts New Dyall STC22426 [1604], d'Olive EF; Stafford-EF, Vertues Due STC20169 [1603], Genealogy STC12872 [1604], Articles of Agreement STC18455 [1607]). For example, the EF-face 'z' (width: 1.25mm) appears a few times in Whore (E2:32, F4v:10) and Gen (p. 9:27) along with Y-face 'z' (width: 1.45mm) (Whore, E2v:22, E3:6,22,23; Gen, p. 14:35) and S-face z1. Sort-pressure is quite high in Whore (compositorial preference for "cozen", "coz" forms), but very low in Fools (EF total: 6), a possible reason for the exclusive appearance of z1 in Fools (E1:33, E1v:2, F2v:3). The EF-face and Y-face 'z' could nonetheless still reside in the cases. In instances such as the exclusive moderate-level g1 in CD, I, (about 17/gathering) and low-density g2 in EF (about 6/gathering), a consistent frequency distribution vs sort-pressure in two gatherings is a fairly reliable indication of the strong probability that a variant is absent (i.e., S-g1 vs Y-g, 17/147, 13/142, no S-g2). The low-density of ligature samples in short texts such as Fools EF requires the examination of every ligature (excepting the exclusive 'ſh' variants) to isolate and classify the S-face ligatures, but even then, the sample is minimal so that the results must be considered tentative until confirmed in other texts. Furthermore, the overall condition of letters in a sort can cause difficulty. Moderate-level (CD, I, 15/gathering) S-face k1,3 can be confused with unclear samples of replenished EF-face k1,2 (EF-k1: C4v:11, D1:19 "spark"; E2:17, E2v:35, E3:2, E4v:21; oblique junction EF-k2: D1:17 "like"); only one anomalous sample (F4:28 "like") of S-k1 in EF emerged after examining all such 'k'. Similarly, an EF-b1 (F2v:30 "tables") with a heavily-inked bowl frequently looks like S-b1 (a single anomalous S-b1 occurs at F2: 29). Finally, White-M's consistent low-riding capitals (A B C F G H K L M N P R S V Y) differentiate it from Stafford-EF's correctly justified capitals (except


146

Page 146
for some 'A B R'); similarly, Stafford-EF lacks the low-riding misjustified ligatures of White-M.

While searching through a text for clean samples of letters in a given typeface, it is wise to maintain a separate composite for sketching and noting locations of alternate-face letters for reference at a later stage of analysis. The fact that both White-M and Stafford-EF are mixed fonts emerges in early stages of analysing the S-face components; each yields alternate-face lower-case letters and ligatures in the stylistically recognisable sorts. The additional "gross features" discriminants provided by the different alternate-faces (Y-face and EF-face subset) is a mixed blessing since several Y-face and EF-face letters are unresolvable in reproductions. The Y-face and EF-face variants in 'b d', for example, require precise measurements in originals to detect differences in structure and size; the EF-face b1, d1,2, and 'h p t' are easily confused with their S-face counterparts. The ovals group illustrates the problem of "too many" potential variants in mixed fonts. The wide (1.2mm) EF-p2 and the expanded S-p1 (width: 1.3mm) are difficult to distinguish, even when juxtaposed (see single sample of S-p1, F1v:26 "ap /p/ oynt"). Similarly, the Y-p in EF (F4v:19) differs from the EF-p2 (E1:2 "su /p/ pose", E4v:33) chiefly in descender-length (1.15mm vs 1.25mm). The EF-d1,2 and S-d1,2 appear together in approximately equal proportions in Stafford-EF. The two d1 exhibit similar dimensions (width: 1.15mm, junction separation: 1.15mm, height: 3.0mm) but differ in junction angles and formation, top-serif, bowl-stroke, and height of the ascender above the top junction (EF-d1: E1:30 "doubt", E4v:30 "doubt", E1:5 "doe"; S-d1: E1:13 "poynted"). The two d2 differ in bowl-width (1.225mm vs 1.2mm), junction angles and junction separation (1.3mm vs 1.15mm), and EF-d2 is a thicker letter; nonetheless, the two are indistinguishable to the naked eye when heavily-inked (EF-d2: E1:31 "day", E3:29 "diuorced", E4v:30 "day"; S-d2: E2v:5 "dycing"). Similarly, a wrong-face 'd' in CD, I (C1:18 "related", C2:26 "dayes", D3v:18 "In /d/ eede", D4v:14 "darkest") is easily confused with Y-face 'd' (C1:29 "feared", C1v:12 "good"), differing in bowl width (1.25mm vs 1.2mm) and junction separation (1.45mm vs 1.35mm).

In addition to the S-face 'b' variants in CD, I, and EF, an anomalous 'b' occurs in each font. The shorter (2.85-2.9mm), narrower (width: 1.15mm) variant of CD, I (C1:25, C1v:22 "but", D2:12) with wide junction separation (1.45mm) can be confused with lightly-inked samples of the anomalous variant in EF which has the same junction separation, is slightly taller (2.9-3.0mm) and wider (1.175mm) (E1:28, E2:15 "business", E4:11,30), and both appear to be EF-b1, since all three exhibit a slight but noticeable right slant (about 3 deg.). Expanded EF-b2 (width: 1.3mm, junctions: 1.45mm) appears less frequently in CD, I (C2:3,19 "bound") than in EF (E1v:10, E3v:19) and can be confused with Y-face 'b' (C1:36 "breath", C1v:6 "be"[2nd], 17), although the top-junction is perpendicular rather than at an oblique angle. Unstressed EF-y (apex: 35 deg., E1:11,13,14) with usually non-printing serifs looks like Y-y2 (apex: 38 deg.) with worn serifs. The low-angle top-stroke of


147

Page 147
EF-a (E1:3) creates a large counter like that of S-a, but the shoulder is a tight curve and it is a thicker letter, and is usually easily distinguished from the Y-face 'a' variants in CD, I; the mixture of Y-face 'k' variants differs in the two fonts (see lists below). The cross of the EF-e actually does slant downward. Another possible source of confusion in a superficial analysis of the lower-case is the fact that both fonts contain several B-face letters which are obvious ('a', C1:26 "and"; 'd', C2:26 "dayes"; 'e', E1:3 "raigne", I3:2 "mee case"; 'm', E4:1 "shame"; 'c n' seen, not noted) as well as a massive 'o' (C3:5 "lose", F1:30 "our", F4:26 "you").

Both CD, I, and EF use a combination of S-face and alternate-face ligatures, limited to the Y-face in CD, I, with low-density remnants of Y-face 'ff' (one at C2v:30) and 'ct' (two at D1v:24, D3:11), and the unique 'oo' (C3:34) and 'th' (C2:4,7). EF uses Y-face 'ſl' (width: 1.15mm, height: 2.75mm, F3v:16, F4:28, [Eld-Y1, G1:33]), miscast ſſi2 with the short 'i' and leaning middle letter ('ſ' and 'i' spacing, x-line: 1.0mm, baseline: 1.05mm, E3v:18, [Eld-Y1, G3:8; ſſi1 with correctly aligned letters [spacing, 1.05mm] and normal 'i', G3:1]). The few Y-fi1 (width: 1.1mm, height: 2.9mm, E1:16) are taller than the EF 'f1' but usually look the same (width: 1.05mm, height: 2.75mm, E1:20, E2:4, E3:27, E4:22, F1:11, F3v:3). The 'ſi' sort includes two EF-face variants (same height: 2.7mm) that can be confused with the S-face 'ſi". EF-ſi1 is vertically stressed with the shoulder terminating in a larger pear and letters spaced 1.1mm (F4v:1) while ſi2 is unstressed and narrower (width: 1.0.-1.05mm, E4:7, 9, F1:28, F2:2,3); the three are difficult to sort out in the worn condition of Stafford-EF. The condensed 'ſſ' (width 0.9-0.95mm) is distinctive (E2:8,15, E1v:32, F1v:21,23,24), yet measurements are frequently required to distinguish it from the Y-face ligature. The EF-ſh ligature exhibits a high junction (1.5mm) and slanted ascender (spacing, x-line: [1.2mm, 1.05mm], baseline: [1.15mm, 1.15mm]) and is easily recognised. One anomalous 'fl' appears at E1:6 (width: 1.15mm, height: 2.85mm).

Appendix IV

List of Exclusive S-face Discriminants in Fools CD, I, and EF

(CD, I, samples of each sort given first, followed by EF samples, separated by ";". The absence of a variant is indicated by [-]. [+] indicates too common to list.)

  • b1: C3v:2, C4v:19; [one, F2:29 "brest"].
  • b2: C1:2, C4v:19 "blowne be"; [-]. b3: C3:25 "booke"; [-].
  • g1: C1:18; [-]. g2: [-]; E2v:9, E4:9, E4v:10,18.
  • h1: C1:12 "how", 18 "he"; E2v:3 "Whic /h", E4:22 "fathers".
  • h2: C1:4, I3v:25 "this"; E2v:14 "heauen", F1v:16, "With".
  • h3: C1v:9 "him", I2v:11 "have"; [-].
  • p1: C1:7; [one, F1v:26 "ap /p/ oynt"]. p2: [-]; E1:2 "su /p/ pose".
  • t1: [+]; [-]. t2: [+]; [-]. t3: C1:9, C3:3, D4v:6, I2:16,17; [-].
  • u1: C1:6 "you" [2], new C1:16 "perceiue"; E1:5 "should" "jealousy".
  • u2: C1:6 "preserue", C2v:3 "much" [2nd]; E2v:16 "euery", bold 17 "musicke".

  • 148

    Page 148
  • u3: [-]; [-]. u4: C1:2 "doubt", 12 "Nature"; [-].
  • w1: C1:15 "Newes, 26 "would"; E1:9. w2:C1:22 "how"; [F4v:31?].
  • w3: C1:17, 22 "t'was", C1v:11; E1:4,6. w4: [-]; E4:19, F1:7,9.
  • z1: [-]; E1:32, E1v:2, F2v:3. z2: D2v:7,8; [-].
  • D1: [-]; E1v:23, E2:32. D2: C4v:14; [-].
  • S1: [-]; F2v:33. S2: I1v:7; [-].
  • ff1: C4v:25; [-]. ff1b: C1:33; [-].
  • ff2: I1:33, I3:7; F1:20, F3:22, F4:33. ff2b: C4v:12, D2v:6, D3:12; [-].
  • fl1: I2v:7, I3:21; F2:21. fl2: C3v:5, D2:25; E1:34, F3v:11. fl3: I3:26; [-].
  • ſh1: C1v:17, C3:25; [-]. ſh2: C1v:20; [-]. ſh3: C2v:12, C3:18, D3:28; [-].
  • ſl1: D2:9; F1:11. ſl2: D4:17, I1v:8; [-].

Appendix V

Table 1. Frequency Distribution of Simmes-S Y-face 'g' by Gathering

           
Hamlet  
Mal Q1 
Mal Q2 
Mal Q3 
Sup Q1  10 

Table 2. Frequency Distribution of 'g' in Stafford-EF and Simmes-S by Page

                                   
1v  2v  3v  4v  Total 
AF-E  12 
HW-E 
CC-E  0[*]   0[*]   2[*]   2[*]  
Ham-E 
Q1-F 
Q2-F 
Q3-F 
Q1-G 
Q2-G 
Q3-G 
Sup-A 
Sup-B 
Sup-C 
Sup-D  10 
Sup-E 
Sup-F 
Sup-G 


149

Page 149

Table 3. Frequency Distribution of Simmes-S 'p' by Gathering

                   
Total 
Hamlet Q1 
Y-face  15  19  20  17  21  14  10  124 
S-face  82  92  58  56  101  84  71  54  598 
Mal Q3 
Y-face  15  25  22  17  23  11  20  133 
S-Face  44  99  95  85  105  72  106  606 
Sup Q1 
Y-face  25  24  33  26  33  34  13  188 
S-face  105  85  137  151  129  134  53  794 

Table 4. Distribution of Identified Wrong-face Types in Braddock-Y1

             
A2  T1  T2  T3  A3  A4  A7  R1  S1  A1  I
21225 
17483 
12322 
12718 
12312 
22302  2+ 

A summary of the distribution of identified or identifiable types vs the total number of appearances of S-face 'A' in three Braddock-Y1 texts is worth noting. Of 20 total appearances in 17483: A2: 5, A5: 2, A6: 2, identifiable nonrecurrent: 3, undamaged: 8. Of 17 total appearances in 12718: A2: 3, A4: 1, A7: 4, identifiable non-recurrent: 3, undamaged: 6. Of 28 total appearances in MND: A1: 6, A3: 3, A4: 3, A7: 1, identifiable non-recurrent: 3, undamaged: 11.

Table 5. Distribution of Foul-Case Italic Types in Braddock-Y1

         
B   C   G   H   I   L   Q   S   T  
21225  13  11 
12322  27  14  13  17 
17483  10  21  10  24 
12718 
Texts referenced by STC number in the above Tables are:
  • 21225, Celestial Elegies (entered 22 January, 1598), prose 8to, text: A3-D1r.
  • 12322, Two Treatises (imprint 1598), prose 8to, B1-O7r.
  • 12312, A Short Forme (imprint 1599), prose 4to, Dd1-Nn2v.
  • 17438, Edward II Q2 (imprint 1598), 4to play-text, A2-K1v.
  • 12718, Virgidemarium: the Last Three Bookes (imprint 1599), verse 8to, B1-H5v.
  • 22302, MND (entered 8 October, 1600), 4to play-text, A2-H4v.


150

Page 150

Notes


163

Page 163
 
[*]

The research for this paper was in part supported by travel grants from the General Research Fund, Regents of South Dakota, and a generous grant from the National Endowment for the Humanities, Division of College Teachers, for which I also am indebted to Fredson Bowers, Peter Blayney and Donald H. Reiman for their supporting recommendations. This paper would not have been possible without this support. I also thank Fredson Bowers for suggesting the approach of the paper and encouraging and guiding its progress, and Peter Blayney for commenting and sharing his vast knowledge of the subject.

[1]

"Edward Allde as a Typical Trade Printer", The Library, 4th ser., 10 (1929), 121-162. A. E. M. Kirkwood, "Richard Field, Printer, 1589-1624", The Library, 4th ser., 12 (1931), 1-39. Harry R. Hoppe, "John Wolfe, Printer and Publisher, 1579-1601", The Library, 4th ser., 14 (1933), 241-288. These studies concentrate on printer's devices almost entirely with only passing reference to type fonts (see note 5 below). Edwin Eliott Willoughby, A Printer of Shakespeare [William Jaggard] (London, 1934). W. Craig Ferguson, Valentine Simmes, Printer to Shakespeare et al. (University Press of Virginia, 1968) was the first to describe and use a printer's type fonts in identifying his work in sections of shared books. Dennis Woodfield, Surreptitious Printing in England, 1550-1640 (Bibliographical Society of America, 1973). C. L. Oastler included quite unsatisfactory specimens of John Day's type fonts in John Day, the Elizabethan Printer (Oxford Bibliographical Society, 1975) with identifications of punchcutters where possible, pp. 40-43. Akihiro Yamada, Thomas Creede, Printer to Shakespeare and his Contemporaries (Shinsu University, 1981) catalogued appearances of ornamental stock. Clifford Chalmers Huffman, Elizabethan Impressions: John Wolfe and his Press (AMS Press, 1988). See also Peter H. M. Blayney, note 10 below.

[2]

"The Prevalence of Shared Printing in the Early Seventeenth Century", Papers of the Bibliographical Society of America, 67 (1973), 437-442.

[3]

This limitation does not necessarily apply to longer books, especially prose texts with chapter and section divisions which invite the insertion of ornaments and initials. Several of the quarto and folio prose texts referenced later begin sections with ornamental stock as a matter of course.

[4]

This is not a new idea. There are about a half-dozen extant early records of legal proceedings involving the identification of the printers of pirated or seditious books by their type fonts. For some examples, see Talbot Baines Reed, A History of the Old English Letter Founders (London, 1887), pp. 123-133.

[5]

McKerrow overlooked the potential of font analysis and defined what seems to have become a prevalent attitude: "Of Allde's types [type fonts] there is little to say. . . . So far at least I have observed, there is nothing in the types used to distinguish Allde's work from that of a dozen other printers" (p. 147) (Allde actually used four easily distinguished pica roman fonts between 1590-1605). Some years later, Fredson Bowers noted the potential value of the description of fonts in a book, Principles of Bibliographical Description (Princeton University Press, 1949), p. 127, and later commented (p. 305): "The font of the type used in the text, whether roman, italic, or gothic, is given; when possible, this type should be identified by reference to books on printing types, but for books of this period precise identification is usually difficult." See my comment on this problem, note 36 below. One purpose of this paper is to lay a foundation for future work leading to the "precise identification" of fonts by reference to Elizabethan books and printers' fonts.

[6]

This comment applies to every study of a book printed in two or more fonts that I have seen. The consequent analyses of compositors and presswork inevitably are incorrect with respect to the assignment of compositors, running titles, skeletons, number of presses, sequence of setting and printing and other aspects of printing. For example, W. W. Greg distinguished three sections in The Honest Whore on the basis of type fonts, but missed the EF and G-K division, correctly noting it in the second edition by reference to running titles, "'The Honest Whore' or 'The Converted Courtesan'", The Library, 4th ser., 15 (1935), 54-60, repeated by Fredson Bowers (see note 40 below); and G. Blakemore Evans, "Textual Introduction", All Fools, ed. G. Blakemore Evans, in The Plays of George Chapman: The Comedies, gen. ed. Allan Holiday (University of Chicago Press, 1970), 227-232. Evans attributed three compositors to Eld although two belong to the sharing printers. Even though W. W. Greg had specifically noted that a different font appears in one section, Evans begged off: "I cannot see any significant difference, but Greg's eye was sharp and trained and I hesitate to disagree with his years of experience on such a matter" (p. 229). Likewise Akihiro Yamada ignored the issue and attributed four compositors to Eld in "A Bibliographical Study of George Chapman's All Fools (1605)", The Shakespeare Society of Japan, 3 (1964), 73-99. Both concluded that All Fools was printed on one press,


151

Page 151
although they disagreed as to whether casting-off of copy was required. See also: G. K. Hunter, "The Text", The Revels Plays: 'The Malcontent', ed. G. K. Hunter (London: Methuen, 1975), xxxi-xl, and comment below, note 44; Akihiro Yamada, "Q1-3 of The Malcontent, 1604, and the Compositors", Poetry and Drama in the English Renaissance—in Honor of Professor Jiro Ozu (Tokyo: Kinokuniya:shoten, 1980), 107-132, "Press Variants in John Marston's The Malcontent", Studies in the Humanities, 17 (Shinsu University, 1983), 93-113, "Simmes's Compositor A in The Malcontent, Q1-2", Studies in the Humanities, 14 (1980), 121-125, and comment below, note 64; David A. Blostein, "The Text", The Revels Plays: 'Parasitaster' or 'The Fawne', ed. David A. Blostein (1978), 45-52; C. G. Petter, "Note on the Text", The New Mermaids: 'Eastward Hoe!', ed. C. G. Petter (1973), xxxix-xlvi; Allan Holaday, "Textual Introduction", Monsieur d'Olive, in The Plays of George Chapman: The Comedies, 397-402; Akihrio Yamada, "Bibliographical Studies of George Chapman's Monsieur d'Olive (1606), Printed by Thomas Creede", Studies in English Literature (The English Literary Society of Japan: Shinsu University, 1963); Robert F. Walsh, "Textual Introduction", May Day, in The Plays of George Chapman: The Comedies, 311-314. See my discussion of several of the above in "Reproductions of Early Dramatic Texts as a Source of Bibliographical Evidence", TEXT, 4 (1988), 237-268.

[7]

Two forms of measurement indicate the size of an early typeface: (1) the "20-line" measurement in millimeters is the distance from a point in one line to the corresponding point in the twenty-first line of type above or below set without leading (a thin strip of lead placed between lines in later centuries), the line from which the measurement starts being included in the count; (2) the "bare" measurement in millimeters is the distance from the top of an ascender to the bottom of a descender in a single line (i.e., top of the 'h' to the bottom of the 'y' in "thy") multiplied by 20. The complementary measurements are necessitated by the fact a given typeface could be cast on taller bodies than that for which it was designed.

[8]

The technical terminology for the various elements of letter-forms has never been standardized to any degree of unanimity. Philip Gaskell, "A Nomenclature for the Letter-forms of Roman Type", The Library, 5th ser., 29 (1974), pp. 42-51, is the most comprehensive attempt with a glossary of terms and a complete roman alphabet with the term for each element of each letter. Gaskell differs in some respects from the nomenclature traditionally used in historical typography. The physical space occupied by a line of type is defined by four imaginary lines: (1) letters sit on the "baseline"; (2) the "x-line" (Gaskell: "mean line") defines the shoulder height ("x-height") of lower-case letters (m, n, b etc.); (3) the "h-line" ("ascender line") defines the height reached by ascender above the x-line; (4) the "descender line" defines the lowest level reached by letters which protrude below the baseline (g p q y). In regard to other terms, I prefer the traditional "pear" to his "bulbous terminal". "Cross-stroke" (or "cross") seems descriptive of the horizontal strokes in 'A E F T'; the "arm" and "leg" (or "tail") of the 'k' seems preferable to an "upper diagonal" which is topped by an "upper diagonal serif" and a "lower diagonal" terminated by a "lower diagonal serif"; the lower part of the 'g' has always been a "loop", and calling it a "tail" seems inappropriate because it does not look like the "tail" of a 'y' or 'R' or 'Q'; it seems inconsistent to gloss "ascender" as "the extended stem of b, d, f, h, k, l, tall-s" and then use "stem" exclusively to refer to all vertical elements of both short and tall letters, especially when the latter are traditionally "ascenders".

[9]

A font composite is compiled by sketching the stylistic features seen in a font's several groups of letters, punctuation and fouled sorts. The value of the font composite arises from the fact that it defines a font in terms of sets of specific distinguishing features. The diagonal letters form one set, the ligatures another and so on. When comparing two fonts, the sets provide an efficient method for quantitatively estimating the chances that the two are the same. See my discussion in "Reproductions as a Source", pp. 251-255.

[10]

Reference will be made in the following discussion to typeface specimens of the period. The major sources of reproductions of extant founders' specimens are found in a few ground-breaking facsimiles collections specifically intended as a basis of historical study of early types and punchcutters. (1) J. Dreyfus & others eds., Type Specimen Facsimiles, Vol. I, Nos. 1-15 (London, 1963). The anonymous Folger c.1565 specimen sheet attributed


152

Page 152
to Francois Guyot is reproduced as no. 1; the individual specimens of Guyot typefaces are also reproduced in their appropriate section of ref. 4 below. (2) Stanley Morison, John Fell, the University Press and the 'Fell' Types (Oxford, 1967). (3) Hendrik D. L. Vervliet & Harry Carter eds., Type Specimen Facsimiles, Vol. II, Nos. 16-18, "Reproductions of Christopher Plantin's 'Index Sive Characterum 1567' & 'Folio Specimen of c.1585' together with the 'Le Be-Moretus Specimen c.1599'" (University of Toronto Press, 1972). The three extant specimen collections indicated in the title are numbers 16, 17 and 18 in the two volume sequence; each of the three are independently numbered and paginated, a possible source of confusion in references. For clarity's sake, I refer to a specimen by title and specimen number within the section, i.e., "'Le Be' 12" signifies specimen number 12 in the "Le Be" section (no. 18), or by title and page number within the section, i.e., "Le Be" note 13, p. 14. (4) Hendrik D. L. Vervliet, Sixteenth-Century Printing Types of the Low Countries (Menno Hertzberger & Co., Amsterdam, 1968). Specimens are both labeled according to group (i.e., "R17" signifies "roman typeface specimen number 17") and by figure number ("Fig. 186") in this sequentially paginated volume. (5) Joseph Moxon, Mechanick Exercises on the Whole Art of Printing, Herbert Davis and Harry Carter eds. (Oxford, 1962). (6) Fournier on Typefounding. The Text of 'The Manuel de Typographigue', Harry Carter tr. & ed. (The Fleuron Books, London, 1930), esp. Ch. XI "Striking Matrices", pp. 81-88, and Ch. XII "Justifying Matrices", pp. 89-98. The most comprehensive application of historical knowledge about the type production process to a printer's pica roman font is found throughout Peter W. M. Blayney's The Texts of 'King Lear' and Their Origins Vol. I (Cambridge University Press, 1982). The history of typography found in the previous references is complemented in Harry Carter, "The Types of Christopher Plantin", The Library, 5th ser., 11 (1956), 170-179, which reproduces the Varia II folio specimen of the Plantin-Moretus Archives. Also valuable is A. F. Johnson, "The Supply of Types in the Sixteenth Century", The Library, 4th ser., 22 (1944), 47-65; and Paul Beaujon, "The 'Garamond' Types: Sixteenth & Seventeenth Century Sources Considered", The Fleuron: A Journal of Typography, 5 (London, 1926), 131-179, reprinted in Fleuron Anthology, Francis Maynell & Herbert Simon eds. (University of Toronto Press, 1979), 181-214. A frequently referenced paper by Frank Isaac, "Elizabethan Roman and Italic Types", The Library, 4th ser., 14 (1933-34), 85-100 and 212-228, is misleading in its grouping and discussion of pica roman fonts (see further comment, note 36 below). Some sources of potential confusion are worth nothing. The frequently referenced "Specimen Charactervm seu Typorum Probatissimorum" (1592) is variously designated by combinations of "Conrad Berner", "Engelnoff", "Frankfurt", and "1592". Similarly, typeface body-sizes are referenced in the English ("pica"), French ("cicero") and Dutch ("mediaan") vernacular terms that appear in extant specimens. Nomenclature cross-reference tables and measurements are found in The 'Fell' Types (p. 260), Phillip Gaskell, A New Introduction to Bibliography (Oxford, 1972) (p. 15), and Low Countries (p. 16).

[11]

The height of a typeface, as measured from the top of the ascenders to the bottom of the descenders, defined the shortest body that could carry the face. However, a given typeface could be cast on a taller body; the Claude Garamond "cicero" (i.e., pica) punches in the Plantin collection, for example, produced fonts with heights ranging from 80-84mm (see Type Spec. Facs. Vol. II, "Le Be" note 13, p. 14). Similarly, new punches were cut with shortened ascenders and descenders and the typeface cast on a smaller body than originally intended (see Type Spec. Facs. Vol. II, "Folio c.1585" specimens 38 and 45). In one extraordinary instance, a typeface with a bare measurement of 80mm was cast on a body with a 20-line measurement of 67mm without modification of ascenders and descenders (see Type Spec. Facs. Vol. II, "Folio c.1585", specimen 57, and note 57, p. 10). Vervliet does not explain how the type body was modified to make possible the setting of the specimen and other texts. The body of every sort had to have been kerned to allow for the overlapping of ascender and descender lines.

[12]

In the remaining discussion, I will label specific typefaces, variants and subsets by the printer who used them in combination with an initial that was chosen according to an imperfectly defined convention. The "-Y" was selected arbitrarily for Eld's Y-font as first studied in The Malcontent, then the "-Y" was used to designate the same typeface in other


153

Page 153
printers' work. The recognition of the second of Eld's typefaces in Sejanus suggested using the first letter of this title for the label in "Eld-S", and so on for other printers using the S-face; similarly, Windet-F was first noted in The Fawne. Lacking the identity of the printer, the short-title of the book and the sheets in which the typeface occurred seemed the logical approach, hence Fools-CDI or Fools-EF. Mixed fonts present a complex problem in that they are combinations of two or more typefaces, and ideally could be described by the printer's name and the label for the faces, as in "White-S+Y", thereby indicating the two faces that were mixed. However, the compromise "-M" for "Mixed" is most expedient until the typefaces in these mixed fonts are isolated and labeled. I rely upon W. Craig Ferguson's labeling of Creede's fonts, although consistency would call for Creede-1S, Creede-2M, Creede-3M, Creede-4S etc. This cumbersome method is necessitated by the fact that we lack seminal specimens of many typefaces that are attributable to specific punchcutters; it was suggested by Paul Beaujon: "by fixing our attention upon the identity of the type [typeface] rather than assigning anything to a definite punch-cutter, we may hope in time to establish a scientific method of approach based on actual evidence rather than rash speculation or (worse) that learned credulity that makes a printed reference a reliable authority. Most of the types [typefaces] to be mentioned cannot be assigned with confidence to any engraver; their labels, for our purposes, may be the titles of the earliest books in which their appearance has been noted," "The 'Garamond' Types", p. 146. I must note that an essential element of the bibliographical description of a font is overlooked in my labeling, namely, its physical size, because of the slant of this paper. Vervliet's discussion of his naming convention should be followed in bibliographical descriptions of the type in a book (see Low Countries, pp. 14-19). I should comment also on my line-numbering system on pages that include titles and stage directions. The line-count begins at the first stage direction, speech prefix or line of text while titling material is ignored; blank lines setting off stage directions etc. are included in the count, although this sometimes presents a problem since the blanks frequently are not equivalent to the height of a line. I use the simple expedient of inscribing a line-scale on self-adhesive labels on one edge of a six-inch ruler and numbering every fifth line.

[13]

See reproduction, Texts of 'King Lear' (p. 455), and Blayney's attribution (p. 500) of the font to Francois Guyot by reference to the anonymous c.1565 Folger specimen (in Low Countries, R27, Fig. 203; Type Spec. Facs. Vol. I, No. 1) and the specimen of Henry Bynneman's Guyot font in Isaac, Fig. 6.

[14]

Generally, high quality reproductions are useful for discriminating gross features such as the height difference of S-face vs Y-face capitals and stylistic differences (i.e., diagonal group) between two typefaces, provided the images of the letters are distinct and clear. This preliminary stage of analysis is frequently adequate for concluding that two fonts are not the same. However, precise measurements at high magnification such as are referenced in the following discussion should not be attempted in reproductions. The dimensions cited in the present study were taken in originals at The Huntington Library as follows: width is measured from points dead-center in the letter elements as established by the faint trace of the elements in the ink-blot; height is measured at points 0.025mm from the edges of elements. Angular measurements are taken with a specially designed and produced micro-protractor. Consult a graphics design/production shop with a PMT machine for reducing the camera-ready protractor to a radius of about 0.25-inch on celluloid or other suitable material.

[15]

The specimens in this monumental work are an extremely valuable resource for bibliographers desiring to become familiar with actual type appearance at high magnification (20X-30X). The edges of the letters are quite distinct; the only drawback is that modern printing technology has eliminated most of the distorting effects produced by early methods of hand-inking with the balls: originals have to be examined to appreciate the impact of that variable. Such crisp inking is rarely encountered in early books, a fact which renders them quite a bit more interesting in typographical study. Since frequent reference will be made to variations in size as a means of distinguishing variants, a tabulation of the Fell pica typeface (see Appendix II) will provide a useful overview of the accuracy that can be expected of a master punchcutter like Garamond while filing steel


154

Page 154
punches by hand over a period of one to two years, the time necessary to complete the entire set.

[16]

Bibliographers tend to think of type in terms of the two-dimensional image on the printed page. Several references provide discussions, drawings and photos of the various tools and stages of the type production process which are helpful background for the following discussion. Absolutely outstanding photos of punches and matrices appear in Stanley Morison's The 'Fell' Types, particularly plates 14, 15, 17 and 18, and decorated initials in plate 19, with a detailed glossary of technical terms, pp. 256-259. See also Low Countries for a drawing of the type-casting mold and discussion (pp. 10-11); and Harry Carter, A View of Early Typography (Oxford, 1969), Figs. 2, 3 and 5 for drawing and photographs of a mold, newly cast types, and type-caster at work.

[17]

A perceptive former graduate student, Mary S. Radigan-Hunter, responded to this possibility: "Of course: the punchcutter had a bad day, missed a few strikes and crushed his left thumb, so naturally he wouldn't be able to hold the punch perfectly perpendicular, would he?" Fournier explains: "It sometimes happens that the matrix has been struck more deeply on one side than the other, the punch not having been continuously held quite upright in relation to it. . . . Striking the matrix needs a steady and practised hand; which has given rise to a saying that 'A matrix well-struck is half-justified'." (Manuel de Typographique, p. 90).

[18]

See Fournier's explanation of correcting height-to-paper, Manuel de Typographique, p. 90. The depth of impression obtained during the striking of the matrix was a "judgment call" as Fournier notes that the punch "is beaten in with a hammer as perpendicularly as possible until it has gone in as much as a twelfth of an inch or thereabouts [my emphasis]" (p. 83). See Blayney's explanation of matrix-wobble during casting, Texts, pp. 179-180.

[19]

Morison defines the justification of a matrix: "reducing it by filing to a true rectangular shape such that, when fitted to a mould, it will produce type with a level face and of standard height for a character that is optically satisfactory for uprightness, alignment, and distance from adjoining characters," The 'Fell' Types, p. 257. When the metal of the matrix was displaced during impact by the punch, the shape of the rough matrix blank was distorted by the vectorial dispersion of the impact force. In simple terms, the sides of the matrix bulged around and beneath the impression of the letter. This can be seen in the clear photo of an unjustified matrix in A View of Early Typography, Fig. 6 (between pp. 8-9). Fournier's explanation of the arduous process of striking a matrix reveals the latitude for errors in judgment and provides a possible insight into why certain letters exhibit sharp-edged elements while others seem rounded or blunted: ". . . with a gentle hammer-blow a shallow impression is made [in the blank]. If the letter is observed to slope either to right or to left, the punch is turned round a little and a second impression is made on the first, and after this a third, the punch being carefully got into the right position. So the place for the strike is gradually found. When at last the punch is shown by the preliminary trials to be in its proper position, it is beaten in with a hammer . . . ." (p. 83). Justification involved correcting the kinds of misalignments referred to in my discussion of miscast letters: "The matrix has to be reduced at the sides until it is of the right width . . . in the case of a letter which ought to be upright but leans a little to right or left through having been struck awry, the matrix must be corrected by being filed down on the side away from which the letter leans. In this example . . . where the h leans to the right, the letter must be corrected by having some of the left-hand side of the matrix towards the top filed away" (p. 93). Low-riding letters are corrected: "the distance between the end of the matrix and the head of the strike [i.e., top of the letter] is too great, and some of it [i.e., end of matrix] must be taken off"; high-riding letters are corrected: "the face of the strike is too short. This is cured by a few hammer-blows upon that end of the matrix to lengthen the copper," thereby lowering the letter on the matrix (p. 93).

[20]

For example, the fine S2 that seems to lean right (as if the rectangle enclosing the letter were rotated slightly clockwise) and to sit low on the baseline appears in significant proportions in White-M (LLL, A3:16,21, A3v:2,21,36), and in Creede-3 (R&J, A4v:26, B2v:29, B3v:19, C1:32, C3:13,31) amidst normal erect-appearing letters, most of which exhibit some form of distorting damage. A very low portion of Eld's 'S' sort seems to lean although


155

Page 155
they are erect, and all sit correctly on the baseline. It may be that Creede's and White's are actually miscast, but there is no point in pursuing the issue, since they are noticeable just the same. I should note that the preceding discussion has focused upon S-fonts because, in general, Y-fonts seem free of horizontally misaligned capitals with the exception of the low-riding 'C L' in Braddock-Y2 in Book I of Antichrist STC7120. The absence of variant Y-face capitals indicates a single set of seminal punches, with the possibility of duplicate or replacement punches only in the 'S T'. The fact that the capitals are uniformly justified strongly suggests that, if several founders supplied Y-fonts, they purchased their sets of matrices already justified by the same workman.

[21]

Some comment by Paul Beaujon shocks the industrial-age mind into an awareness of the accomplishment of these craftsmen: "As Garamont himself wrote that he had cut punches and cast type since the end of his earliest childhood, it is possible to credit the unsupported statement of Lottin that he was practising in 1510, though our earliest firsthand news of him comes some thirty years later. It takes fifteen years, say the punch-cutters, to learn to cut a twelve-point g [i.e., large pica]: the smaller sizes are cut by men who have had more practice," "The 'Garamond' Types", pp. 132-133.

[22]

Normal ſh1: Eld-S, Sej, B1v:21; Simmes-S, Mal Q3, B4:13; Creede-3, R&J Q2, A4v:1, 23; White-M, LLL, A4:13, I1v:21. ſh2: Eld-S, Sej, C3:5; Simmes-S, Mal Q3, A4:13; White-M, LLL, A2:10,15; Field-S, Arte, D4:34; Creede-4, d'Olive, A2v:33; Windet-S, Fawne Q1, E2:18. ſh3: Creede-4, d'Olive, A4:30; Windet-S, Fawne Q1-2, I2:5.

[23]

In contrast, the cross of the 'e' in early pre-Garamond romans, which the B-face subset in Danter-M resembles (see below), exhibited a upward slant of as much as 24 degrees or so, obviously an intentional stylistic feature. Similarly, the cross of the Le Be 'e' consistently slants upward, while Henry Denham's (An Apology STC6077 [1566]) slants downward; these variant 'e' definitely derive from two punches, while the S-face 'e' from a single punch.

[24]

Commas are difficult to work with, given their physical shape and size. Another problem is created by medial vs terminal justification which affects the spacing between a comma and the preceeding letter. See Blayney's discussion, Texts, pp. 179-180.

[25]

Generally, "obvious to the naked eye" means that the presence of a possible variant can be sensed; however, some form of magnification is always useful in confirming the variant, such as a 3-inch high-quality magnifying glass. Bear in mind that improper orientation of the glass or the viewing angle distorts the viewed image. The AGFA LUPE 8X viewer, a quite useful magnification device designed for examining slides and contact prints, is pocket-portable, inexpensive and can be purchased at most photo shops. The precise resolution that is required for the kind of measurements indicated by the term "high magnification" simply cannot be achieved with "a powerful magnifying glass and a fine scale [ruler]" as one typographer suggested. In the typical steel or plastic ruler calibrated in millimeters, the tick marks are 0.2mm or wider; what's worse, the scale becomes more inaccurate as the lenght increases. Most simple high magnification devices, such as the Radio Shack or the shirt-pocket varieties, invert the image, decrease light about 75%, and distort the image. I use a Wolfe 8x30 Field Microscope with lens extender, an internal scale graduated in 0.05mm, and negligible-light-loss non-inverting optics. Since it took about two years to locate a source for this instrument, I pass along the information: Catalogue #591305, Carolina Biological Supply Company, Burlington, NC 27215 (about $125 or higher, depending upon the dollar-yen exchange rate).

[26]

A comment by Harry Carter, one of the great scholars of early typography, is worth noting in this context: "Some of the types [typefaces] . . . can be reduced to three or four very diverse elements, more of them than I thought at first. These conjunctions and confusions are material for another, more technical discourse, of interest to people who underrate the difficulty of identifying types [typefaces] by looking at them . . .", "Types of Christopher Plantin", p. 171.

[27]

I have discussed the inherent deficiencies of reproductions for typographical study in "Reproductions as a Source", see esp. pp. 238-239 regarding some of the factors that distort the image of a type; other problems of textual transmission that undermine the


156

Page 156
editorial process are detailed in G. Thomas Tanselle, "Reproductions and Scholarship", Studies in Bibliography, 42 (1989), p. 29. A more specific comment about the deficiencies of the reproduction processes upon which scholars rely may be useful at this point, since they vary greatly in their imaging fidelity. In general, regardless of the kind of process, essential information about a typeface is either obscured or lost in reproductions. This is partly due to the fact that the end-product used by the scholar is of at least "twice removed" quality, that is, an original is usually photographed, and this copy is then duplicated for the printing process. Fidelity decreases with each step of duplication. A xerox is a high-contrast image in which lightly inked areas are either cancelled or filled-in, depending upon the density of inking in the original or the adjustment of the machine. Essential information is inevitably lost. The trade-off between the usefulness of a xerox and the damage to the original that results from the duplicating process suggests the ethical principle that direct xeroxing be avoided without exception. If the point to be made is important enough, it will require the examination of an original for final confirmation in any event. When examined in a microfilm viewer, microfilms are virtually useless for typographical study except for sorting out categorical stylistic differences (i.e., roman vs italic typeface). The usefulness of hardcopy produced either with the photoflow (wet) or xerographic (dry) processes varies with the quality of the microfilm itself, of the copying machine, and the reproduction ratio, which is never "actual size". It is frequently impossible to judge the size of a typeface (i.e., 82mm vs 94mm) unless a ruler scale is reproduced along with the book (technicians are extremely erratic in this essential step), but this is useful only for distinguishing between different categories of typeface sizes such as pica (82mm) vs english (96mm), and certainly not for distinguishing fonts within one size range (i.e., an 80mm vs 84mm pica) or measuring 20-line height. Once it has been determined that a font is of a given size such as pica, the usefulness of the hardcopy depends entirely upon the clarity of the images, but it should never be considered an adequate substitute even for high-quality facsimiles. In regard to these, I am not really technically knowledgeable about the actual processes used, but note the following from my examination of the end products. Early W. W. Greg Shakespeare Quarto Facsimiles such as Pericles (No. 5, 1940), and Hamlet Q1 (No. 7, undated), and Shakespeare's 'Titus Andronicus': The First Quarto (Charles Scribner's Sons, New York, 1936) were reproduced by the photographic unscreened collotype method and are of quite high quality; however, at high magnification, the edges of the type-images are fragmented so that details such as nicks and precise locations of bends are lost. The collotype facsimiles vary in quality. Later Greg quarto facsimiles (such as Romeo & Juliet Q2, No. 6, 1949) were screened collotypes with a dot resolution of 0.18mm. Others used considerably finer screened lithographs with a dot resolution of 0.12mm (Henry V, No. 9, 1957, and LLL, No. 9, 1957). Screened type-images, regardless of dot resolution, do not transmit details of the image between the dots (see discussion and specific examples, "Reproductions as a Source", note 13, p. 258). The Scholars's Facsimiles & Reprints (New York, Gainesville) series varies widely but generally are of poorer quality, as are the hardcopy reproductions from University Microfilms. The Scolar Press Limited (Menston, England) offset reproductions are quite good, as is the Shake-speares Sonnets from the Nottingham Court Press (London, 1984) despite the fact its source is a British Library microfilm. The facsimiles in Shakespeare's Plays in Quarto, ed. Michael J. B. Allen and Kenneth Muir (University of California, 1981) were printed from screened line-negatives (as I understand the note, p. 899) with a dot resolution of 0.18mm; a very interesting editorial decision occurred at the final stage: "To avoid a disjointed appearance on the facsimile page, each Quarto page was overprinted with a faint, uniform, tint block. No loss in image accuracy has resulted [my emphasis], although the book has become easier to use." Numerous sheets such as R3 M1-3v definitely suffered from the tinting. Tanselle comments on the philosophical implications of such editorial modifications of the original in the duplication process ("Reproductions and Scholarship", pp. 30-32).

[28]

A mandatory exercise for the new typographer consists of recognising the effects of the reproduction process on the transmitted image of a letter by comparing the Fell 'a' with the "CICERO DE GARAMOND" screened facsimile (p. 124; also in Type Spec.


157

Page 157
Facs. II, "Le Be" specimen 10). The AGFA LUPE 8X magnifier is marginally adequate for this exercise until a proper instrument is obtained. The features of the 'a' samples are quite uniform in the Fell specimen: the pear-tipped fineline shoulder extends beyond the bowl and rises at an acute oblique angle (28 deg.); a top-side medial bend is formed where the shoulder-stroke reverses direction and broadens into the erect back; the long up-curving base-serif symmetrically balances the stressed top of the back; the bowl's top-stroke at 28 degrees produces a high junction with the back and a crimped small counter between shoulder and bowl; the top-stroke curves down to form the left edge of the bowl. A comparison with the "CICERO DE GARAMOND" facsimile reveals that, although some 'a' samples (line 7 "coagulasti") in the facsimile seem to conform to the Fell 'a' traits, the bowl's top-stroke seems more horizontal and shoulder-pear detail is lost in the others. The direction of ink-flow away from the type face and the displaced paper fibers seen at high magnification in an original usually permit detection of such features, which are irretrievable by any means in a reproduction. Next, a xerox of a xerox of the Fell specimen can be compared to the original to appreciate the "twice removed" quality of reproductions. Note particularly the cancellation and fill-in effects; examination of a xerox of the "GARAMOND" specimen reveals how the screened dot pattern is transformed into a solid image. In general, if variant letters exhibit "major" stylistic differences, an analysis of a moderately-large sample in high-quality reproductions can lead to a tentative conclusion that they derive from different punches. For example, Carter and Vervliet trace most (but not all) of the matrices of the "CICERO DE GARAMOND" specimen ("Le Be" 10) to the set of Garamond punches responsible for three other sets of matrices represented in the "c.1585" specimen 38 and "Le Be" specimens 12,13 ("Le Be", note 13, p. 14). A sufficient number of 'a' occur in the four reproductions to reveal such stylistic differences: the shoulder is formed by a symmetrical, tight curve with no medial bend and terminates short of the bowl; the back leans to the right and is frequently bent in at the bowl so that the letter seems to lean right; the bowl's top-stroke curves closer to the horizontal (junction angle of 15 degrees) and joins the back at a lower point, resulting in a larger counter and a more open letter. Although no single 'a' precisely embodies the combination of traits seen in the group of samples, the consistency of the major stylistic differences suggests a different punch. However, an examination of the Le Be 'a' in originals is necessary to confirm these differences and the existence of a second punch (Jean de Tovrnes [Lyons], De la Vie des XII Cesars 1569 [Hn Shelf No. 404078], A4v; Christopher Plantin [Antwerp], B. Svlpicii Severi . . . quae extant Opera 1574, [Hn Shelf No. 388522]).

[29]

See Gaskell, A New Introduction, p. 38.

[30]

Blayney's statistics and discussion of the approximate size of Okes's pica font is most informative. See Texts, pp. 145-148.

[31]

It is interesting to note the use of Eld-Y1 and Eld-S in the former's final appearance in Eastward Hoe! Q3 F3v. The compositor first drew an 'H' and two 'ſſ' ligatures from Eld-Y1 on F3, then switched to Eld-Y1 on F3v. Eld-Y1 apparently was so depleted that substitutions were required with S-face 'k fi ct ſi ſſ A C E M P S'.

[32]

W. Craig Ferguson notes that Simmes acquired a new font in 1606 which prints all of Eliosto Libidinoso STC13509 except for D3-E4v, the final appearance of the long-lived Simmes-S. He printed only four books in the new font before he "was officially replaced as a printer" in March, 1607. See Valentine Simmes, Printer to Drayton, Shakespeare et al., pp. 9, 40-41.

[33]

See W. Craig Ferguson's discussion, "Thomas Creede's Pica Roman", p. 149.

[34]

Stanley Morison indicates that the set of Fell pica punches, for example, includes later replacement punches in the 'C J N R U W g' sorts: "Their presence in a uniform set of matrices suggests that a strike from Garamond's punches with these additions or replacements was justified in Holland in the seventeenth century" (The 'Fell' Types, p. 133); the english (96mm) roman matrices were supplemented by punches cut at Oxford c.1688 to produce duplicate matrices in 25 sorts (pp. 144-145). See also Type Spec. Facs. II, "Index 1567", note 26, p. 4.

[35]

The letters are reminiscent of pre-Garamond early romans, especially in the bold


158

Page 158
'e' with the up-slanted cross. These letters provide a better size match than normal S-face letters to the sometimes overwhelming S-face ligatures, especially the gross fl2, apparent ſſ2, and ſt2, which match the height of the tallest capital ('G').

[36]

Modern typography benefits from the fact that a Caslon, Garamond or Times Roman letter is easily identified by reference to standard specimens. However, the current state of the history of early typography is a serious obstacle to historical font analysis since only a few early pica roman specimens are known that can be definitely attributed to identifiable punchcutters. The problem of extant specimens in font analysis merits further comment. The most frustrating aspect of working with Elizabethan and Jacobean books is that the published specimens do not match with the typefaces in these books, with few exceptions like Henry Bynneman's Francois Guyot pica font. In general, typographers have long resorted to comparatives such as "very similar to" or "a close copy of" in recognition of the fact that similarities in design, but not precise correspondences, can be detected between specimens and fonts in books. A. F. Johnson deals directly with this practice in "The Supply of Types". I believe that the old, frequently invoked, axiom "the same type can look quite different in the hands of different printers" can be rejected out of hand. The bare measurement of a typeface and individual letter sizes are constant physical facts: precise measurements of letters of a typeface simply do not vary significantly from one book to another or one printer to another. This is obvious from thousands of such measurements: a Y-face 'm' is always narrower than the S-face letter, and always by an amount that hardly varies. Similarly, referring to differences as "a later development in" an identified typeface is legitimate only if it can be shown that the seminal set of punches, or an intermediate set of matrices, is identical except for the differences. Ignoring these basic physical facts leads to nebulous references to typefaces which are a hindrance to font analysis and the bibliographical description of the typefaces in a book. Frank Isaac, for example, notes of the "extremely popular" pica roman: "most of the quarto plays are printed in this size. They were modelled on the 'Cicero de Garamond' of the Berner sheet which is also found in Plantin's Index," "Elizabethan Roman and Italic Types", p. 89. Isaac groups these typefaces according to the styles of 'g' and 'w' and makes a few supplemental distinctions in surveying the varieties of the two letters used by some 18 printers (e.g., "A w with a short second stem . . . was used in this group by Bishop in 84[mm], Bynneman in 86, Day in 82, Orwin in 82, Purfoot in 82 . . .", p. 90). Isaac's listing lacks references to sample books and dates, a real problem. For example, Purfoot indeed used this 'w' at least from 1570 (The Hateful Hypocrisy STC10951) in an F-font that by 1579 (The Forest of Fancie STC4271) had become mixed with an S-font. Determining when Purfoot-Y was introduced needs further investigation, but it does not use the Guyot 'w' and appears in Fawne with Windet-F which does, an occasion for confusion. Beyond that, pica roman fonts simply cannot be grouped according to a few variants. The confusion caused by such a simplistic rationale is apparent when the six pica roman specimens are examined. Henry Bynneman's font (Fig. 6) derives exactly from the Francois Guyot punches that produced the Guyot specimen of the anonymous Folger folio sheet (Type Spec. Facs. I, No. 1; Low Countries, Fig. 203, p. 268), except that Bynneman's font is cast on an 82mm rather than a 78mm body; the bare measurements of the two match (72mm). Richard Field's and Valentine Simmes's S-fonts (Figs. 8, 9) are stylistically distinct in all respects from Garamond's pica romans. Similarly, John Windet's pica (Windet-F) (Fig. 12) represents a different design concept and scale (the x-line is closer to the h-line) and is closer to faces cut by Ameet Tavernier (see Low Countries, Fig. 206, p. 267). Thomas Orwin's is a mixed font (Fig. 10) with assorted S-face letters; John [Henry?] Denham's font (Fig. 11) combines letters from "Le Be" specimens 12,13 with others probably cut in the Netherlands, but this font needs further study. [I append this brief reference to W. Craig Ferguson's Pica Roman Type in Elizabethan England (Scolar Press, 1989), which arrived too close to press-time to permit detailed comment on what seem to be fundamental problems in approach and documentation. Too many conflicts between this paper and Pica Roman Type occur to permit specific discussion; hence a general statement must suffice. Although my identifications of fonts and printers have been


159

Page 159
presented in the context of fonts evidence because of the focus of this paper, a fundamental principle has been tacitly followed: identified recurrent types provide the final, absolute evidence of the identity of a font. Hence, my identifications, based upon such evidence, can be trusted to either confirm or supersede those presented in Pica Roman Type.]

[37]

The combinations of typefaces noted in regard to mixed fonts suggests many hypothetical explanations for the minor variations in same-face fonts. For example, the fonts were purchased at different times from one founder who had replaced certain matrices in the meantime; or from two founders who were supplying type cast in hybrid sets of matrices that differed in but a few sorts; or one of the two printers was forced by circumstances to replenish the ligatures from a founder who supplied only S-face pica type. The Plantin and Le Be inventories of punches and matrices suggest another possibility. Plantin had at least four sets of pica roman matrices from two sets of punches, one by Claude Garamond; the Le Be inventory listed eight sets of pica roman matrices (see Type Spec. Facs. II, "Le Be", notes on specimens 10, 12, 13, 19, p. 14). Secondly, although it illustrates an 18th-century foundry, a woodcut of William Caslon's casting shop shows casting stalls and casters at work along two walls of the room (see fold-out illustration at the front of Talbot Baines Reed, A History of Old English Letter Foundries [London, 1887]). Taken together, these suggest a fundamental economic rationale that could explain the mixing of typefaces either piecemeal or more extensively. A temporary heavy demand for pica roman types could be met by assigning several casters to the task using available sets of pica roman matrices; or, by assigning two casters to high-frequency sorts (a e p etc.) using duplicate matrices not necessarily in the same typeface. I must admit these explanations are purely speculative although attractive.

[38]

The following examples are based upon my own oftentimes regrettable experiences. Trips to rare books collections inevitably place a premium on time that, at least in my case, often invites short-cuts in methodology that cause later problems.

[39]

The search process will be discussed in a later paper. Relevant details about this case are as follows. While previously identifying Eld-Y1 in G-K of Whore, I had noted that EF used the large awkward 'W', the Guyot '?' and several other potential discriminants. The completed Fools EF composite demanded a follow-up analysis of EF of Whore and the second edition, The Converted Courtesan. Some Fools EF discriminants emerged in sections of Limbo-Mastix during a survey of other books by Eld, but it is not the same font. Fools EF discriminants were later encountered in books by Simon Stafford and Edward Allde, leading to the former's identification. I had noted some traits of d'Olive EF while analysing Creede-4 in A, B(i), and these suggested the comparison to Fools CD, I.

[40]

See Fredson Bowers, "The first quarto of 1604 was set up in three sections comprising sheets A-B, C-D, E-K. . . . This final section of the play, E-K, appears to have been machined on two presses, one of which printed sheets E, F, H; and the other, sheets G, I, K" (p. 4) and other discussion of the compositors, running titles and skeletons in the two editions, "Textual Introduction", The Dramatic Works of Thomas Dekker Vol. II (Cambridge, 1955), pp. 3-14. I might note that identifying running titles is a difficult proposition simply because they are set in same-face letters. As I note in "Reproductions as a Source", most of the letters in typical running titles are unidentifiable because of the lack of specific damage (see pp. 242-244); hence, without specific evidence to the contrary, concluding that the same running titles move through a book is based upon an assumption of regularity in compositorial procedures at one press or in one printing shop. Shared printing is inherently irregular in this sense since a minimum of two compositors, two presses, two skeletons and casting-off of copy is required. The confirmation of printing in one shop by a font analysis must, therefore, precede running-title analysis. In the event that two fonts are found together within a single gathering (as in Eastward, Fawne E-I, and d'Olive B), a survey of the printer's work is an additional prerequisite in order to confirm the possibility of simultaneous printing in a two-press shop. (Aside from d'Olive, I am unaware of any instance of two printers' fonts appearing together within a gathering.) Otherwise a dangerous trap ensues when the running titles in sections of a shared book are set in same-face fonts without distinguishing peculiarities such as a distinctive damaged letter


160

Page 160
or stylistically distinct letters (e.g., italic 'swash-curl-M, swash-P' in A-D, Granjon 'M P' in E-I). This is evident in several of the studies cited in note 6, where the same-face running titles of one printer are assigned to another of the printers in a book.

[41]

Locating low-density foul-case letters such as these is especially tedious. I usually do a minimum of three searches of a text (or texts). One cardinal rule is to note the location when the letter is first detected, since a repeated search is no guarantee that it will be found again. It is worth noting again that non-appearance does not mean that the letter is not in the font.

[42]

The complementary black letter font was a major source of foul-case punctuation, but I know of no resident fouling with letters from this typeface except for punctuation marks. However, infrequent instances of textual fouling with the black letter 'ye' occur (Antichrist STC7120 [1603], [2]H3v:32; The Woeful Cry of Rome STC1833, B2:3, B2v:14; Regiment STC1827 [1606], I1:18; W. Craig Ferguson noted 'ye' fouling in Creede-3 also, "Creede's Pica Roman", p. 152). This 'ye' fouling should be viewed as transient fouling resulting from the compositor's justification habit in prose texts, and not as resident in the roman cases. A rather unique kind of textual fouling with a large pre-Garamond roman font occurs in Isle of Gulls (STC6412, 1606), where the left-third or so of F4v:31-39 are set in this oversized Venetian(?) type. Fouling with 'C' occurs elsewhere throughout this section (EF) (F4v:13 etc.). The discovery of this unique font in a signed book would automatically identify the printer, given its rarity in Elizabethan printing.

[43]

The three styles of italic capitals in use during the period provided obvious discriminants. In the most common, stems and ascenders of the capitals are straight and terminate in serifs (as if a roman face were titled to the right); this style appears in the pica italic faces cut by Robert Granjon (see The 'Fell' Types, p. 139) and Ameet Tavernier (Low Countries, Figs. 233-234, p. 305). Granjon also cut a second style of italic capitals characterized by extended "swash" strokes. The left stem of the 'A M' is a long, graceful curve; the bowl-stroke of the 'B D P R' begins with a swash, the 'G' terminates in a descender swash, the tail of the 'K Q R' terminates in a swash below the baseline, the diagonal stroke of the 'N' incorporates both the shoulder and base swash, and the 'E' is like a reversed numeral "3". Both styles are mixed in Eld's, Braddock's, Purfoot's, Short's and other italic fonts. Read's (and Eld's early) font included a variant 'A M' in which the swash terminated in a curl, fairly common variants. Francois Guyot cut a third style of capitals (see Low Countries, Fig. 231, p. 301) in which the bowl-stroke of the 'B D R P' begins almost vertically at the x-line to the left of the ascender, the left top-serif of the 'M' hooks over, and the swash diagonal of the 'N' extends below the baseline. The Guyot style was less common than the Granjon.

[44]

G. K. Hunter noted of Mal Q1: "One of the problems of printing The Malcontent seems to have been the very large number of italic Ms required for the names, and this put the printer to all kinds of shifts. But at the beginning of sheet F, and incessantly thereafter we find a new kind of italic capital M [Guyot], which we must assume to have been in one compositor's box but not in the other's. That this M was in Sims's possession is shown by its use in QC [Q3] of The Malcontent and in that part of The Honest Whore printed by Sims," "Introduction: The Text", The Revels Plays: 'The Malcontent' (1975), p. xxxii. Simmes's section (F-H) of Q1 also uses the Guyot 'P' in large numbers, and both Granjon styles of 'B M P'. Eld's section (B-E) uses the straight-stem Granjon and the curl-swash variants of 'A M', swash 'B D P', and what appears to be a few Ameet Tavernier 'M' (B3v:34, C1:29, C4v:8, D3:32, E2:19, E3:32). The Guyot 'M P', and the curl-swash 'A M' and Tavernier 'M', are exclusive to the two sections of the book (and Simmes's and Eld's sections of Whore as well). Given the number of 'M' and roman substitutions, the situation suggests two separate printers' fonts rather than two cases, and calls for a survey of a few other books by Simmes to determine whether the letters found exclusively in Eld's sections were ever used by Simmes. Unfortunately, Hunter seems not to have been aware of W. Craig Ferguson's Valentine Simmes, Printer to Shakespeare (1968).

[45]

I must note the possibility that the replacement of a large portion of the italic font could be responsible for the shift in these sorts. The italic of Q3 prints much more


161

Page 161
crisply than that of Q1-2, a fact which strongly suggests new type. The difference cannot be noted in reproductions.

[46]

See Robert K. Turner, Jr., "Printing Methods and Textual Problems in A Midsummer Night's Dream Q1", Studies in Bibliography, 15 (1962), 33-55. The major portion of the study aimed at establishing the setting and distribution sequence by correlating 115 reappearances of identified types with evidence from substitutions. Turner's use of a reproduction for the typographical identifications produced an error rate of 44% which, in combination with my identification of 47 additional appearances of damaged types, discounts his conclusions about the setting and distribution of the text and his interpretation of substitutions; see my discussion of Turner's distributions (pp. 240-242) and the Table of Identifications (pp. 266-268), "Reproductions as a Source".

[47]

The distribution of foul-case letters deeper into sort boxes or into a different case, or the setting of a text from alternating cases, one of which is not used after a certain point, can only be surmised from the behavior of foul-case letters which, in any event, is random. Although it can be assumed that types in a sort emerge in a "last distributed, first to be reset" pattern, the pattern is merely probable, since a certain randomness is introduced by the order in which pages of a forme are distributed: the last page to be set is not necessarily distributed first. The genuine exhaustion of a sort prior to substitutions would seem to pull all normal letters from a sort box, but in order to establish a real number, the point in the text at which setting from the sort began would have to be established, and that is possible only through an exhaustive identified-types survey to accurately pinpoint distributions. Hence, this discussion of the behavior of foul-case letters ignores the possible effects on recurrence behavior by setting from two cases and by distributions. Nonetheless, distribution factors seem responsible for recurrence in many instances. For example, the damaged 'ſſ' in the 'ſl' sort appears eight times (A2:9,12[text], A5:11, A6:15, A7v:22, B1:20, B4:11, B4v:11, C1:10) in 3H6 but not later in this book in either the 'ſſ' or 'ſl' sorts; seven times in R3 (H3v:28, I4:35, I4v:29, K2:27, L4:4, M1v:21, M3:27); none in 1H4; then six in Secretary (F3v:20, I1:25, L3:13, T3:36, Ff3v:8, Gg3v:10). The damaged ligatures were purged and newly damaged ones wrongly distributed in the sequence of texts except for the one which reappears (R3, L4:4, M3:27; Sec, F3v:20). It is worth noting that the right shoulder of the foul-case 'ſſ' at Sec I1:25 is not sheared away, but merely bent downward rather severely. This alteration of its appearance was sufficient to occasion incorrect distribution into the 'ſl' sort.

[48]

Recurrences of Y-face capitals across three gatherings in the verse text of Sejanus: 'S': B3:38, B3v(5), C1v:16,24, C2v:24, C3:10, D1:16, D1v:9,11, D3:2, D3v:2,3; 'T': B2:4,16,20, C3v:14, D3v:28, D4:20,21,30; 'A': B4:31, C2v:24, C4v:17,22, D3v:33, D4:18. Despite the extremely low demand on the sorts in the prose text of Spider, it was sufficient to produce the following appearances in the first gathering of Eld's section: 'S': H1:13[text], H1v:4; 'A': H2:30, H4:29; 'T': H4:20. In addition, new fouling occurs with Y-face capitals in 'F' (H4v:13) and 'G' (H3:4); 'C' also appears later (I1:3). Incidentally, the mixture of S- and Y-face 'I' occurs in both texts.

[49]

See Sejanus: 'g': B1:32, B1v:6, B2v:9; 'k': B2v:18; 'p': B3v:37. In Spider, the prose text requires a greater number of lower-case letters, but this does not seem to increase the appearance rate of low-density fouled letters: 'g' appears at H1:6[text], H2:28, 'p' at H2:28, but 'k' is lacking until I1:25, I1v:23.

[50]

I think it fair to observe that consistent performance in the purging of fouled speech prefixes, emphasis names and stage directions (roman capitals with italic lower-case) exhibits no real significance with respect to compositorial skill: the roman capitals preceding 2-10 italic lower-case letters present a minimal challenge, if any. The capitals are almost invariably distributed back into the roman pica cases as numerous compositorial analyses have shown. However, the practice of automatically interpreting roman substitutions (capitals) in speech prefixes irrespective of the italic font's history merits comment. The possibility of fouling the italic cases with roman capitals calls for caution in invariably assuming that substitutions are transient fouling. The S-face 'S' substitutions in Creede's printing of Henry V Q1 (1600) may represent an exception. The ratio of S-face 'S' to italic 'S' is: D:


162

Page 162
4:14, E: 6:17, F: 7:10, G: 0:4. When Creede printed R&J Q2 in the previous year, the italic 'S' sort set 20 speeches prefixes in A and 10 in B; the first apparent substitution with S-face 'S' occurs at C3:2 in a gathering which used only 8 italic 'S'. Then in Pastor Fido STC12415 in 1601, a single S-face 'S' appears among 24 italic 'S' in B, and another at K4:15 amidst a few italic 'S'. I am at a loss to explain how the C3:2 and K4:15 appearances could be viewed as substitutions in view of the quite light sort-pressure. Even so, it is possible that Creede's italic 'S' sort was severely depleted between 1599 and 1600, and then replenished to par by 1601 (or was being used simultaneously in another text). However, resident fouling with S-face 'S' is an equally plausible explanation. A lapse in the compositor's attention cannot be ruled out either, especially in view of the inexplicable shift of the text typeface from roman to italic in most of B4 and parts of B4v in R&J Q2.

[51]

The density is fairly consistent throughout 3H6 which was printed in eights. Dividing into units of eight pages (equivalent to a quarto in fours) gives the following numbers of black letter queries: 0, 11, 9, 14, 5, 8, 6, 9. The numbers in 1H4 are: 0, 4, 8, 14, 16, 1, 9, 7, 8, 6. In Short's section of R3 (H-M3v): 2, 1, 15, 2, 1. The Y-face and italic query numbers are omitted to emphasize the variation that occurs in R3, although the pressure on the '?' sort in the three texts is not significantly different. What remains, I believe, is a reflection of compositorial intention either to purge or to tolerate obvious foul-case letters. Since both cases were used to set R3, Short-Y as a whole was imperfectly purged of the black letter query prior to setting.

[52]

See MacD. P. Jackson, "Two Shakespeare Quartos: Richard III (1597) and i Henry IV (1598)", Studies in Bibliography, 35 (1982), 173-190. Jackson did not consider the possible connection between foul-case evidence and the cases from which each compositor set his respective section of the text.

[53]

Foul-case italic T1 occurs in 3H6 (D5:19, E6v:3?) and R3 (I1:24, L4v:9), but not in the two later texts, suggesting a purging of the text of R3, but not the cases. The 'T' in 3H6 and Secretary do not correspond to any seen in R3 with one probable exception (R3, L4:26, M2:33; Sec, Ll4v:29).

[54]

The former may simply skip Fawne in the manner characteristic of low-density fouling. However, the high density of the 'I T' fouling precludes the possibility that it could skip gatherings or texts. Furthermore, both cases were fouled since the 'I T' appear in both formes and halves of each gathering A-E. Deliberate purging of large-scale fouling is clearly indicated. The 'I' in Fawne Q1 C2v(5) and D2v(5) is probably fresh transient fouling. DC was entered 26 June, 1605, and Spider on 4 February, 1606, about six weeks prior to Fawne. Appearances of Purfoot-Y1 italic cluster members: 'B': DC, C3v:16, D1v:11; Fawne Q1, C4:21. 'C': DC, D3:24, E4:13, E4v:12; Spider, C4:4; Fawne Q1, C1v:13, Q1-2, D2:10, D3:36. 'G': DC, D2v:14, E1:16; Spider, B3:37, B4v:34, C2v:9, E3v:20, F2v:17. 'L': DC, A2v:9, B3:21, 25, C1:29, C2:18,20, C3v:15,17, D1v:38,39, D4:12,34; Spider, G1:5, G2: 31, G4v:22; Fawne Q1-2, D2v:34. 'S': DC, B2:24, B3:3, B4:3, C1:28, C1v:1; Spider, C4v:20, D3v:31; Fawne Q1-2, D2:30.

[55]

Appearances of common low-density cluster members in Whore: 'C': D1:5, D1v:37; 'F': D4v:25; 'G': C1v:5, D1v:31; 'swash-G': C3v:13, C4v:7; 'P': D2:23; 'W': D2:32,33; Y-face 'M': D4v:31. Appearances in d'Olive: 'C': B3v:28,30, B4:2,26; 'F': B1v:30, B2:35; 'G': A4v:29, B4:17; 'swash-G': B2:15,27, B3v:15, B4:5; 'P': A2v:6, B1v:17, B4:21[2]; 'W': A4v:22; Y-face 'M': A3v:22. Incidentally, Windet-F prints Ee1-Ff4r of Regiment STC1827.

[56]

Windet-S2 sections: G1:1-8, G1v, G2, G3:1-28 except most of 17-18, G3v, G4, G4v; F1v, F2, F3; E1, E2v, [E3?], E4v.

[57]

The Fawne was entered 12 March, 1606, five days before Sophonisba, but this does not definitely indicate the order in which the two were printed. In any event, the resident roman fouling is the same in both and was not purged between books.

[58]

S-face k1: Mal Q2, C3v:24; Q3, I1v:14; Whore, G1:19, G2:18, G4:23, Iv4:8; Fools, A4:14[text], B3:29, G3v:37, K1:7; Eastward, B1v:15, B2:28, C2:4, H2v:11, H2v:35. S-face g1: Mal Q1, B2:16, C4v:28; standing type Mal Q1-2, D4v:16, E4v:11; Q3, H2:24, H2v:10, H3:24, H4:26, I1:34, I2:22, I2v:4,8, I3v:27; Whore, G1:8, G2:18, H1:5, H1v:30, H2:33, I2:6,30, I4v:7, K1:32; Fools, H1:5, H2:28; Eastward, D4v:37, E1:14, E2:5, F3:27, H3v:17,29, H4:33. Expanded S-face x2: Mal Q3, H1:8, I3:29; Eastward, E1:4, F3:20, H4:20.

[59]

Errors and omissions in the list of the 'k w' appearances noted in "Reproductions as a Source" (note 39, p. 264) are corrected as follows. Windet-F 'k': Q1, E3:8; type k2: Q1, F4v:30, H3:4, Q2, E3v:14, H3:14; both Q1-2: H1v:17, H3v:27, H4:17, I1v:29, I2:23, I3:6. Windet-F 'w': Q1, E2v:8,36; Q2, E3v:11, E4:15; both Q1-2, H1v:17, H3v:27.

[60]

The recurrence of the S-face A1 in MND Q1 has been noted earlier. The number of appearances (28) of S-face 'A' indicates that replenishment introduced this letter along with the S-face 'H S' (with a few 'Q R V') and 'ſt' to form the S-face cluster in Braddock-Y1. See also later discussion and the Braddock-Y1 Tables in the Appendix.

[61]

This 'W' also appears regularly in 3H6 (B6v:12,32, C3:15, C5:24, D1:30, D2v:26, E2v:12) but does not seem to occur in 1H4 or Secretary.

[62]

Spectres was entered 11 January 1605; Fawne on 12 March 1606.

[63]

The damage is of the kind that a compositor or corrector could not ignore. It is not clear whether Antony Hammond meant to imply a general principle about compositorial behavior in noting that "such types are likely to be noticed by the compositor or proofreader, and discarded rather than distributed," "The White Devil in Nicholas Okes's Shop", Studies in Bibliography, 39 (1986), p. 160. To be on the safe side, it seems wise to take his statement as true of Okes's workmen; it does not apply to every shop.

[64]

Implicit throughout this discussion is the categorical necessity of the survey as a prerequisite to compositorial and presswork analysis. Unless this principle is clearly understood, the potential for error is limited only by the inferential ingenuity of the scholar who misunderstands or ignores font evidence. For example, an extraordinary sequence of inferences in Akihiro Yamada's "Q1-3 of The Malcontent, and the Compositors" leads to the rhetorical question: "But was it in fact practical to use only one copy [printer's copy] when two compositors were working at different shops, as must be the case with Q3 according to the implication [actually an unqualified conclusion] of Ferguson's statement that the type used in H-I of Q3 is one 'which Simmes is not known to have used'. Or should it be believed, in order to settle this question . . . that this particular type was in fact in Simmes's possession?" (note 14, p. 130). The implicit answer is absolutely wrong. The "tail" of compositorial and presswork analysis, however ingenious and cogent, simply cannot "wag the dog" of a survey of nearly all the books (in the original) by a printer. Ferguson's survey indeed proved that Simmes did not use this type. The extreme improbability that he nonetheless possessed it is disproved by my survey of Eld-Y1 which yielded about 450 appearances of 85 identified types from The Malcontent in a dozen other books by Eld. Such simple facts as are yielded by the survey can, in effect, "trash" studies of printing which overlook font evidence. (I was tempted by insatiable curiosity to tabulate the number of pages "trashed" by the font evidence noted herein, but the thought of adding even one more statistic tipped the scale to pity, leading doubtlessly to a much-deserved mutual sigh of relief).

[65]

A few of the previous isolated examples can be combined to illustrate the point. Given the fact that both Purfoot-Y1 and Eld-Y1 are used in two-font play-texts 1605-1606 by John Marston (Fawne, Eastward Hoe! etc.), a comparison of the two same-face fonts would be the logical first step in determining whether either printer appears in both books. A-D of Fawne exhibit high-level replenishment of upper-case sorts with S-face letters, a foul-case cluster of italic 'B C L S', both erect and leaning italic query (D4:5), colon and exclamation point (A4v:17), while the Y1-font sections of Eastward show a low-density (foul-case) cluster of S-face 'D E G H K M O P S T g k p x!?', black letter and leaning italic'?', and damaged 'ſſ' ligature in the 'ſl' sort. Or, Windet-S in E-I of Fawne could be compared to Eld-S, since Windet printed Sophonisba, another Marston book. In this instance, the fouling is in contrasting typefaces: Eld-S is fouled by Y-face 'g k y', while Windet-S is fouled by Windet-F 'a k w x y' and 'ſt' ligature as well as italic capitals. One final qualification is necessary in concluding this discussion of the random behavior of wrong-face letters. The fonts have been treated as a whole while ignoring the probability that two sets of cases fed the texts included in the examples. It seems likely that the use of statistical methods for analysing the frequency distributions by pages will distinguish the cases from which texts were set, given clusters of appropriate dimensions. However, more work is needed in this area.

[66]

Although the group of fonts considered in this discussion represents but a small


164

Page 164
fraction of the pica roman fonts used during the period, I believe that the cluster phenomena noted in these fonts suggests a definite possibility that two same-face fonts could include clusters with some lower-case duplicate members, although I have not encountered such an instance. The commonness of italic 'I S T' fouling and wrong-face replenishment in sorts such as 'g k p y' certainly point in that direction. Replenishment of the upper-case suggests a second hypothetical consideration. The progressive replenishment with wrong-face capitals in an increasing number of sorts, for example, could conceivably transform a font with S-face capitals in just a few sorts (Braddock-Y1, 'A H Q R S V') into one with S-face capitals across the upper-case (Purfoot-Y1), given an adequate time interval. This is implicit in the continuum of proportions of normal and wrong-face capitals seen in mixed fonts.

[67]

See my discussion of the inadequacies of reproductions for typographical study of the kind needed to demonstrate printing from standing type and to identify running titles, "Reproductions as a Source", pp. 242-245.

[*]

standing from Whore