Early modern printers and the standardization of English spelling.
Modern linguists do not explain convincingly how English spelling advanced to its present mainly standardized form. They suggest various influences, principally the writings of sixteenth-century spelling reformers, grammarians, and teachers rather than printing. Study of reprints of a Caxton work shows that modernization occurred before these could have operated. The economy of the printing-house, notably the use of apprentices for distribution, established common spellings, usually shorter than earlier forms, which were widely transmitted as compositors moved to different printing-houses. Modernization was largely effected by 1700.
Although many influences on the development of substantially standardized orthography have been suggested, the mechanism by which modern spelling was achieved by the end of the seventeenth century has not hitherto been identified. (1) To whatever extent the writings of orthographical reformers like Sir Thomas Smith or John Hart, (2) teachers like Richard Mulcaster, (3) and the writers of spelling books and dictionaries may have affected the movement towards a modern standard spelling, their works were disseminated most widely by the press. It is crucial to understand that early modern printers--unlike present-day printers--did not follow the spellings of their copies. Consequently, the printed works of spelling reformers and dictionary-makers used spellings that were not sanctioned by their authors. For instance, Mulcaster's 'The Generall Table', intended as a substantial contribution to orthographical uniformity, demonstrates many instances of unreformed spellings frowned on by Hart and other reformers. There are also other non-modern spellings only partly reformed by Coote in 1596. Of the 136 spellings on the first page of the table (p. 170), 36 do not achieve modernity, with redundant doubled consonants and terminal -ie for y throughout the table, providing, as in the later works to be discussed, persistent classes of older unreformed spellings.
An even more telling example of the influence of printers on the spelling of school books is the earliest substantial guide to English spelling, Edmund Coote's The English School-Master (1596). Alston records a long sequence of editions of his work from the first in 1596 through the 48th in 1696 to the last edition in 1787. (4) Examination of the first 242 words of the spellings (on pp. 74-7 of the first edition) shows first, that these words in the earliest edition contained 93 non-modern graphemes (i.e. elements of spelling); the number of non-modern graphemes declined to 83 in 1614, 65 in 1630, 33 in 1684, and 21 in 1696: patently the spelling became more modern with each successive edition. (5) Within a century (1596-1696), 78 per cent of the spellings of Coote's English School-Master were modern. (6)
Spellings in fact moved towards modernity from the beginning of printing in England. N. F. Blake examined the successive reprintings of Caxton's Reynard the Fox (1481), which Caxton reprinted in 1489, his Alsatian assistant and successor Wynkyn de Worde in 1515, and Thomas Gualtier, a naturalized Frenchman, in 1550. (7) Blake observed that 'One of the most remarkable features [ ... ] is how the grapheme ea appears suddenly and becomes accepted as the standard spelling in some words in such a short time'--words such as teache, head, great, and beast (p. 68)--and noted other forms that had become regularized in the direction of modernity in the 1550 edition. (8) His is the only longitudinal study of early printers' spelling known to me. What is important is not that printed books continued to evince a considerable amount of spelling diversity but that the variations that tended towards regularity also moved towards modernity, i.e. the spellings that were increasingly favoured during the next century and a half. Obviously printers had not yet arrived at a consistent orthography but they had begun to influence the development of spellings towards the modern standard.
My purpose here is not to show exactly when standardization occurred or at what rate or with what kinds of words but, rather, to define the mechanism by which spellings became modernized from the beginning of printing of English books. I have chosen to prepare my demonstration (not proof) from readily available materials and texts of which the bibliography is well established so that, if desired, my limited studies can be easily enlarged. The further progress of printers' spellings towards a modern standard can be illustrated conveniently by successive reprints of two Shakespearian plays, for which the stemma of editions is more straightforward than it was for Blake's study, each successive edition being printed from its immediate predecessor. Small portions of verse chosen to avoid the possible effects of spelling variation for justification were selected for convenience from Q1 of Titus Andronicus in 1594 and the reprints in 1600, 1611, and the four Folio editions in 1623, 1632, 1664, and 1685; and from Q1 of A Midsummer Night's Dream in 1600 and its reprint in 1619 and in the Folio editions, together amounting to 200 verse lines. Although the sample is small, it was thought to be large enough at least to illustrate broad trends in spelling in the seventeenth century, particularly as the similar folio formats of the folios might be expected to inhibit compositors from employing their own variant spellings. (9)
The variation of spellings from edition to edition supports the following observations. (1) Surprisingly perhaps, nearly 22 per cent of the text of the first Quarto editions at the end of the sixteenth century was already in modern spelling; (2) by F4 (1685), 171 of the 200 lines, say 85 per cent, were completely in modern spelling. Therefore, on the basis of this sample it is a simple fact that English spelling was predominantly modern by the end of the seventeenth century, the greatest progress towards standardization in printed works occurring from 1630 onwards. (3) Words that resisted variation in all editions were the eye-rhymes coffe (cough) and loffe (laugh) at the end of a couplet in MND (TLN 425-6)--which suggests that successive compositors were sensitive to the spellings of their copy--and among others, some words which individual compositors would not usually encounter often enough to have a personal spelling of them. (10) (4) Words altered to modern spellings in one reprint were sometimes altered in a subsequent reprint to a non-modern form, e.g. (Tit.): forrest/Forest/Forrest (677); Here/Heere/Here (1241): such variations can only be compositorial. Further, whereas variation of phonemes usually involving vowels inside words are generally useful to linguists as possible indications of shifts in pronunciation, in a different context words that may be subject to contraction in the printing house are more illuminating: the spellings hadde and had, for instance, do not indicate different pronunciations. Accordingly, in the Shakespearian sample (5) there were 219 spellings from which a modern spelling could be formed by contraction. By 1685 only eleven spellings remained to be contracted, most of them the non-modern forms mentioned above and therefore subject to the same kind of modernization in later editions.
The question of why these spellings changed at all when it was to the benefit of the compositors to follow their copy in successive reprints--and why spelling became standardized--lies within the ambit of the derivative sense of economy as 'the thrifty and efficient use of material resources', according to which we can expect an avoidance of unnecessary or unproductive effort (by compositors) with a desire for maximum return from labour and materials (by master printers), and then, more particularly, 'the arrangement or mode of operation of something', here, the printing house. Economy is not a modern notion, nor is the conception that spelling standardization would produce economies a modern one. John Hart maintained in 1569 'how much more easie and readie, it [spelling reform] will be for the writer or printer' (fol. [4.sup.V]) and pointed out that 'we should not neede to vse aboue the two thirdes or three quarters at most, of the letters which we are nowe constreyned to vse, and to saue the one third, or at least the one quarter, of the paper, ynke, and time which we now spend superfluously in writing and printing' (fol. [5.sup.R]). The long-winded title-page of Bullokars Booke at Large (1580) suggests the benefits of using both the old and reformed orthographies, 'to saue expences in Bookes for a time, vntill this amendment grow to a generall vse'. Most human operations tend towards economy, and the regularization of spelling was a human operation, located in the economy of the printing house. To this we must at last turn.
The personnel of an early printing house consisted of a master printer, journeymen printers (compositors or pressmen), and apprentices, as well as (in larger establishments) a warehouse keeper and sometimes boys to aid the pressmen. It is only to the extent that any of these set type that they are relevant here. Apprentices usually started in the printing house around the age of fourteen or fifteen and served for five to seven years. Consequently, there was often a wide spread of age and experience in the early printing house, a fact that is significant in the context of spelling variation.
Compositors brought with them to the printing house their 'traditional spellings' that--to the extent that spelling had retained a phonetic basis--reflected their regional origins. It is helpful, therefore, to consider the role of apprentices in spelling standardization in the printing house. D. F. McKenzie has listed the 'Country Origins of [Stationers' Company] Apprentices Bound, 1562-1640' and his list has been analysed by Paul Morgan. (11) Besides London and Westminster and 'Unknown', 3, 780 apprentices came from 42 counties and Ireland and Scotland, Yorkshire providing the third largest number (213). Clearly there was a considerable amount of native linguistic variation in the London printing houses before 1700, and in the very early period, before the movement towards reform or modernization of spelling had reached its strongest point--if indeed it had any effect at that time--there must have been considerable diversity of compositors' spellings when they did not follow copy. For the years 1562-5, which pre-date Hart, Coote, Mulcaster, and the other spelling reformers, McKenzie records London apprentices who came from 34 counties.
Apprentices (not all of whom were destined to become compositors) were usually bound to a master printer and were not freed from the Stationers' Company before they were twenty-four years old. They could have had about seven years' education in a grammar school, but such an education stressed Latin rather than English. (We should not forget that a major portion of books printed before 1700 were in Latin and Greek.) The wastage was considerable. Cyprian Blagden says that 'it has been given as 50 per cent' (12) and Morgan confirms this figure for the Warwickshire apprentices (p. 6). Unlike journeymen printers, early apprentices received no wages. They lived as part of the master's household: 'printing house' is entirely appropriate in the early period of printing. In return for the apprentices' labour (and a premium), masters were bound only to teach them the mystery and craft of printing. Although the number of apprentices a master might bind was restricted, and an attempt was made in 1587 to prevent printers binding apprentices when journeymen were available to do the work, there was a clear economic advantage for masters to employ as many unpaid apprentices as they could get away with. (13)
Apart from the general provisions of articles of apprenticeship, there is no contemporary information about how sixteenth- and early seventeenth-century apprentice compositors learnt their craft. Moxon (1683-4), whose manual falls towards the end of the period with which we are concerned, barely mentions apprentices at all. (14) But the small size of the early seventeenth-century printing houses--when most master printers were limited to own only two presses, (15) and could have employed very few journeymen--suggests that apprentices learnt composing under the guidance of a senior compositor, the master himself or a journeyman. A demonstration of this can be taken from the intensive bibliographical investigation of the most important printed work of the early seventeenth century, the Shakespeare First Folio of 1623. McKenzie records four workmen who were employed at William Jaggard's press in 1622-3 when the F1 was being printed. (16) Jaggard's son Isaac was freed by patrimony (as the son of a master printer) on 23 June 1613, probably after his father became blind: his name appears in the F1 imprint. Miles Shirborne, the son of a baker in Islington, Middlesex, was bound on 30 September 1611 and freed on 4 January 1619: he might have remained in Jaggard's employ as a journeyman. A Warwickshire man, John Shakespeare, the son of a butcher, was bound on 20 March 1610 and freed on 22 May 1617: Hinman quite reasonably suggests that he was the compositor B with whom the final apprentice to be mentioned shared the composition of many of the tragedies.(17) Finally, John Leason, the son of a Hampshire yeoman, was bound on 4 January 1622--around the time that Hinman estimates work on F1 began. (18) As there is no record in the Stationers' Company that he was freed, it is possible that he absconded after working on the Folio. However, Hinman makes a good case for identifying him as the compositor whom he had previously described as an apprentice compositor. (19)
In Printing and Proof-Reading Hinman shows compositor E (Leason) working closely with the experienced compositor B--with whom, however, he could not keep up: the formes of type compositor E set were 'intercalary' to the established sequence of composition, correction, and press-work, and at the beginning he was unable to set from manuscript copy efficiently. Most of his work in the Folio was set from the less demanding printed quartos that served as copy for many of the tragedies. (20) Hinman concludes that E was a compositor who was
less expert than we can reasonably suppose any full-fledged typesetter of the time would have been, and it is clear that his superiors were perfectly familiar with his limitations and considered it necessary to proof-read most, at least, of what he set. He was not supplied with type-cases and working space of his own, and he is occasionally found doing some of the more routine work of the regular Folio compositors. (pp. 225-26)
In brief, compositor E was an apprentice compositor in the first months of his apprenticehood. The most significant of his attributes for my argument is, first, that he started his compositorial career under the tutelage of a senior compositor with whom he worked very closely, and second, the 'more routine work' that Hinman mentioned included E's distribution into type-cases of type set by other compositors.
Now to make the connection of apprentices (using E conveniently as an example), composition, and distribution with the standardization of spelling. I suggest that standardization of spelling was brought about not primarily by authors' responses to the profusion of written materials that issued from the early press but rather by the mechanical economy of the printing press, centred on composition. (21)
Distribution and composition were reciprocal parts of the typesetting process that compositors had to master. All type selected from type-cases to be printed must eventually be returned to the type-cases, i.e. distributed. The first kind of distribution an apprentice would encounter was restoring pied sorts (individual types) to the correct compartments of the type-cases. Pie--sorts mixed promiscuously by sizes and founts--was the compositor's bane. It could result when a compositor put so much type into a compartment of the case that sorts overflowed into adjacent compartments, or when a compositor dropped his composing-stick, or fumbled so that types fell to the ground. The most serious pie of all occurred when a forme was not locked up securely and a page of type collapsed when the forme was being transferred to the press. Moxon records solaces (fines or penalties) for such offences, but of course the greatest penalty was having laboriously to distribute the pied type back to its correct cases and compartments. (22) Someone had to determine the fount and size of the vagrant sorts and return them to their correct places. For this ungracious and laborious task no one was better suited--through his lack of standing--than the least experienced apprentice. However, beneficially, distributing pie was an excellent introduction to the characteristics of the several founts and to the lay of the case, without secure knowledge of which both composition and distribution would be inefficient. Timperley's Printers' Manual (1838), which was directed at learners, notes that 'Sorting pie is generally the first employment' of the learner. (23) Although his work should not be regarded as authoritative on printing-house procedures two hundred years earlier, it seems sensible that the beginner, knowing nothing about printing, should be employed as Timperley indicates.
The new compositor did not learn of the spellings used in the printing house from distributing pie. However, experience of distribution equipped the inexperienced apprentice to begin composition. After a digression on the lay of the case, Timperley continues: 'Having made himself perfectly acquainted with the upper and lower cases [ ... ], he [the learner] is put to distributing, or conveying the different sorts of letter to their respective apartments' (p. 15).
To learn the lay of the case was not in itself simple, especially as early on the lay was not necessarily uniform from one printing house to the next. (24) Blake illustrates a hypothetical single lay with 250 compartments needed to accommodate all the single-, double-, and triple-letter type-forms (allomorphs) employed in Caxton's printing house in order to emulate handwritten script: not only the ligatures that have survived to this day, ff, fi, fl, ffi, and ffl, but also numerous other combinations of consonants and vowels (sa, sc, se, sh, sl, so, su, and ssu), vowel combinations (ee, ei, en), and so on. (25) More recently, for the twenty-four letters of Caxton's alphabet, Bridget Cusack identified about 270 pieces of type, all needing to be accommodated in a single type-case. (26) In distributing sorts the novice compositor would learn what accented and special characters and ligatures such as ffl existed, and where they belonged in the cases. Only by learning the different combinations of letters available to him in the cases could a compositor begin to set type efficiently. Practically, spellings are combinations of metallic sorts, and it was in that light that the compositor had to perceive them. (These devices of course reflect the enduring concern of economy: if a compositor could pick up or distribute three letters as one sort, he saved time and effort. However, there are diseconomies in expanding the case too far, and the tendency for a hundred years or so was to reduce the number of special characters the standard case accommodated, as illustrated by the large differences between Caxton's hypothetical case and Moxon's case in 1683-4.)
That an apprentice would be set to distribute not only pie but also type set by another compositor is confirmed by Hinman's analysis of compositor E's participation in the setting of the First Folio. He notes several occasions on which compositor E, besides distributing his own work, distributed type set by a different compositor, most often compositor B, with whom E usually worked. (27) It was during distribution that the new early seventeenth-century (and earlier) compositor became practically informed of the different spelling practices in the printing house among his fellows before he was allowed even to attempt composition himself. (28)
Having described in considerable detail how the compositor takes up lines of type to be distributed, Moxon continues in these words:
then he brings what he has taken off towards his Sight to read; then [ ... ] he spreads and Squabbles the shanks of the Letters between his Fingers askew; and remembering what Letters he read, he nimbly addresses his Hand with a continued motion to every respective Box, which his Fingers, as they pass by, lets a Letter drop into, till his Taking off be quite Distributed. (pp. 201-2)
No doubt early compositors trained their eidetic memories to remember the individual spellings of the lines they were assigned to distribute, and visualized the combinations of sorts (ligatures etc.) that made up the line of type. As Knowles explains, 'What has always mattered to skilled readers, processing the text for meaning rather than form, is efficient word recognition' (p. 90, emphasis added). Fast reading requires rapid recognition of word-shapes as a whole, and there is no time to map individual letters onto the sounds of speech. (29) The necessity to memorize variant spellings made distribution a difficult task to accomplish at all. However, distribution did not depend entirely on the compositor's memory: each individual sort had its own characteristics and a compositor would feel when he took the sort into his right hand whether it was, say, a thin space when he expected a terminal -e or an m when he expected an i. Nevertheless, distribution would become most efficient when spellings were uniform, when the words the compositor read were in his spelling, when--despite his regional origin and dialect--he spelt by and large in the same way as everyone else in the printing house. Timperley strikingly illustrates the benefits of standard spelling in distribution:
With many compositors much time is unnecessarily lost in looking at the word before they distribute it. By proper attention in the learner he may avoid this. [ ... ] By proper attention and practice he will become so completely acquainted with the beard or beak of the type, as to know the meaning of the word he takes up, with the cursory view he may have of it, while in the act of lifting it. (p. 15, emphasis added)
We see, then, that a hundred or so years after spelling had become largely uniform, a distributing compositor need not bring 'what he has taken off towards his Sight to read' (Moxon, p. 202) but need only take a 'cursory view' of the text while lifting it. This significant increase in the efficiency of compositors is not only a product of a process of spelling standardization in the printing house; it is also a significant cause of it.
It is unnecessary to speculate what compositor E's preferred spellings of such common words as do, go, here were before he started to distribute and compose under compositor B's experienced eye. Hinman noticed that at the start of their collaboration E's spellings were the same as B's for the common compositorial discriminants do, go, and here (i.e. heere) but that the novice was more likely to reproduce the spellings of his printed copy, putting his own preferred forms in abeyance. There were other spellings that served to distinguish the apprentice from the journeyman, notably -ie- in griefe and its cognates. One can only suggest rather than prove that compositor E learnt his mentor's most frequent preferred forms, and it is possible to claim some measure of support from the observation that some of E's common spellings changed during his stint on the Folio, in the direction of modernity. For most of the Folio both B and E used both the short and the double-e spelling in words like he, me, she, we and their cognates, but by the last plays of which he shared the composition with B, compositor E had moved to employ the shorter variants almost exclusively. Apparently composition as well as distribution modified his spelling preferences--and in the direction of modernity.
It is well known that when setting type by hand, compositors selected sorts from their cases, letter by letter, making up and justifying lines of type (i.e. filling the measure) in their composing sticks before removing them eventually to a galley for eventual imposition. Initially, apprentices had to be shown how to set their composing sticks to the desired measure (unless they were using fixed-width wooden sticks), how to hold the stick in their left hand with a setting rule to ease justification, and how to hold the incomplete line of type secure with their left thumbs, while selecting type and placing it in the stick with the right hand. Most important, they had to be shown how to fill the measure tightly with types, quadrats, and spaces so that the lines of type would lock up firmly in the forme. (When they did not, the result was instant pie.) There is no need here to go more deeply into composition: details are readily available in Moxon and other manuals of the hand-printed book era.
The necessity to translate his copy from one medium to another obliged the compositor to consider spellings intensely, not only for the purpose of casting-up his copy (which process will be discussed later) but in order to accommodate the text to the assigned or determined format. In about the first two centuries of printing in England many of the spellings of his copy were not those that a compositor habitually employed. Consequently, even when setting a close reprint of printed copy, when the format, measure, and fount were identical, by varying the spellings of his copy the compositor created fresh typographical situations with which he had to cope. The spellings of his copy did not appear to the compositor as the product of a system for the representation of sounds but rather as a more or less plastic medium whereby he might convey the text into a new print format through mechanical means. The notion that in early times ordinary compositors knew of or attended to theoretical schemes of spelling reform--or would recognize a reformed spelling when they saw it--is untenable: there is no evidence at all that compositors generally considered spelling as a system at all. When Moxon in 1683-84 referred to the 'task and duty incumbent on the Compositer, viz. to discern and amend the bad Spelling and Pointing of his Copy' (p. 192), the movement towards standard or 'present traditional' spellings--which it is my object to explain--was largely accomplished: a 'bad Spelling' was an unmodernized, non-standard spelling.
Copy was adjusted for printing by justification. When considering justification, most of us think of prose where lines of text extend to the full width of the measure, but there may be long lines of verse that require to be justified, and indeed all lines of text, whether short or long, require justification by nonprinting sorts: a space is a piece of metal. However, justification requires most skill when text has to be accommodated to the measure in 'long' lines, i.e. lines of text that extend to the full extent of the measure. Early compositors had a variety of means available to them in order to fit their text to the measure, including the use of spaces thick and thin and contractions such as the ampersand, but the most common method was to vary the spelling of the copy: bedde could be set as bedd or bed in order to fill out a line so that it did not end with white space, without affecting meaning or pronunciation.
Linguists have noticed spelling variation for the purposes of justification but, by failing to examine it in bibliographical situations, have exaggerated the extent of its influence on orthographical variety. They may not have appreciated that such variations as that between the -ee- and -ie- spellings of grief might have been used in order to secure the benefits for justification of employing a wider or narrower character. In fact, there are very few compositorial spellings in justified lines that were not acceptable spellings at the time they were made. Analysis of spelling variants in concordances to Q1 Titus Andronicus (1594) and A Midsummer Night's Dream (1600) (30) gives the results shown in the accompanying tables. (31)
A remarkably small number of spellings can be identified confidently as the product of variation for the purpose of justification. On this evidence, justification in the printing house is not a major cause of the inconsistency of early English spelling. Those who have claimed otherwise have been looking at all spellings in long or justified lines, not merely at the spellings that compositors varied in order to fit the text to the measure. (1) Justification by contraction is the most favoured method (other than the use of spaces, of course) in each text. (2) The removal of a redundant terminal -e used mostly to indicate a long vowel, and the reduction of double consonants, are the two most favoured methods of justification in both texts. The significance of this for the movement of spellings towards modernity can readily be appreciated. (3) So useful was the removal of a superfluous terminal -e for justification that the addition of a redundant -e was also an occasional device employed to justify a long line.(32) (4) The difference of the incidence of justification between the two texts may be explained by different compositorial practices.
In short, from his senior mentor the apprentice would learn not only what typographical devices he could use in justification but also what spelling variations were possible in order to fit the text to the measure. In this context, the apprentice's spellings would tend to conform to those of his instructor. Both of them were obliged in the practice of their craft to consider the nature of the spellings in their copy and their deployment in the practical and technical (not theoretical) context of their craft. Spelling was the primary material at the command of the early compositor, enabling him to produce a readable book. Variation of spellings was an inescapable professional obligation until such time that spellings lost their personal character and became 'correct' and customary, therefore fixed. Only after both personal and compositorial spelling had become as near to standardized as possible could the injunction to compositors to 'follow copy' have any force. (Then, compositors set what they had before them and readers corrected aberrant spellings; later, copy-editors more efficiently regularized the manuscript before it went for composition. Later still, computer spell-checkers enforced conformity in authorial typescripts, and it is not impossible to conceive that sooner or later computer word-processing programs will 'correct' punctuation, grammar, style, and content.)
By perceiving the printing process itself as the essential cause of spelling standardization, we avoid the need to arbitrate between the claims of spelling books, grammars, dictionaries, teachers, and so on to have brought it about: all physical means of spelling instruction were printed, and all printers employed the same methods to communicate their texts in acceptable orthography. Every book printed bore witness to the reformed orthography, as it does now. The standardization of spelling occurred in the printing house, at first unselfconsciously, when (unavoidably) motives of efficiency, through which any technology advances, came into play.
In order to get to the heart of my argument I earlier omitted discussion of the first stage in the production of a book at which efficiency is of primary importance: casting-off the copy, manuscript or printed, in order to ensure the best use of the resources available to the printer. Moxon provides a detailed account of casting-off, with the bases of calculations to estimate how the handwriting of copy can best be translated into an estimate of the type required, taking account of breaks and different founts (e.g. italic and black letter) and the necessity to expand authors' abbreviations. He does not consider the question of how different variable spellings in copy might affect the correctness of the estimate. That may be because, writing as late as 1683, when spelling was mostly standardized, he did not expect much variation in copy, or, more likely, because when the compositor 'Composes one Line in his Measure' (p. 240), selected to represent the character of the manuscript fairly, Moxon knew that in that line and throughout subsequent composition, the compositor would employ the by now standard spellings. Because the test line to aid estimation was set in the compositor's spelling, we need not regard casting-off as a primary factor in spelling standardization; but, again, it was certainly a beneficiary of it.
Nevertheless, in the early years of English printing variation of spellings among compositors might have been perceived by frugal masters as diseconomic. It was not until about 1774 that piece-work (or payment by the total length of type, expressed as a number of ens a compositor set) was introduced into the London trade, (33) under which system of payment the speed of the compositors was crucial to their making adequate wages. Because spellings were mainly standardized by then, compositors could not increase the output and thence their pay by shortening spellings. However, the early masters might well have observed that although a compositor whose spellings were consistently longer than those of another may have set more ens of type, the compositor with the more economical (i.e. modern) spellings actually set more text in a similar period.
In considering a compositor's operations I have necessarily proceeded synchronically. It is at last desirable to outline how the influence of distribution and composition illustrated here by a single apprentice compositor multiplied from one compositor to another, from printing house to printing house. Some of the figures I have mentioned for different texts show that modernization did not proceed at a consistent rate or uniformly from edition to edition. There were different intervals of years between editions, and without validation from diachronic studies the texts chosen need not be considered fully representative. Yet the overall movement towards modernization is clear. It may be associated--though not securely until further investigations are undertaken--with the increase of the number of printers from Caxton onwards, the increase of books issued from the presses, and thus the increasing number of compositors influenced by the factors I have discussed above.
The volume of entries for titles or editions published each year tabulated in Philip R. Rider's 'A Chronological Index' (STC, vol. III) affords a rough visual index to the volume of printing, increasing in England almost without respite from its beginning. The entry for 1477 occupies 3 1/2 lines, that for 1640 over 1 1/2 pages. John Barnard and Maureen Bell provide statistical tables of different forms of publication extending as far as 1700 in the Cambridge History of the Book in Britain. (34) There appears to be a positive relationship between the rate of standardization of spellings and the increase of books printed.
The most significant instruments of spelling uniformity were the journeymen compositors--increasing in numbers year by year--many of whom, having served out their time, were unable to find permanent employment and were therefore obliged to move from printing house to printing house. Had compositor E completed his apprenticeship, he might have been forced to leave Jaggard's establishment, taking with him his do, go, and perhaps a 'reformed' here spelling, as well as the efficient single-e pronouns, to a different printer's. There he would employ his more efficient, i.e. more modern, spellings, perhaps passing them on to a learner in the craft, before moving on to another establishment. If compositor E is multiplied by hundreds of compositors from the 1620s (and extrapolated backwards but in decreasing numbers), how English spelling achieved its present form from Caxton onwards is readily perceived.
The standardization of spelling occurred because printers, like every other artificer--like the medieval scribes whose work the first printers emulated--strove to perform their work more efficiently. There may be some resistance to the purportedly modern concept of efficiency being applied to printing, even though the history of technology shows that there is nothing especially modern about the notion of efficiency. Timperley illustrates the point. He is describing how the type-case is arranged so that the sorts the compositor uses most often come readily to hand:
Some may consider the distance from one box to another too trifling to demand our notice; so it really is, in an abstract point of view; but when that space, however short, is multiplied by the number of times which the hand has to traverse over the case in the course of one day, much less a week, it would make the comparative trifle amount to a space almost beyond credibility. (p. 14) (35)
Modernization was initially slow to emerge in the print-shop but accelerated with the growth of the printing trade.
Printers employed their spellings, readers were influenced by them and employed them in their writings (inconsistently at first), printers standardized the older spellings of their copies, by then unacceptable, and published more books, and so on. Schoolmasters taught the reformed printers' spellings that were printed in their spelling books, grammars, and dictionaries and all such influences, working together under the aegis of printing and the operations of the printing house, brought about the graphemic--but not phonemic--uniformity we experience today.
The following abbreviations are used: STC = A. W. Pollard and G. R. Redgrave, Short-Title Catalogue of Books Printed in England, Scotland, and Ireland, and of English Books Printed Abroad, 1475-1640, 2nd edn, rev. by W. A. Jackson and others, 3 vols (London, 1976-91);TLN = Through-Line-Number (in Shakespeare First Folio); Wing = D. G. Wing, Short-Title Catalogue of Books Printed in England, Scotland, Ireland, Wales and British America, and of English Books Printed in Other Countries, 1641-1700, rev. by J. J. Morrison and others, 4 vols (New York, 1982-98).
(1) Besides spelling, orthography, defined in 1616 as 'the art of writing words truely', includes punctuation, capitalization, and such devices as the use of the characters i and j or u and v with a single meaning in early printing. My concern here is with spelling alone. D. G. Scragg, A History of English Spelling, Mont Follick Series, 3 (Manchester: Manchester University Press; New York: Barnes and Noble, 1974), is the most comprehensive authority.
(2) 'English spelling seems to have been particularly resistant to the interference of linguistic philosophers' (Scragg, p. 81).
(3) In his The First Part of the Elementarie, Which Entreateth Chefelie of the Right Writing of Our English Tung (London: T. Vautrollier, 1582) (STC 18250)--the only edition before 1700.
(4) Robin C. Alston, Spelling Books, A Bibliography of the English Language from the Invention of Printing to the Year 1800, 4 (Bradford: printed for the Author by E. Cummins, 1967). Note: these are STC 5711-16; Wing C6067-78a; the last edition Alston records is the 54th, printed in 1737. From the 1614 edition Coote was published by the Company of Stationers as part of the English Stock--apart from the sole Dublin edition--which suggests that the schoolmaster was dead by 1614.
(5) The largest classes of variation were doubled consonants (as in alleddge), terminal -ie for y (as in accessarie), and the use of terminal -e to indicate a long vowel, usually redundantly (as in arraigne).
(6) There is convincing evidence that the successive editions--at least those surveyed here--were, from the 1630 edition, merely reprints. Gross but not immediately obvious misprints were taken over from edition to edition. As the illustrative spellings changed their form, they did not, however, change their places in the alphabetical list: by the 1696 edition the order of spellings is seriously misleading and the list is difficult to use.
(7) N. F. Blake, 'English Versions of Reynard the Fox in the Fifteenth and Sixteenth Centuries', Studies in Philology, 62 (1965), 63-77.
(8) He also observed that 'Although in WC [i.e. 1481] -wh and -ugh are found finally in such words as though and enough, by the time of TG [i.e. 1550] only -ugh is found and most versions show a gradual spread of -ugh forms over -wh ones' (p. 71). And a 'similar trend to standardisation is apparent in the use of the graphemes -tch [as in fetche] and -dg- [as in pledge]' (p. 68).
(9) The sample consisted of Tit. Folio TLN 666-94, 1228-50, 2566-616 = 101 lines in the quartos; and MND TLN 420-505 and 1560-75 = 99 lines in the quartos.
(10) Compositors followed copy in Tit. for atchieve (668), Didoes (2586), Emperours (2: 689, 2602), forrest (677), cursie (2578), chappes (2581), and knives (knife's, 2566), and in MND for tailour (425), forrest (458), dewlop (421), queint (474), Rheumaticke (480), Hyems (484), Tytania (434-5), and flouriets (flowerets, 1570).
(11) D. F. McKenzie, 'Apprenticeship in the Stationers' Company, 1555-1640', Library, ser. 5, 13 (1959), 292-99 (p. 298); Paul Morgan, Warwickshire Apprentices in the Stationers' Company of London, 1563-1700, Dugdale Society Occasional Papers, 25 (Leeds: printed for the Dugdale Society by W. S. Maney, 1978), pp. 1-2; he also gives totals by county of origin to 1700 (p. ).
(12) 'The Stationers' Company in the Civil War Period', Library, 5th ser., 13 (1958), 1-17 (p. 3).
(13) Ellic Howe, The London Compositor, 1785-1900 (London: Bibliographical Society, 1947), p. 16. Philip Gaskell, A New Introduction to Bibliography (Oxford: Clarendon Press, 1972), citing a document in the Public Record Office, states 'There were only 198 printers altogether in London in 1668, of whom 75 per cent were journeymen, 13 per cent masters, and 12 per cent apprentices. The total number of printers in London around 1600, when there may have been 40 presses at work, was probably under 175' (p. 176). For figures charting the increase in the number of London presses--and therefore of compositors--see also The Cambridge History of the Book in Britain, iv: 1557-1697, ed. by John Barnard and D. F. McKenzie, with the assistance of Maureen Bell (Cambridge: Cambridge University Press, 2002), p. 18 n. 110.
(14) Mechanick Exercises on the Whole Art of Printing (1683-4) by Joseph Moxon, ed. by Herbert Davis and Harry Carter (London: Oxford University Press, 1958).
(15) Marjorie Plant, The English Book Trade: An Economic History of the Making and Sale of Books, 3rd edn (London: G. Allen and Unwin, 1974), pp. 86-7, gives details of 'The Size of the Firm'; see also H. S. Bennett, 'The Printers', in English Books and Readers, 1475-1557: Being a Study of the Book Trade from Caxton to the Incorporation of the Stationers' Company, 2nd edn (Cambridge: Cambridge University Press, 1969), pp. 178-97, for an overview of the condition of the earlier printing trade.
(16) 'A List of Printers' Apprentices, 1605-1640', Studies in Bibliography, 13 (1960), 109-41, substantially reprinted in Stationers' Company Apprentices, 1605-1640 (Charlottesville, VA: Bibliographical Society of the University of Virginia, 1961). This does not mean that other workmen were not employed, who had not been bound; many journeymen compositors did not find permanent positions and drifted from shop to shop, without leaving traces in archives.
(17) Charlton Hinman, The Printing and Proof-Reading of the First Folio of Shakespeare, 2 vols (Oxford: Clarendon Press, 1963), II, 513.
(18) Hinman places this 'early in 1622' (I, 342).
(19) 'The Prentice Hand in the Tragedies of the Shakespeare First Folio: Compositor E', Studies in Bibliography, 9 (1957), 3-20.
(20) '6. "The Prentice Hand"', in Printing and Proof-Reading, I, 214-26. The extent of E's participation in F1, including the composition of plays from manuscript copy, is enlarged to tragedies with manuscript copy by T. H. Howard-Hill, 'New Light on Compositor E of the Shakespeare First Folio', Library, 6th ser., 2 (1980), 156-78; ibid., 4 (1982), 328.
(21) Standardization was an ongoing process, as it is today, and affected different kinds of spelling at different rates. N. E. Osselton documents the accepted differences between private documents such as letters and the increasingly standard 'formal' spelling system of eighteenth-century printers ('Informal Spelling Systems in Early Modern English, 1500-1800', in English Historical Linguistics: Studies in Development, ed. by N. F. Blake and Charles Jones, CECTAL Conference Papers Series, 3 (Sheeld: Department of English Language, University of Sheeld, 1984), pp. 123-37), leading to a notion familiar to modern copy-editors that authorial spellings are always less reformed than those employed by the press. Incidentally, he indicates that the change from -all to -al in unstressed final syllables of words like general, critical, and rural was completed between 1640 and 1680, and finds that 'the final double consonant in monosyllabic words with a short vowel such as cutt and fitt' declined rapidly from 1580 'to about 1620, and [ ... ] vanished by mid-century' (p. 132): these are economical spellings.
(22) 'Ancient Customs Used in a Printing-House', in Moxon, Mechanick Exercises, ed. Davis and Carter, pp. 323-31.
(23) C. H. Timperley, The Printers' Manual, Containing Instructions to Learners, with Scales of Impositions, and Numerous Calculations, Recipes, and Scales of Prices in the Principal Towns of Great Britain: Together with Practical Directions for Conducting every Department of a Printing Office (London: H. Johnson; Manchester: Bancks and Co., 1838), p. 12. Moxon writes at length about the process of distribution but not about who did it.
(24) Philip Gaskell notes that 'individual printers are likely to have varied the patterns of their lay in minor respects' ('The Lay of the Case', Studies in Bibliography, 22 (1969), 125-42 (p. 126)) and points out that Moxon's lay, the first English illustration of the later divided lay, lacks small capitals (p. 128) .Moxon himself observed that cases 'are not in every printing house disposed alike' (p. 194). Printers' attempts to improve the arrangement of type-cases are documented in the later literature of the press. Timperley, who illustrates an old and a new lay, also observes that 'cases partially differ in most offices' (p. 13).
(25) N. F. Blake, Caxton: England's First Publisher (London: Osprey, 1976), p. 94.
(26) '"Not wreton with penne and ynke": Problems of Selection Facing the First English Printer', in Edinburgh Studies in English and Scots, ed. by A. J. Aitken [and others] (London: Longman, 1971), pp. 29-53 (p. 37). She also notes that 'the reduction in Caxton's stock of ligature-characters as his work progressed is one of the clearest trends in the evolution of his founts. For where his first type has about seventy-five different combinations of letters in ligature, his last cuts this figure down by two thirds' (p. 43). Caxton's founts became smaller (and less expensive), the type-case had fewer compartments, distribution and composition became easier and, one thinks, faster, a movement of economy foreshadowing the later development of modern founts and type-cases.
(27) Hinman, II, 193, 206, 258, 282, 283-84.
(28) Fundamentally, distribution was necessary to replenish the cases with types. Cases could hold only limited amounts of type. In the early period, in general terms, in order to provide employment, printed formes were not allowed to be left standing for use for other impressions, the size of which was limited.
(29) Gerry Knowles, A Cultural History of the English Language (London: Arnold, 1997; 2nd impr. 1998), pp. 60-1, 83-91, gives a refreshingly lucid overview of language changes, including spelling, in the early period. If compositors had spoken the words out loud to aid their recollection of the line, there would be a tendency to distribute the lines as if they contained their own habitual spellings, and so they would foul the case.
(30) 'Titus Andronicus': A Concordance to the Text of the First Quarto of 1594, Oxford Shakespeare Concordances (Oxford: Clarendon Press, 1972); 'A Midsummer Night's Dream': A Concordance to the Text of the First Quarto of 1600, Oxford Shakespeare Concordances(Oxford: Clarendon Press, 1970).
(31) The numbering of the following observations relates to the superior numerals in the tables (spp. = spellings).
(32) It should also be noted that variations of syllabic -ed for the sake of the metre (as at MND dischargd ----- 2145; cf. discharged 2008) have not been counted as justification variants.
(33) The London Compositor: Documents Relating to Wages, Working Conditions and Customs of the London Printing Trade, 1785-1900, ed. by Ellic Howe (London: The Bibliographical Society, 1947), p. 19 n. 3; Ellic Howe and Harold E. Waite, The London Society of Compositors (Re-established 1848): A Centenary History (London: Cassell, 1948), p. 43.
(34) IV, 779-93. Bell offered a foretaste of these in 'A Quantitative Survey of British Book Production, 1475-1700', in The Scholar and the Database: Papers Presented at the CERL Conference Hosted by the Royal Library, Brussels, ed. by Lotte Hellinga, CERL Papers, 11 (London: Consortium of European Research Libraries, 2001), pp. 15-21. The same volume contains discussions of the difficulties of quantifying book production. There is no disagreement, however, that it increased continually from 1475 and that later in the seventeenth century there were significant increases in the number of presses and, therefore, of compositors.
(35) Those who think that Timperley's observation is exaggerated may be instructed by the following report from a European statistician in 1884: 'an expert printer, working ten hours a day, sets up 12, 000 letters (this is allowing time sufficient for distributing and correcting), and counting 300 working days to the year [ ... ] the compositor's hand in a twelvemonth makes 3, 600, 000 movements. Estimating the distance from the case to the stick and the stick to the case at two feet, [ ... ] the total distance [is] 7, 200, 000 feet. [ ... ] the distances travelled under these conditions by a printer's hand is, in round numbers, 1364 miles a-year, or over four and a half miles a-day' (Scottish Typographical Circular, ser. 3, 11 (1884), 54).
UNIVERSITY OF SOUTH CAROLINA
|Printer friendly Cite/link Email Feedback|
|Publication:||The Modern Language Review|
|Date:||Jan 1, 2006|
|Previous Article:||The teleological element in Richard Rolle's Contra amatores mundi.|
|Next Article:||Genius loci: placing place in Gerard Manley Hopkins.|