Paradigm, No 4 (December, 1990)


Textbook design: chronological tables and the use of typographic cueing

Michael Twyman

University of Reading,
Department of Typography and Graphic Communication,
2 Earley Gate,
PO Box 239,
Reading RG6 2AU

The theme that links the two strands of this summary paper is the emphasis that seems to have been put on learning through the eye in the second half of the 18th century and throughout the19th century. One publication serves as the focal point of the paper: James Bell’s Compendious view of universal history and literature, in a series of tables (London, 1820), which ran to four further editions, the last of which appeared in a substantially improved form in 1842 with the title A view of universal history, literature, and the several schools of painting.

Little is known about James Bell, who describes himself as a Major in the East York Militia, but his chronological tables achieved some success in the second quarter of the 19th century. Lists of subscribers were published with three editions (1824, 1829, and 1833) and included an impressive array of schoolmasters, university teachers, and private tutors (among them the Tutor to the Duke of Wellington’s sons at Eton). In 1833, before the publication of the third edition, Bell gave an ‘Explanatory and demonstrative lecture’ based on his approach to teaching history to the young Princess Victoria (1833 ed., Dedication). Nearly 20 years later copies of the fifth edition of 1842 were put on display at the Great Exhibition (class 17, no.203), where one in roll form covered a wall space of 16 x 14ft.1

The fifth and most impressive edition of Bell’s publication consisted of 25 mostly double-page tables in Royal folio format and cost œ 2.10s (œ 2.5s to subscribers). Mounted on calico, varnished, and attached to rollers for display on walls, it cost œ 5.5s.2 Sales were not large by the standards of main-line text book publishing:3 the first four editions jointly accounted for 5,500 copies (1842 ed., p.9), and all five together for 7066.4 All editions were handsomely printed by the firm of T. C.Hansard and the contribution of the printer to the very difficult job of keeping the entries in the columns more or less in line with one another chronologically was generously acknowledged by Bell (1820 ed., p.vii). The presentation of the tables was of the utmost importance and substantial graphic improvements were made to them, principally to those of fourth edition of 1833, when hand-colouring was added to code the columns of information relating to different nations, and to those of the fifth edition of 1842 when, in addition, different kinds of type were introduced to code significant textual elements. In 1853 Bell wrote, and published himself, two explanatory books outlining his approach to the use of mnemonics in learning: The first-sight phenomena of historic time, which he described as an ‘elucidatory and illustrative companion’ to the fifth edition of his tables, and a supplement to it, The tabular teacher of the theory and thrift of combinative and corroborative historic memory.

It has yet to be established how influential Bell’s tables were educationally and typographically. All that is claimed for them here is that they exemplify, probably as effectively as any other publication of the period, a substantial shift that took place in the design of educational material in the 150 years or so before the introduction of the fully illustrated textbook in the 20th century. The specific changes to be seen in Bell’s tables relate to what I have referred to in other contexts as the ‘intrinsic’ and ‘extrinsic’ features of graphic language.5 By intrinsic is meant those features that reside in the characters or signs themselves and in the system that is used to produce them: they include the range of characters available in their basic form, the variants of these characters (e.g., italic, condensed, and bold characters), the style of the characters (typeface), and their size. By extrinsic is meant what can be done to these characters or clusters of characters (words, lines, sets of lines) by organizing them in relation to one another to encourage particular reading strategies. The most significant of these extrinsic features is probably the ‘configuration’, or the way clusters of characters are organized. Five configurations can be identified as having played a significant role historically in the effective communication of verbal information graphically: the linear interrupted configuration (commonly called continuous text), list, branching, and matrix configurations, and a much broader category I have referred to as ‘non-linear directed viewing’.6 Other extrinsic features have to do with the amount of space used at a macro-level between large clusters of characters (that is, at the level of the line and above) and at a micro-level between and around characters and small clusters of characters (such as words). This brief diversion into theoretical matters is necessary in this context only because it emphasizes the significance of Bell’s considered use of intrinsic and extrinsic graphic features in his chronological tables.7

Bell’s arrangement of historical information in tabular form (a matrix configuration) was not at all new. He made it clear that his Compendious view of universal history was based on the historical tables of G. G. Bredow (presumably an earlier edition of his Merkwrdige Begebenheiten aus der allgemeinen Weltgeschichte (11th ed., Alton 1820). In any event, there already existed a long history of published chronologies dating back to J. J. Scaliger in the late-16thcentury. From the following century onwards there appeared in Europe, and later on in America, a steady flow of tabular arrangements of historical material, either set in type, engraved intaglio on metal or, in the 19th century, lithographed.9 By the close of the 18th century such approaches were so widespread that a lively debate ensued over whether time should be resented on the vertical axis (which was Bell’s approach) or on the horizontal axis, which was the approach advocated by Joseph Priestley in his A description of a chart of biography (Warrington, 1765) and A description of a new chart of history (London, 1770). The metaphors ‘river of time’ and ‘stream of time’ were frequently used to describe such horizontal and vertical displays of time on chronological charts.

The earliest chronologies merely attempted to put history on a sound footing by basing it on a sequence of known, or supposedly known, events. By the second half of the 18th century, and particularly with the publication of Priestley’s very successful ‘Chart of biography’ and ‘New chart of history’ in the second half of the 1760s, emphasis had shifted towards chronologies with a stronger didactic basis. Priestley’s charts grew out of his own teaching at the Dissenting Academy in Warrington and were originally produced solely for his own purposes; but they reached a large audience when engraved and published, and continued to appear into the 19th century.

Priestley’s contribution to the teaching of history was part of a more general movement towards communicating factual information through the eye that had its parallels in the scientific world with the development of line graphs by Johann Heinrich Lambert in Germany in the 1760s and 1770s and in economics and business affairs with William Playfair’s use from the 1780s of what he called ‘lineal arithmetic’ in the form of line graphs and bar charts. What we know today as ‘information graphics’ seems, therefore, to have emerged in the late-18th century on a broad front. This important cultural shift has yet to be satisfactorily explained: clearly it relates to the growth of science and the Enlightenment, but it cannot be divorced from a whole set of changes in society brought about by the Industrial Revolution, not the least of which was the need for people to communicate more effectively in all walks of life.

What these new graphic approaches had in common is that they made relationships between sets of information visually explicit and so, it was argued, helped people to understand and remember them. These points were made in different ways by numerous writers concerned with graphic displays of information in the late-18thand early-19th centuries. In the texts that accompanied chronological charts, analogies were frequently made with maps and cartography, particularly in relation to the use of colour for the purpose of coding information. Bell’s chronological tables used colour for the first time in 1833, but several other historical charts were hand-coloured long before this, probably the earliest being an anonymous ‘Chart of universal history’, which was engraved and published by Thomas Jefferys in London in 1753. Bell’s special contribution was to explore the use of colour more fully than earlier compilers of historical charts: he used over 30 different colours in the 1842 edition of his tables and provided a key on a throw-out page to help the reader interpret them.

Bell’s contribution to the use of intrinsic features of language was made explicit on the title-page of the 1842 edition of his tables, where he referred to his introduction of ‘variegated forms of type’. As far as I know, the use of the word ‘variegated’ was new in this context; what he meant by it was the use of different sizes and kinds of type to draw attention to specific elements in a text. The idea of cueing significant words in a text had its roots in Egyptian hieroglyphic inscriptions and manuscripts, and was explored more fully in mediaeval manuscripts. These early approaches to cueing in texts relied on colour, conventionally red; but with the invention of printing this became expensive and was restricted mainly to books used in church services. The alternative to colour offered by printing was italic type, which began to be used by Robert Estienne and others in France from the 1520s as a means of coding information within a text set in roman (upright) type. The convention of setting titles of publications in italic dates from these early-16th century approaches to coding information units. Capital letters and different sizes of letters (including small capitals) were also used for coding parts of texts from the 16th century onwards. These conventional methods of typographic cueing were employed by Bell in the first four editions of his tables, but in the fifth edition of 1842 he introduced ‘variegated’ type and developed the idea of typographic cueing to an altogether new level of complexity. Bell’s approach to variegation -- which must surely have been discussed fully with his printer, T. C. Hansard -- involved the use of a wide range of different sizes of type, some of which were more than twice the size of the type used for the main body of the setting. What is more, Bell used some of the bold sanserif and bold slab serif types that had been introduced fairly recently by British typefounders, claiming that they offered ‘greatly increased Advantages of Attraction to Vision, and consequent Impression on Memory’ (1842 ed., p. 9). Such types allowed Bell to reinforce the mnemonic approaches to learning that he had adopted even in the first edition of his tables. Though the term ‘variegated type’ seems to be his, he was not the first to make use of the idea, even in historical tables. The Oxford printer/publisher D. A. Talboys did so a few years before Bell in his series of three Chronological tables of history (Oxford, 1835-40), in which, incidentally, he refers to Bell’s tables in a complimentary way. However, Talboys appears to have used the idea of variegation with rather less conviction than Bell and made no special issue of it.

Together, Talboys and Bell were responsible for initiating an approach to cueing information using bold type which was taken up with enthusiasm later in the 19th century in textbook design generally, and also in other areas of printing and publishing (such as timetables, directories, catalogues, and dictionaries). The long-term consequences of their innovation were not to be fully realized until the 20th century when typefaces began to be designed for the first time with related bolds; but the first step towards designing a bold face that looked well with text faces was made very soon after the publication of Bell’s tables when Robert Besley of the Fann Street Foundry in London registered a type under the name Clarendon in 1845.10 This was a slightly condensed bold typeface with bracketed serifs. Its popularity as a ‘bold’ to be listed in text can be gauged by the fact that it was copied by other typefounders as soon as its three-year period of registration was over. And before long the term ‘Clarendon’ was being used loosely to describe all bolds used in text. At first sight Clarendon types may not appear very different from the slab serif bold types used by Talboys and Bell, but close inspection reveals that they related to most text types of the time very much better than those earlier ‘bolds . In doing so, they paved the way for a distinct change of dress in textbook design in the second half of the 19th century.

This change of dress involved the use of one or more sizes of bold type (often, though not always, proper Clarendons) for headings and other words that needed to be emphasized in the text. Bolds were often so generously used that distinctions between headings and words emphasized in the text were sometimes blurred; they were also frequently used in lists and other non-prose configurations. All this suggests that the writers of such textbooks must have had the use of bolds in mind and that, in turn, they probably had an influence on teaching methods.

Anyone with an interest in late-19th century textbook design cannot have failed to notice the use of bolds outlined above. This form of design can be seen across a wide range of subjects, but seems to have been particularly favoured by those concerned with geography, history and language teaching. J. M. D. Meiklejohn’s A new geography on the comparative method (London, 1889) is a particularly good and well-known example of the genre, and appears to have been an outstanding success commercially. When this change of direction in textbook design began is not yet clear, but by the begrinding off the last quarter of the century the use of bolds (and to a lesser degree of structured texts) was, well established in Britain. It remained a characteristic of some areas of textbook design, possibly as a result of inertia, through to the 1920s. It may have been brought to an end as much by the increasing use of machine composition (which was at its most economical when producing continuous text with little variegation) as by the significant shift in educational ideas in the interwar years.

The use in textbook design of tables, lists, and other non-prose configurations (extrinsic features of graphic language) and of bold variants of type (intrinsic features of graphic language) have in common -- at least as an aim -- that they help to reinforce learning through their visual presentation. A familiar thread running through the introductory texts of the numerous historical tables published in the late-18th and 19th centuries is that they enable information to be impressed on the mind through the eye. This approach to learning (memorising) has a very long tradition dating from ancient times and is well documented by Frances Yates in The art of memory (London, 1966). Bell’s adoption or variegated types for the 1842 edition of his tables seems very much in the spirit of Quintilian’s recommendation that learning a passage by heart should be done from the same tablets on which it had been written, so that ‘the mind’s eye will be fixed not merely on the pages . . . but on individual lines’.11

In the second half of the 18th century and throughout the 19th century printing in general displays a move away from gray rectangles of type towards the articulation of intrinsic and extrinsic features of graphic language for specific purposes. This paper suggests that this pattern is to be seen in textbook design. In the case of textbooks, However, the purposes of this greater articulation may not be absolutely clear. In correspondence with me after the meeting at Reading, Chris Stray referred to differences between ‘memorial’ and ‘reference’ learning in the possible shift in the mid-19th century from the one to the other. Does this help to explain the greater use of typographic devices that help articulate texts, such as tables and lists, and the use of bolds for cueing? And did teaching methods change as a result of changes in textbook design, or were these changes a response to what was already happening in teaching? Or does textbook design merely reflect broader changes in society and the greater emphasis put on structured typography because of the increase in printed matter generally and the consequent need to speed up the ‘transfer’ of information? A greater awareness of precisely when textbook design began to change in the direction described here, and also in relation to particular kinds of texts, might help answer some of these questions. Notifications of ‘sightings’ of Clarendons (and other bolds) in textbooks before about 1870 would therefore be greatly appreciated.



1. Bell, J. The tabular teacher of the theory and thrift of combinative and corroborative historic memory (London, 1853), Advertisement, pp. 217-218.

2. Ibid. p. 220.

3. Michael, I. ‘Aspects of textbook research’. Paradigm, no.2, March 1990, p. 7.

4. Bell, J. The first-sight phenomena of historic time (London, 1853), Dedication.

5. Twyman, M. ‘The graphic presentation of language’. Information Design Journal, vol.3, no.1, 1982, pp. 2-22.

6. For a discussion of graphic configurations, see M. Twyman, ‘A schema for the study of graphic language’. In Kolers, P. A., Wrolstad, M. F., & Bourna, H. (Eds.), Processing of Visible Language vol.1, (New York & London: Plenum, 1979), pp. 117-150.

7. These features are put into a broader historical context in M. Twyman ‘Articulating graphic language: a historical perspective’. In Wrolstad, M. E. & Fisher, D. E. (Eds.) Toward a new understanding of literacy (New York: Praeger, 1986), pp. 188-251.

8. Grafton, A. T. ‘Joseph Scaliger and historical chronology: the rise and fall of a discipline’. History and Theory vol.14, no.2, pp. 156-185.

9. For references to some of these, see Twymam, ‘Articulating graphic language’, and under such headings as Ancient History, Modern History, Chronology, and General History in the various editions of the Catalogue of the Educational Division of the South Kensington Museum (1857 to 1876). See Chris Stray Paradigm, no.3 (July 1990), p. 14.

10. Mosley, 1. ‘An essentially English type’. Monotype News Letter no. 60, 1960, pp. [6-8].

11. Quoted in Yates, F.A., The art of memory (London, 1966), p. 25.

Paradigm Catalogue Textbook Colloquium