-Van Leeuwen(2006)_Towards a Semiotics of Typography

Information Design Journal + Document Design 4(2), 39–55 Towards a semiotics of typography © 2006 John Benjamins Publishing Company

Theo van Leeuwen Towards a semiotics of typography

The new writing

of texts in terms of the communicative work they do. At the level of the clause, functional grammar has moved beyond the formal, structural analysis of sentences, allowing insight into the relations between clause structure and the communicative work that clauses do, for instance through concepts such as theme and rheme, and given and new (e.g. Halliday, 985). The problem is, just as we have developed these concepts and analytical techniques, writing itself has changed. Much of the cohesive work that used to be done by language is now realised, not through linguistic resources, but through layout, colour and typography. Consider the two text fragments below, from the UK version of Cosmopolitan magazine (September 2003, p. 49).

Over the past thirty years or so, a range of methods has been developed for analysing the coherence of English text. Influential studies of thematic structure (Fries, 983), lexical cohesion (Gutwinski, 976; Halliday & Hasan, 976; Martin, 992), reference (Gleason, 973, Halliday & Hasan, 976; Martin, 992), conjunction (Halliday & Hasan, 976; Halliday, 985; Martin, 983, 992) and other aspects of cohesion have provided effective and widely used tools for analysing coherence and cohesion in written text. And linguistic genre studies (Swales, 990; Martin, 992; Van Leeuwen, 993, 2005) have made it possible to interpret the cohesive structures

Figure . Linguistically realised text coherence (UK Cosmopolitan, September 2003, p.49)

Keywords: connotation, distinctive features, document design, ideational meaning, interpersonal meaning, semiotics, medium, mode, textual meaning, typography This article outlines a social semiotic approach to analysing the ideational, interpersonal and textual meaning potentials of letter forms, drawing on Jakobson’s distinctive feature analysis and Lakoff and Johnson’s theory of experiential metaphor. X distinctive features are recognized and applied to the analysis of examples: weight, expansion, slope, curvature, connectivity, orientation and regularity.

39

40

Theo van Leeuwen

Figure 2. Visually realised text coherence (UK Cosmopolitan, September 2003, p. 49)

In the second sentence of Figure , the relationship between the different ‘animal sex personalities’ is indicated linguistically, by the conjunction ‘either...or’, and by the first sentence which announces that ‘there are five animal sex personalities’, and so creates an implicit taxonomy. In Figure 2, on the other hand, the relationship between the different characteristics of the ‘dolphin sex personality’ is indicated visually. Each of the ‘characteristics’ has its own visual identity, its own bullet point, and at the same time visually resembles the other characteristics, creating a visual ‘classification’ syntagm (c.f. Kress & Van Leeuwen, 996, pp. 79–89). As a result we understand that they provide the same kind of information, that they all characterize the same type of ‘personality’. But there is no explicit linguistic formulation of this, no sentence announcing that ‘the dolphin sex personality has three main characteristics’ Again, in Figure , a shift in the use of linguistic resources signals a shift in what the text is trying to do, a shift to a new ‘stage’ in the unfolding of the text’s communicative work. As we move from declarative to imperative sentences, we also move from a first stage of ‘explaining the concepts’ to a second stage of ‘instructing the reader in applying them’. But there is no visual

boundary between these stages. Visually the text just runs on. In Figure 2, on the other hand, the two text elements shown, the ‘enumeration of the main characteristics’ of the dolphin, and the ‘expansion of one of these characteristics’ is indicated by a shift in the deployment of visual resources, in terms of layout (bullet-pointed text versus running text), colour (pink and black versus black only), and typography (a shift to a different weight of the same font). Finally, in Figure , the link between text and image is signified linguistically, through the sentence ‘Study the pictures on the following page’, while in the banner of Figure 2 the link between image and text is expressed by means of layout, through a ‘Given-New’ composition (c.f. Kress & Van Leeuwen, 996, pp.86–92). All this applies, not just to text structure, but also to sentence structure. An advertisement for cat food (Figure 3) shows a fluffy grey kitten lying on a soft, silky sheet. A linguistic analysis of the verbal text alone would not make much sense. But together with the pictures, the advertisement forms a kind of passive clause in which

Figure 3. Visually realised participants with verbally realised process

Towards a semiotics of typography

Figure 4. Visually realised process with verbally realised participants

Figure 5. Vicks television commercial by Jonathan Barnbrook

the process, the ‘verb’, is expressed linguistically, and the participants visually – the agent of the ‘spoiling’ by the pictures of the different kinds of cat food, and the object of the ‘spoiling’ by the picture of the cat. Note the verbalvisual parallelism here: The word ‘spoilt’ is repeated four times, corresponding to the four tins of catfood depicted at the bottom of the page. We could paraphrase ‘This kitten is spoilt by catfood a, and by catfood b ....(etc)’. In the case of Figure 4, the opposite happens. The grammar, the structure of the proposition is realised visually. The participants are demarcated as participants in the structure by means of framing and colour, and the process is expressed by means of an arrow, rather than by means of a verb such as ‘causes’, or ‘leads to’ or ‘results in’. But the ‘lexical’ content of the participants is realised verbally, through nominal groups. In other words, the ‘grammar’ is visual, and the ‘lexis’ verbal. The London typographer Jonathan Barnbrook has used this principle in a series of television commercials. In Figure 5, the elements of the clause are realised verbally, but each is given a distinct identity, defined as a distinct element by different kinds of frame, colour and typography, and these elements are connected to each other by means of lines and arrows. Again, the (‘clause’ level) ‘grammar’ is visual, and the ‘lexis’ verbal.

The problem is, concepts and methods for analysing this new kind of writing, its coherence, and hence its potential effectiveness, lag behind the techniques we have for analysing traditional writing.

The new typography Elsewhere Kress and I (Kress & Van Leeuwen, 996, 2002; Van Leeuwen, 2005) have attempted to outline methods for analysing layout and colour which can be integrated with already existing methods of analysing linguistic text structure such as those referenced above. Here I will try to apply the principles we used in this work to typography. This is a relatively novel enterprise. Most research on typography has concerned itself only with legibility. Typography was not considered a semiotic mode in its own right. In the Thames and Hudson Manual of Typography, first published in 980, McLean says that ‘to a very limited extent, lettering may help to express a feeling or a mood that is in harmony with the meaning of the words’, but for the most part ‘lettering and calligraphy are abstract arts (...) What moves us is something formal, and, in the last resort, inexplicable’ (McLean, 2000, pp. 54–56).

4

42

Theo van Leeuwen

All this is now changing. Typography is facing new challenges, as screen media such as the Internet become more and more oriented towards the written word and ‘page media’ such as books and magazines become increasingly visual. A new typography has emerged which no longer sees itself as a humble craft in the service of the written word, but as spearheading innovation in graphic design, and which no longer sees typography as an ‘abstract art’, but as a means of communication in its own right. Designers Bellantoni and Woolman (2000), for instance, write that the printed word has two levels of meaning, the ‘word image’, i.e. the idea represented by the word itself, constructed from a string of letters, and the ‘typographic image’, the ‘holistic visual impression’, and designer Neuenschwander (993, p. 3, p. 3) calls typography ‘a fully developed medium of expression’, possessing ‘a complex grammar by which communication is possible’, quoting the Swiss designer Hans-Rudolf Lutz who has said that ‘Gestaltung ist auch Information’ [‘design is also information’]. This move towards a new role for typography is not restricted to the work of professional designers, but affects all writers. The time of the relative uniformity of handwriting and, especially, typewriting, is over, and the basic tools of the typographer are now available to every word processor user. The problem is, despite the programmatic announcement of the new typographers, we do not yet have that ‘complex grammar’. And despite the fact that a number of linguists have begun to explore this new field (e.g. Myers, 994; Goodman & Graddol, 996; Crystal, 998; Walker, 2000; Cook, 200), we do not yet have a systematic framework for the analysis of the communicative work done by typography today.

Typography as a semiotic mode In Reading Images (996) Kress and I used Halliday’s metafunctional theory (Halliday, 978) to argue that

the image constitutes a semiotic mode in its own right, a kind of ‘language’. According to Halliday, spoken and written texts always, and simultaneously, fulfil three broad communicative functions or ‘metafunctions’, and specific linguistic resources, specific lexicogrammatical and discourse-level ‘systems’, can be matched to each of these three metafunctions. We set out to show that images, too, can fulfil all three of these metafunctions, and that the ‘grammatical’ resources of images, too, can be matched to specific metafunctions. To briefly gloss the metafunctions, the ‘ideational’ metafunction is the function of constructing representations of what is going on in the world (and in our minds). The most important linguistic systems which realize it, are the lexicon and the grammar of transitivity, which outlines the different kinds of processes (e.g. material and mental processes) that make it possible to create different representations of what must ultimately be the same phenomena. In images, Kress and I argued, this function is fulfilled by certain aspects of composition (e.g. Kress & Van Leeuwen, 996, pp. 79–89) and by systems of vectoriality (Kress & Van Leeuwen , 996, pp. 56–7). The ‘interpersonal’ metafunction is the function of language to constitute social interactions and express attitudes towards what is being represented. One of the lexicogrammatical resources for the former is the grammar of mood, which allows us to do different things with language, such as making statements, asking questions and so on. The linguistic resources for expressing attitudes have recently been reformulated in the theory of ‘appraisal’ systems (Martin, 2000). In images the interpersonal metafunction is fulfilled by the systems of the gaze, size of frame, and angle. The ‘textual’ metafunction, finally, allows us to use language to marshal individual representations-cuminteractions into coherent texts and communicative events, linguistically through the systems of cohesion, thematic structure, and given-new, and in images through the systems of composition, framing and salience.


Figure 6. Illustrative uses of typography

Figure 8. The interpersonal function of typography: expressing attitudes and feelings

Can typography fulfil all three of these functions? I think yes. Typography can, and is, used ideationally, to represent actions and qualities. The examples in Figure 6, for instance, show a ‘scratchy’ font used to illustrate the idea of headache, and bones to illustrate the idea of death. Again, in Figure 3 a soft, smooth, rounded ‘script’ font is used to express the idea of ‘indulgence’. Designers are increasingly interested in such illustrative uses of typography, and in blurring the boundaries between letter forms and images, something which, in the ‘old typography’ was often frowned upon (e.g. McLean, 2000, p. 56)

Typography can also enact interactions and express attitudes to what is being represented. A word can be changed into a ‘warning’ or a ‘question’ through typography and typographic signs alone, as demonstrated in Figure 7, and typography can also be used to express attitudes towards what is being represented. It can ‘interpret’, or, you might say, ‘perform’ texts, or parts of texts, as ‘modern’, or ‘traditional’, ‘capricious’ or ‘serious’, ‘exciting’ or ‘dull’ and so on. Figure 8 shows New York designer’s Kathryn Marshal’s attempt to transform email into a visually expressive communication vehicle. It should be remembered here that not all typographical signs are letter or number forms (cf Stötzner, 2003). Many new non-letter signs are now emerging, and some of them can realise ‘interpersonal’ meanings, for instance the ‘emoticons’ used in email messages. The way typography can realise textual meaning has already been touched on in the discussion of Figure 2. Typography can demarcate the elements, the ‘units’, of a text and express their degree of similarity or difference as textual elements, and it can foreground key elements of a text and background less important elements. Many typographical signs that are not letter forms realise textual meaning, the most obvious example being punctuation marks – and they, too, are now rapidly developing new uses and new signs.

Figure 7. The typographical realisation of ‘speech acts’

43

44

Theo van Leeuwen

Typography and multimodality It is important, however, not to isolate typography from the other communicative modes with which it almost always co-occurs. Just as, in the practice of contemporary designers, the boundaries between the formerly distinct specialisms of design (illustration, typography, photography, etc) are now eroding, so in the new writing the corresponding semiotic means of expression no longer occupy distinct territories, but are interconnected in many different ways. Typography itself, too, is no longer just about letter forms. It is multimodal, integrated with other semiotic means of expression such as colour, texture, three-dimensionality, and movement. In Figure 9, for instance, the words ‘fall in love’ are in red, which both lends them salience (a ‘textual’ meaning), and expresses the idea of love, though by means of colour, rather than by means of typography.

Figure 9. Typography and colour

Figure 0. Typography and texture

Figure 0 shows the logo of the Swiss avant-garde art magazine Parkett, which uses three-dimensionality and texture, and was in fact hand-embroidered by the designer’s mother. In this way it celebrates the values of traditional hand-crafted objects, and opposes itself to the slick, computer-generated logos which are so ubiquitous today. Finally, in film and television titles and commercials, and on Internet websites, typography makes increasing use of movement. A series of Channel 5 programme announcements in the UK, for instance, used kinetic typography both illustratively (e.g. writing the verb ‘cycle’ in a circle and making it rotate, or stretching out the word ‘long’ in the phrase ‘a long wait’) and interpersonally, by creating visual equivalents of intonation and speech rhythm. This means that the key concepts we need to analyse and evaluate document design should not apply just to language, or to any other specific, single semiotic mode. They should be functional concepts, concepts that label a particular communicative function, and can be applied to all semiotic modes that have developed resources for realising it. Salience, making a given text element stand out from its immediate textual environment, is such a concept. It can be realised through a wide range of semiotic modes, and, within each mode, by a number of different means. Typography for instance, can realise


salience through size, colour contrasts, movement, or indeed anything that can make a word or phrase or clause stand out from others (different font, different set, different weight, etc). Framing, the demarcation of the elements of text, be they verbal or visual, is another. And the means by which these communicative functions are realised can also cross over between modes, and be applied in many domains of semiotic endeavour. Creating salience through colour, for instance, is not restricted to typography, but is possible also in images, fashion, product design, interior decoration, architecture, etc. And colour, in turn is not restricted to expressing salience, but can also express ideational and interpersonal meanings (Kress & Van Leeuwen, 2002) In other words, if we are to do justice to the common semiotic functions of different semiotic resources, and if we are to be able to bring out the specific strengths of specific semiotic resources and to explain what difference it makes whether a given communicative function is realised through one semiotic mode or combination of semiotic modes, or another, we need to extend the scope of linguistics, and to incorporate it in a broader theory of multimodality. But this cannot be done without first separately exploring the communicative potential of the different semiotic resources involved, and it is this I am trying to do here with respect to typography.

Typography as medium and as mode Semiotic resources can be organized as a medium or as a mode (c.f. Kress & Van Leeuwen, 200). If a semiotic resource is organised as a ‘mode’, it has both a grammar and a ‘lexis’. If it is organised as a ‘medium’, it has only a ‘lexis’. Perhaps this is best explained by means of an example. As Kress and I have described elsewhere (2002), in Medieval art the semiotic resource of colour was organised as a medium. Pigments had value in themselves. Ultramarine, for instance, had to

be imported from across the sea (as the name indicates) and was expensive, not only for this reason, but also because it was made from lapis lazuli. Therefore it was used for high value subjects, such as the mantle of the Virgin Mary. Such pigments were not mixed, but used in unmixed form, or at most only mixed with white. Each pigment was a very concrete, material resource, with its own, unique identity and character. Around 600, in the Netherlands, a new type of oil paint was introduced. It was not only cheaper, it also made mixing possible. As a result colours lost their individual identities. Colour was no longer conceived of as ‘lexis’, as a large collection of distinctly different, individual pigments, but as a combinatory system with five elementary, ‘abstract’ colours (‘red’ in general, rather than a specific red, and so on) from which all other colours could be mixed, just as language is conceived of as a system with a limited number of speech sounds from which all words can be constructed, and a finite number of words from which all sentences can be constructed. Typography has mostly been seen as a ‘medium’, a collection of distinct, individual typefaces, with distinct provenances, to be listed alphabetically, as in the word processor, or at best grouped together on the basis of historical principles and ‘influences’, rather than systematically, as in this example, from the Thames and Hudson Manual of Typography (McLean, 2000, p.60). Didone types, invented by Didot and perfected by Bodoni, are classified in England by the meaningless term ‘modern’. They are characterized by vertical shading and hairline serifs, introduced in the middle of the eighteenth century when improvements in presses and paper-making made such fine lines possible to print.

The 920s Bauhaus designer Jan Tschichold attempted to change this. He analysed letter forms into their basic building blocks, in order to create what he called a Skelettschrift (‘skeleton lettering’), a rational and functional typeface, suitable for the modern, industrial age. Just

45

46

Theo van Leeuwen

as phonologists describe language as having a limited number of discrete phonemes and regard the many variations of pronunciation that result from the co-articulation of different phonemes as variations that do not affect meaning, so, here too, the ‘meaningless’ variation that resulted from typography’s roots in handwriting was eliminated, and as many interchangeable components as possible were created (e.g. the ‘bowls’ of ‘a’, ‘b’, ‘p’, ‘d’, ‘g’ and ‘q’ were all made identical, which usually they are not). But this move towards typography as ‘system’ was, and often still is, rejected by traditional typographers. McLean, for instance, compares it unfavourably with the work of the famous British typographer Eric Gill (McLean, 2000, p. 67). It ‘reduces difference’, he says, and eliminates ‘subtlety’, ‘refinement’ and the link with tradition.

Figure . ‘Circuit’ typeface (Peter Grundy, 982)

[Tschichold’s] seductive theory had to be paid for in loss of legibility, since the effect was to reduce the differences... Eric Gill’s sans was different in that it was drawn by an artist and designer who was already deeply involved with the classical roman alphabet. His letters contained subtleties and refinements which the German designers, preferring the logic (or dictatorship) of rules and compasses, could not admit.

When a semiotic resource is organized as a ‘medium’, meaning comes about in a relatively adhoc, unsystematic way, through one of two principles, connotation or experiential metaphor (c.f. Kress & Van Leeuwen, 2002). The term ‘connotation’ is used here in a specific sense. It refers to the idea that signs may be ‘imported’ from one context (one era, one social group, one culture) into another, in order to signify the ideas and values associated with that other context by those who do the ‘importing’. Many aspects of the ‘Circuit’ typeface in Figure  are ‘imported’ into typography from the domain of the electric circuitry diagram. It can therefore be used, for instance, to connote ‘technicality’. The ‘Herculanum’ typeface in Figure 2 imports aspects of

Figure 2. ‘Herculanum’ typeface (Adrian Frutiger, 988)

the form language of informal Ancient Roman inscriptions an papyri into a contemporary typeface and can therefore be used to connote the values we associate with Antiquity and the Roman Empire. The other principle that can be used to endow meaning to the items in the typographic lexicon, is that of ‘experiential metaphor’. The idea, inspired by the groundbreaking work of Lakoff and Johnson (980), is that a material signifier has a meaning potential that derives from our physical experience of it, from what it is we do when we articulate it, and from our ability to extend our practical, physical experience metaphorically,


to turn action into knowledge (see Van Leeuwen, 2005, Chapter 2, for a more extensive discussion). This is again best explained by means of an example. Speakers and singers often adopt a soft, breathy voice quality to signify ‘sensuality’. As experienced producers and interpreters of speech we know both that soft, whispered speech can be associated with intimacy or conspiracy, and that breathiness is associated with being ‘out of breath’, for instance as a result of excitement or exertion – thus ‘intimate excitement’ and ‘intimate exertion’ can become part of the meaning potential afforded by this signifier. A key aspect of the letter forms in figure 3 is their irregularity. They differ in size and thickness, and indeed in shape – different a’s, for instance, are drawn differently. The distribution of ‘weight’ (thickness and thinness), too, goes against the norms of typography, in which it is usually the upright stem of the ‘n’, rather than the descending line in the middle, which is thick. In our own physical experience of writing, such irregularities stem from an inability or unwillingness to apply the rules of ‘neat writing’ we are taught in school. As a result, irregularity has, amongst other things, the potential to signify a kind of rebellion against the norms of the school, or, by extension, other coercive institutions. In Figure 3 this rebellion is of course neatly contained by the controlled symmetry of the overall layout.

Figure 3. Cover of a Monie Love single (lettering by Ruth Rowland, 989)

Distinctive feature analysis In phonology, a breakthrough was made when Jakobson and Halle (956) described phonemes, not as the minimal, not further analysable units of speech, but as ‘bundles of features’, different combinations of ‘distinctive features’ such as ‘voicedness’, ‘frontality’, ‘openness’ and so on. Although Jakobson and Halle did not see these features as having a semiotic potential, it is possible to argue that they do by using the principles of connotation and experiential metaphor, and in earlier work I have attempted to do just this for the semiotic modes of sound (Van Leeuwen, 999, Chapter 6) and colour (Kress & Van Leeuwen, 2002). Once this is established, it follows that not just formal features, but also meaning potentials can be ‘bundled’, ‘mixed’, as in the example of the ‘sensual’ voice, which ‘blends’ the meanings of ‘softness’ and ‘breathiness’ to get ‘sensuality’. Here is a first attempt at identifying the distinctive features of typography, and outlining their semiotic potential. I am restricting myself to the actual letter forms rather than also including other features such as letter spacing, interlineal space, etc, which of course also belong to the semiotic resources of typography. I would like to stress that the list below is not a kind of ‘dictionary’, listing the ‘authoritative’ meanings of letter forms. What I am doing here is presenting proposals for explicitly ‘semioticizing’ typography, for making something meaningful that was previously was not regarded as semiotic. But I am doing so on the basis of what I argue to be shared experience, and hence on the basis of principles which promise at least the possibility of successful communication. The principle of connotation of course also makes shared meaning possible, but on a different basis – on the basis of shared cultural knowledge and values.

47

48

Theo van Leeuwen

Weight

The metaphoric potential of this feature, which is, again, a continuum, relates to our experience of space. Maximally condensed typefaces make maximal use of limited space. They are precise, economical, packing the page with content. Wide typefaces, by contrast, spread themselves around, using space as if it is in unlimited supply. But the values of the contrast may be reversed. Wide typefaces may also be seen in a positive light, as providing room to breathe, room to move, while condensed typefaces may, by contrast, be seen as cramped, overcrowded, restrictive of movement.

This is the difference between bold typefaces or versions of a typeface, and regular typefaces or versions of a typeface, as shown here by the difference between Arial black and Arial. As with many of the features to be discussed below, this is not a binary but a gradual contrast – there is, at least in principle, a continuum of boldness, even if technologies like the word processor reduce it to a binary choice. Increased weight is of course frequently used to increase salience, but it can, at the same time, be used metaphorically, to signify ideational and interpersonal meanings. Bold can be made to mean ‘daring’, ‘assertive’, or ‘solid’ and ‘substantial’, for instance, and its opposite can be made to mean ‘timid’, or ‘insubstantial’. But the values may also be reversed. Boldness may have a more negative meaning. It may be made to mean ‘domineering’, ‘overbearing’. Other, co-present signifying elements will narrow down the meaning potential and the values invoked, and make them more specific. I have glossed the typographical meaning potential by means of adjectives, and that may suggest that their meaning is primarily ideational. But while adjectives like ‘daring’, ‘assertive’, ‘solid’, ‘substantial’ and so on signify qualities of what is being represented, they can also have interpersonal significance. They can also signify attitudes towards what is being represented, or do something to readers. Boldness, for instance, can typographically ‘hector’ readers and the smooth, rounded letter forms in Figure 3 can simultaneously signify the idea of ‘indulgence’ and symbolically ‘pamper’ and ‘soothe’ the reader.

This refers to the difference between cursive, sloping, ‘script’-like typefaces and upright typefaces, as shown here by the difference between Lucida Bright and Lucida Calligraphy. Again, there are degrees of ‘slope’, and slope can also be either right-leaning or left-leaning, although the latter is less common in typefaces. Even when we cannot ‘place’ a cursive typeface in a particular era, the contrast can be recognized as that between handwriting and printing. The meaning potential of this contrast is therefore predominantly connotative, based on the meanings and values we associate with handwriting and printing. Depending on the context, it might signify a contrast between the ‘organic’ and the ‘mechanical’, the ‘personal’ and the ‘impersonal’, the ‘formal’ and the ‘informal’, the ‘mass-produced’ and the ‘handcrafted’, the ‘new’ and the ‘old’, and so on.

Expansion

Curvature

Typefaces may be condensed, narrow, or they may be expanded, wide, as shown here by the difference between Arial and Arial narrow.

A letterform can stress angularity or it can stress curvature, as shown here by the difference between Copperplate and Century Gothic. ‘Black letters’,

Slope


as shown here by the ‘Old English Text MT’ font, have pronounced angularity. Curvature may also be realized by the difference between, on the one hand, rounded ascenders and descenders, e.g. in fonts which use loops and fonts which apply flicks (curved hooks at the end of ascenders and/or descenders), as shown here by Script MT Bold and Pristina, or, on the other hand, predominantly straight ascenders and descenders, as in typefaces like Agency FB. Many typefaces mix and match the two. Although particular types, such as Old English Text MT, may have clear cultural connotations, this feature also has experiential meaning potential, based both on our experience of producing straight, angular forms, which requires controlled, brisk, decisive movement, and round forms, which require a more gradual, ‘fluid’ control of movement, and its significance may also be based on experiential and cultural associations with essentially round or essentially angular objects. Roundedness can come to signify ‘smooth’, ‘soft’, ‘natural’, ‘organic’, ‘maternal’, and so on, and angularity ‘abrasive’, ‘harsh’, ‘technical’, ‘masculine’, and so on. Both may either be positively or negatively valued. Modernity, rationality, functionality etc have often favoured the values of angularity, as e.g. in the paintings of Mondrian, while postmodernity has brought back round forms, for instance in car design and architecture. Clearly the field of possibilities is very wide. But it will be narrowed down by other, co-present features, and by the context generally – a particularly important feature of the context is the genre in which a font occurs, and the expectations this sets up in the reader.

Connectivity Letter forms can be connected to each other, as in running script, have hooked feet that extend to various degrees to the next letter, or almost touch it, or lack

any of these features so that the letter forms are quite separate and self-contained, as shown here by Lucia Handwriting, Lucida Calligraphy, and Lucida Console. Connection and disconnection can be external, between letter forms, as in the examples above, or internal, within letter forms as in the Bauhaus 93 typeface. Connectivity is, again, associated with handwriting, and therefore shares much of its meaning potential with ‘slope’ (see above). But it also has its own metaphoric potential. External disconnection can suggest ‘atomisation’, or ‘fragmentation’, and external connection ‘wholeness’, or ‘integration’. But the values may also be reversed, with disconnection signifying the distinctive individuality of the elements of the whole, and connection its opposite. Internally disconnected letter forms, finally, have a sense of not being ‘buttoned up’, which may be negatively valued, as ‘unfinished’, or ‘sloppy’, or positively, as, say, ‘easy-going’.

Orientation Typefaces may be either be oriented towards the horizontal dimension, by being comparatively ‘flattened’, as shown here by Bodoni MT Black, or oriented towards the vertical dimension by being comparatively elongated, stretched in the vertical direction, as shown here by Onyx. The meaning potential of horizontality and verticality is ultimately based on our experience of gravity, and of walking upright. Horizontal orientation, for instance, could suggest ‘heaviness’, ‘solidity’, but also ‘inertia’, ‘self-satisfaction’, while vertical orientation could suggest ‘lightness’, ‘upwards aspiration’, but also ‘instability’. Other related aspects of orientation are (a) the difference between typefaces with short ascenders and descenders that hardly extend beyond the ‘x-line’ and the ‘base line’, as for instance in Bernard MT Condensed, and typefaces with long descenders and ascenders, as

49

50

Theo van Leeuwen

for instance in the aptly named High Tower Text, and (b) the difference between a downwards orientation in which the descenders are longer than the ascenders, as exemplified here by Viner Hand ITC, and an upwards orientation, in which the ascenders are longer than the descenders, as exemplified here by Poor Richard. In the former case the letter forms seek roots, as it were. In the latter case they aspire to some form of metaphorical ‘elevation’. When the extent of the ascenders and descenders is minimalized, the letter forms ‘stay within their allotted space’, and neither aspire to take root nor to some form of metaphorical ascension.

Regularity The contrast between regular and irregular typefaces has already been commented on in relation to Figure 3. Many typefaces have deliberate irregularities, through an apparently random distribution of specific features, for instance curvature (e.g. some descenders with, others without a playful flick or ligature), and through entasis, which can also be interpreted in terms of regularity – in some cases the different parts of a letter form differ in weight, and in others they do not. Entasis may either be fairly regular and systematic, in ‘traditional’ oblique shading and ‘modern’ vertical shading, or differ from these two standard forms, as in this typeface called Chiller. Irregularity may also be created by not staying within the lines, going above the x-line or below the baseline, for instance, as here in Kristen ITC, or by variation in slope, as in Ravie (compare the ‘t’ and the ‘l’, for instance). Traditional typography has set great store on regularized forms of differentiation, for the sake of the distinctiveness of letter forms, and hence of legibility. But regularity and irregularity also have their metaphoric potential, as seen in the brief analysis of the letter forms in Figure 3.

Non-distinctive features Some features of letter forms are, strictly speaking, not necessary for telling them apart, although they may be said to contribute to legibility, as in the case of serifs. Typography has developed a wide range of flourishes, ligatures and capricious additions, and they, too, can be said to have a meaning potential, in many cases derivable from that of the distinctive features described above. The flourishes of Edwardian Script IT, for instance, are both ‘rounded’ and ‘expansive’, while the curls of Curlz MT are irregular, including pearl-shaped loop terminals, circular dots on the i’s and, capriciously, within the bowls of the ‘o’, the ‘p’, the ‘g’, and the ‘q’. I am not able to do justice to this complex area within the space of this paper, and hope to be able to explore it more fully in further work.

Typography as a semiotic mode

pp. 151 & 152 not included

Table  summarizes the discussion above in the form of a system network (e.g. Halliday, 978). The curly brackets signify ‘parallel systems’, that is, ‘both...and’ rules (for instance, ‘a letter form must have both a certain weight and a certain degree of expansion and...’). The square brackets signify binary systems, ‘either...or’ choices (for instance ‘disconnection must be either internal or external’). The double-headed arrows signify graded contrasts, continuums. This brings out that, overall, this aspect of typography operates as a parallel, rather than a linear system. At the same time, at least some of the parallel systems can be modelled as binary systems, usually because they have, in the practice of typography, been standardized to the degree that they have become a set of discrete alternatives, whether for technological or other reasons. The system of ‘serifs’ (Table 2) is a good example (I use typographical terminology here, rather than functional labels)


The same applies to the textual meaning potential of typography, which I have not explored in detail in this paper, and which is closely related to the meaning potential of layout (c.f. the theory of layout in Kress & Van Leeuwen, 996).

Typography and document design I would like to end by using the framework developed in this paper in a slightly more extended analysis of two examples from the everyday practice of document design. The example in Figure 4 comes from the brochure of a real estate agent. It uses two different fonts, one for the first sentence, a statement of self presentation in which the company headlines the nature of its operations, the other for a list of the areas in which it operates. The textual meanings expressed here are obvious. The statement of self presentation is bolded and coloured, and hence more salient. It is a headline. But other features are also relevant, and they can not all be explained on the basis of salience – salience could also have been achieved, for instance, by bold uprights, rather than italics. This self-presentation statement also uses typography ideationally (the company is constructing a representation of itself) and interpersonally (the

Figure 4. Real estate information brochure

company is also addressing its potential clients in a certain way). In terms of the distinctive features I have discussed, the typeface is not only coloured (a royal purple) and bold, but also fairly wide, sloping, fairly rounded, not entirely regular, and with just a hint of flourish. The other features seem to be relatively neutral and do not appear to play much of a role in the contrast between the two typefaces. The typeface of the list, by contrast, is not only black, but it also has less weight, is more condensed and upright, less rounded (compare the a’s and the e’s for instance), more regular and without even a hint of flourish (compare the f ’s, for instance). The differences are slight, but slight differences matter in typography. Applying my discussion of the meaning potential of these features, the company here presents itself in ‘personal’ way (the sloping font, reminiscent of handwriting), as ‘human’ rather than ‘mechanical’ (rounded, and slightly irregular), but also quite assertively (bold and wide). The typeface of the list, by contrast, is less personal, more formal (the upright font), and ‘mechanical’ (more angular and regular), and lacks the ‘assertive’ features of the presentation. It is, in short, factual and informative only, oriented towards legibility rather than expression. This contrast is also realised in the semiotic modes of language and colour. The language of the self-presentation has a personal element (the use of a first person pronoun), while the language of the box is a neutral, factual list, eliminating all the interpersonal resources of language. And the colour of the self-presentation sets up connotations of value (the royal purple), while the black and white of the list is neutral in this regard. In short, typography plays a role both in expressing what kind of company this is, and in expressing what communicative work it is doing in this text fragment. The second example is from an information brochure of an insurance company, National Mutual Life.

53

54

Theo van Leeuwen

phy. In other words, text and typography do not always double each other. Some meanings may be realized in both modes, others only in the one or the other.

Conclusion

Figure 5. Heading of National Mutual Life information brochure for independent financial advisers

The text in this example uses a ‘dialogic’ question and answer format, mixing information (‘n the past it has not always been easy to compare the effect of charges of the different pension providers’) and self promotion (‘One company really stood out ... Who is this company? Answer. National Mutual Life’), and adopting a relatively informal, personal tone. The font used for this information is a serif, with differentiated letters, wide, fairly rounded, with more than usually long ascenders and descenders (the other features would appear to be relatively neutral). In terms of the discussion above, this could be interpreted as an attempt to appear both personable, ‘showing a human face’ and assertive – within the limits of the factual genre and the legibility requirements of running text. The treatment of the phrase ‘for the Independent Financial Adviser’ is of particular interest. The font has clear connotations of traditional calligraphic script. It is wide, differentiated (two different d’s, for example), sloping, rounded, connective, quite assertive in terms of the extent of the ascenders and descenders, and it has pronounced flourishes. Thus the independent financial adviser is flattered as being personal, flexible, enterprising, but also rooted in solid traditional values, someone with a ‘pedigree’. Nowhere are these meanings expressed linguistically. They are expressed solely by the typogra-

My conclusion will be short, because I feel that this work has only just begun, and not yet reached a stage in which conclusions can be drawn. This paper should therefore be relatively open-ended, inviting others to join in the enterprise, rather than presenting a finished product. So let me just briefly recapitulate. I have suggested, persuasively I hope, that typography can be seen as a semiotic mode - systematic, multimodal and able to realize not just textual, but also ideational and interpersonal meaning. I have argued that developing a detailed ‘grammar’ of this semiotic mode, and detailed approach to analysis, is important, and that this should be done in a way that can be integrated with the theory and methods of other semiotic modes. In the age of the ‘new writing’ it has become imperative to analyze and evaluate documents multimodally, rather than on the basis of the linguistic text alone - however important language is, and will always remain. To integrate the study of typography into such a multimodal analysis, it is necessary to go beyond the formal approach that has characterized it so far, to put it on the basis of a theory of communicative functions, and to develop that ‘complex grammar’ Neuenschwander called for. I fully realise, of course, that what I have done so far covers only part of the territory and captures only the broadest outline of the ‘grammar’ of this complex and fascinating field. Note I would like to express my thanks to the two anonymous readers who reviewed the manuscript and spotted a number of inaccuracies and inconsistencies.


References Bellantoni, J. & Woolman, M. (2000). Type in motion – innovations in digital graphics. London: Thames and Hudson. Cook, G. (200). The discourse of advertising, (2nd ed.). London: Routledge. Crystal, D. (998). Towards a typographical linguistics, type 2(), 7–23. Fries, P. H. (98). On the Status of Theme in English: arguments from discourse, Forum Linguisticum 6(), –-38. Gleason, H. A. Jr. (973). Contrastive Analysis in Discourse Structure. In A. Makkai & D. Lockwood, Readings in Stratificational Linguistics (pp. 258–276). Alabama: University Press. Goodman, S. & Graddol, D. (996). Redesigning English: New texts, new identities. London: Routledge. Gutwinski, W. (976). Cohesion in Literary Texts: a study of some grammatical and lexical features of English discourse. The Hague: Mouton. Halliday, M. A. K. (978). Language as a social semiotic. London: Arnold. Halliday, M. A. K. (985). An Introduction to Functional Grammar. London: Arnold. Halliday, M. A. K. & Hasan, R. (976). Cohesion in English. London: Longman. Jakobson, R. & Halle, M. (956). Fundamentals of Language. The Hague: Mouton. Kress, G. & Van Leeuwen, T. (996). Reading Images: The Grammar of Visual Design. London: Routledge. Kress, G. & Van Leeuwen, T. (200). Multimodal Discourse – The Modes and Media of Contemporary Communication. London: Arnold. Kress, G. & Van Leeuwen, T. (2002). Colour as a semiotic mode: notes for a grammar of colour. Visual Communication (3): 343–369. Lakoff, G. & Johnson, M. (980). Metaphors We Live By. Chicago: University of Chicago Press. Martin, J. R. (983). Conjunction, the logic of English text. In J. S. Petöfi & E. Sözer (Eds.), Micro and Macro Connexity of Texts (pp.–72). Hamburg: Buske Verlag. Martin, J. R. (2000). Beyond Exchange: APPRAISAL systems in English. In S. Hunston & G. Thompson (Eds.), Evaluation in Text – Authorial Stance and the Construction of Discourse (pp. 42–76). Oxford: Oxford University Press.

Martin, J. R. (992). English Text: System and Structure. Amsterdam: Benjamins. McLean, R. (2000). The Thames and Hudson Manual of Typography. London: Thames and Hudson. Myers, G. (994). Words in Ads. London: Edward Arnold. Neuenschwander, B. (993). Letterwork – Creative Letterforms in Graphic Design. London: Phaidon. Stötzner, A. (2003). Signography as a subject in its own right. Visual Communication 2(3), 285–303. Swales, J. (990). Genre Analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press. Van Leeuwen, T. (99). Conjunctive structure in documentary film and television. Continuum 5(), 76–5. Van Leeuwen, T. (993). Genre and Field in Critical Discourse Analysis. Discourse and Society 4(2), 93–225. Van Leeuwen, T. (999). Speech, Music, Sound. London: Macmillan. Van Leeuwen, T. (2005). Introducing Social Semiotics. London: Routledge. Walker, S. (2000). Typography and Language in Everyday Life. London: Longman.

about the author Theo van Leeuwen is Dean of the Faculty of Humanities and Social Sciences, University of Technology, Sydney. Previously he was Director of the Centre for Language and Communication Research at Cardiff University. He has published widely in the areas of social semiotics, critical discourse analysis and multimodality. His latest book is Introducing Social Semiotics (Routledge, 2005). A second edition of his Reading Images – The Grammar of Visual Design (co-authored with Gunther Kress) will appear in 2006.

Contact University of Technology, Sydney PO Box 23 Broadway NSW 2007 Australia theo.vanleeuwen@uts.edu.au

55

-Van Leeuwen(2006)_Towards a Semiotics of Typography

Recommend Documents