17.2: What is writing? and Figures

The term ‘writing’ embodies two meanings — a process, involving the interaction of human bodies with materials normally mediated through various tools (pen, keyboard, etc.), and substance (cf. Piquette and Whitehouse, this volume), the material residues of those bodily actions on, or in, the surface of media of many kinds, permanent to varying degrees, even virtual, but visually legible, often tangible, at any time after writing has happened subject to preservation of the medium itself (cf. Cessford, this volume, on the range of media and their preservation in 18th- to 20th-century Cambridge). For those of us who study writing in the past, the former is rarely visible (although writing or writers are sometimes represented visually and we occasionally recover writing tools; cf. e.g. Coe and Kerr 1997) and has to be deduced from the latter. In this respect, writing is like many other material products of the past: we have to reconstruct, or ‘reverse- engineer’, the process of production, the chaîne opératoire, from the artefacts themselves.

Writing, of course, implies a complementary, but distinct and sequential process of engagement with the product, a process that is visual and/or tactile and embodied and requires the presence of the material residues of writing. For the purposes of this epilogue, I loosely use the term ‘reading’ for this process, but suggest that ‘reading’ can potentially encapsulate a broader set of engagements than merely making sense of and absorbing a representation of language. Although this is perhaps clearest in contributions to this volume that explicitly deal with non- language-based ‘writing’ (e.g. Salomon, on khipu; or Perego on ‘iconic literacy’), it underpins the material-based approach that pervades the whole collection. The diagram in Figure 1 seeks to encapsulate concisely the processes associated with writing and reading. Because it is a material practice, writing requires a physical surface, on or into which it is applied, and requires bodily movements particular to different techniques for its production, the techniques dependent on the nature of the interaction between tool and medium. The process of writing might involve a single action — in the case, for example, of stamping or application of a transfer print (‘indirect’ forms of writing) — or multiple actions: flowing strokes (cf. Ingold 2007), where a brush or pen is used; staccato incisions or impressions, where a stylus is used on clay; repeated staccato actions, where a chisel is used on stone (‘direct’ forms of writing). The ‘choreography’ of writing — how the implement (brush, stylus) was held; how the surface was rotated, or not, to facilitate marking — was brought vividly home to those attending the original UCL Institute of Archaeology conference in demonstrations of Egyptian scripts written with a brush by professional scribe and calligrapher, Paul Antonio, and of cuneiform impressed into clay by British Museum Assistant Keeper of Cuneiform Collections, Jonathan Taylor. It also lies at the heart of Kidd’s (this volume) exploration of the bilingual, bi-scriptural world of Ptolemaios, a late 3^rd-century BCE ‘Egyptian’.

Although it occupies a period of time, we can think of the ‘moment of writing’ (cf. Piquette, this volume) in contrast to the potentially multiple ‘moments of reading’ that might take place many times, minutes, days, even centuries or millennia afterwards, including the ‘special case’ of reading involving decipherment because knowledge of the original system has been lost. Depending on the portability of the material written upon, ‘reading’ may take place in the same location or at a distance; it may be achieved by the same actor, or a different one, or even be only an ‘implied’ reading aimed at supernatural beings. ‘Reading’, too, requires bodily movement: at a minimum, relatively small head, eye (e.g. Dehaene 2009: 12–18) and hand movements, but potentially quite extensive movements to appreciate writing laid out in (for us) unusual formats — an ‘athletics of reading’ (cf. Johnston; Whitehouse, both this volume).

But this material process is not unique to writing. Because writing as a product is a visual medium (cf. Robertson 2004), the scheme sketched above could be applied to other forms of representation applied to surfaces and loosely termed ‘art’ (cf. Perego, this volume); indeed, in many ways, writing is a particular form of representation. Like all forms of representation, its appreciation is not limited to the visual dimension: it may involve, even require (as in Braille) touch, or movement, and may, directly or by association (e.g. the text applied to consumption vessels in 18^th-century Cambridge [Cessford, this volume], or classical Greece [Johnston, this volume]), instil bodily experiences of taste or smell. Denise Schmandt-Besserat (2007) has also suggested that writing and art in early Mesopotamia ‘co-evolved’, with the formatting and layout of writing affecting that of other visual media. It is no accident that theorists of writing from antiquity to the 20^th century saw its origin in pictures (e.g. Evans 1908; Gelb 1963; Tylor 1865: 83–106; Warburton 1765, Book IV). The same idea drove the interpretations of Egyptian hieroglyphs by Horapollo (e.g. Cory 1840) and arguably delayed the decipherment of Maya writing by over a century after Abbé Brasseur de Bourbourg’s publication of Bishop Diego de Landa’s account of the Maya ‘alphabet’ (Brasseur de Bourbourg 1864; cf. Coe 1992: 99–106).

The above considerations are appropriate to a volume on writing as ‘material practice’, but definitions of writing as a technology often emphasise the content of writing systems: how they work as systems, rather than their material manifestations. Powell (2009: 13), for example, defines writing — simply and concisely — as “a system of markings with a conventional reference that communicates information”. We might add at the end of this definition “through time and (potentially) space”; writing removes the need for a reader to ‘be there’. For Powell there is no necessary link between writing and the representation of speech, a point also made by Boone (2004: 313), who defines it as “conventional, permanent, visual marks to communicate relatively specific ideas”, also illustrating other “semasiographic representation systems”, such as music notation, dance notation, algebra, and chemical formulae (Boone 2004: 317–335; cf. also Boone and Mignolo 1994).

Others insist on a systematic link to spoken language for a representational system to qualify as ‘writing’. Robertson, for example, maintains that “writing is truly writing when it systematically represents speech” (Robertson 2004: 20). This is slightly ironic, since Robertson invokes Peirceian semiotics to explain how writing can work as a representational system and Peirce was attempting to develop a much broader, inclusive understanding of how signs work (Robertson 2004: 18–19). As Powell (2009: 18) points out, such definitions echo de Saussure (1983 [1992]: 24): “[a] language and its written form constitute two separate systems of signs. The sole reason for the existence of the latter is to represent the former”. Definitions that insist on writing’s systematic relationship with speech break down at the khipu (Salomon, this volume), Mixtec pictorial codices (e.g. Boone 2004) or even early Sumerian ‘numerical’ and ‘numerico-ideographic’ tablets (e.g. Cooper 2004: 75–76, figs 4.4–4.5). Systems that are deeply implicated with images, like the Egyptian (cf. Baines 1989), Mayan (cf. Jackson, this volume), or Cretan Hieroglyphic (cf. Flouda; Whittaker, both this volume), also challenge definitions that limit writing to the representation of spoken language. Equally, not all writing is for reading by human eyes (e.g. certain inscriptions in Egyptian tombs or Greco-Roman curse tablets), nor strictly representative of human language (e.g. magic spells of the ‘abracadabra’ type).

Although we might not wish to limit a definition of writing to that of notating speech, most of the papers in this collection treat writing systems with precisely this limitation. At the other extreme, we might wish to constrain a broader definition, so as to avoid the possibility that any visual marks can constitute a writing system. There is a risk in doing so, however, because it presumably sets writing systems, however broadly defined, apart from other representational or mnemonic practices (cf. Gosden 2008, for example), equally material in basis. We may wish to keep in mind the possibility that writing, in some cultures, was one of a number of elite material practices that demanded a broader cultural, rather than a narrow linguistic literacy (cf. Perego, this volume).

In the case of writing, the term ‘conventional’ is critical. Even khipu had conventions, as does musical notation, for example, and texting, as used on mobile phones (cf. Crystal 2008). Among other factors, it is convention that constrains the spread of a particular script beyond its use community (cf. Kidd, this volume, for an example of the strong link between script and language in Ptolemaic Egypt); for it to do so, the ‘convention’ must change (as it did, for example, when Greek-speakers adapted the Phoenician system to record their language). Extensions of script use across linguistic boundaries are much easier in the case of a script with a limited number of signs (especially an alphabet), although they are possible where prestigious and/or specialised literacy existed, as in the cultural inertia that kept Sumero-Akkadian cuneiform in use into the 1^st century bc (e.g. Black 2008; Brown 2008) to record structurally different languages (Sumerian versus Akkadian, Babylonian, Assyrian versus Hittite versus Old Persian) through both time and space (from Mesopotamia to both east and west, where it ran into the even more tenacious Egyptian tradition [e.g. Stadler 2008]). It is perhaps significant that both were replaced by alphabetic scripts: Aramaic, Greek, and Coptic.

Resistance to destabilisation through changing the ‘convention’ represents one reason why, in a multi-ethnic, multi-lingual, multi-scriptural area like the Levantine coastal region (cf. Sparks, this volume), at the interface between Egyptian and cuneiform systems, new ‘simplified’ written notations (proto-Canaanite; alphabetic Ugaritic cuneiform) emerged in response to, rather than as an adaptation of any one system; something similar might be implied by the Aegean syllabic scripts that arose on the margins of literate communities of the eastern Mediterranean (cf. Finlayson; Flouda; Tomas; Whittaker, all this volume).