On roman letters and other stories

An essay in heterographics

in Journal of World Literature

This essay questions the assumption that the roman alphabet is more purely phonetic than any other, and that other scripts and writing-systems are less efficient, whether for the production of texts or for their comprehension. Those who habitually use roman letters are asked to consider their competence to understand other writing systems. The work of Stanley Morison emphasizes the ideological significance of alphabets and of particular letter-forms. M.B. Parkes and Paul B. Saenger are cited to indicate how punctuation and spacing are aspects of the roman-letter writing system that cannot be treated as purely phonetic. Beyond the world of roman letters there is a focus on Syriac and the Xi’an stele, which was printed by Athanasius Kircher in 1667 and marks the first publication in the west of a substantial text in Chinese.


This essay questions the assumption that the roman alphabet is more purely phonetic than any other, and that other scripts and writing-systems are less efficient, whether for the production of texts or for their comprehension. Those who habitually use roman letters are asked to consider their competence to understand other writing systems. The work of Stanley Morison emphasizes the ideological significance of alphabets and of particular letter-forms. M.B. Parkes and Paul B. Saenger are cited to indicate how punctuation and spacing are aspects of the roman-letter writing system that cannot be treated as purely phonetic. Beyond the world of roman letters there is a focus on Syriac and the Xi’an stele, which was printed by Athanasius Kircher in 1667 and marks the first publication in the west of a substantial text in Chinese.

Humans, characters, move, taking with them not only their languages but also their writing systems. The crisis of refugees that confronts Europe today may be figured “literally” in terms of scripts: for the Roman alphabet is universal throughout the nations of the European Union (excepting Greece and Bulgaria), yet some of those seeking a new life in Europe are not only other-tongued; they may also be other-lettered. When two writing systems meet, each ought to recognize the limitations not only of the other but, more challengingly, of its own. Such is the heterographic (Lock “Heterographics”).

The principle of order and sequence is enshrined in the very word “alphabet”, from the sequence of the names (in Greek, from Hebrew) of its first two characters: the alphabet enforces sequence as rigorously as numbers do (Sampson 109). Thus an alien “alphabet” holds a threat rather more serious than does a foreign tongue. A different writing system poses a fundamental alternative to the very idea of alphabetical order and sequence that—often enforced by rhyme—is instilled in every “roman” child.

A graphic system that we lack competence to read can hardly be identified as an alphabet, or even as a writing system. For what are the markers that, for the uninformed, might distinguish the graphic from the decorative? If we do not even know the direction that reading should follow, a script cannot be distinguished from ornament, as in a scroll or frieze that only the informed can read, though all can admire. The difference here exposed by ignorance is one that tests all our attempts to understand and define writing.

There are many “purely phonetic” writing systems, but by far the most familiar, globally, is the roman. Western scholarship on writing-systems and scripts tends to take roman for granted, and then to describe other systems—notably Chinese and Japanese—in iconic terms, using such composite words as pictogram, ideogram, hieroglyph. Each of these words draws attention to a property of written signs that cannot be reduced to the phonetic.

The prestige of the roman alphabet is such as to render other writing-systems, to its “native readers”, eccentric and, in many accounts, inferior; what remains visible, what is not subsumed in voice, belongs to a mode of writing that, still bound by material and the iconic, has not yet attained to the purely conceptual. However, readers of roman may see rather more than they acknowledge; this, the iconic repressed, makes the Roman alphabet a writing-system like any other.

The triumph of the West can be construed as the triumph of roman lettering: Rome’s empire declined and fell, but its successor-states retain its alphabet. In this respect at least, Horace’s line rings hollow: Graecia capta ferum victorem cepit (Horace, Epistles II i 156). Captive Greece captured its wild conqueror in all sorts of ways, but not at the level of script, even though it would be many centuries before Greek script was thoroughly subjugated. “Roman” is a late appellation, dating from the Renaissance, prompted by the new concern with scripts brought about by printing, and motivated by the realization that, exceptionally, this alphabet is used by other languages than Latin.

“Roman” as a modifier in these contexts indicates what in English is often called “Romanesque” or, of languages, “romance” or “romanic”: Roman as “Romanesque” signifies a decline into a derivative mode, epitomized in 800 by Charlemagne’s “Roman Empire”—though not modified as “Holy” until 1157. It was only in 1535 that the word “Roman” was used to modify “Catholic”; such a contradictory and belittling nomination was Calvin’s. The term “Roman Catholic” thus emerged after “lettre rommaine” (sic), first attested in 1528. (OED sv. “Roman” adj. 1, etymology.)

The Eternal Letter, published in 2014, celebrates “Two Millennia of the Classical Roman Capital”. Though its editor, Paul Shaw, argues that the Roman capital was established canonically with Trajan’s Column in the year 14 CE, he has trouble explaining the breaks in continuity between the collapse of Rome and the brief Carolingian revival of the lettering, and then between c. 800 and the Renaissance. The accepted date of the revival is 795, when Charlemagne commissioned the epitaph for Pope Hadrian I in St Peter’s; the epitaph’s significance was recognized in the sixteenth century by its conspicuous preservation. Bernini’s St Peter’s bears within it—displayed as an anachronism, a “literal” relic—the inscription of Hadrian’s epitaph made seven hundred years before: the lettering chosen by the first (Holy) Roman Emperor thus appears to span the ages from Classical Rome to the neo-Classicism of a Christian Basilica.

Shaw acknowledges that “the earliest examples of ‘ancient’ Roman capitals” in Florentine humanist lettering “were actually Romanesque in origin” rather than classical (Shaw 2–3) while overlooking the lacunae in the lettering’s history. That it was largely unused for four hundred years before c. 800, and was then redundant for a further seven centuries, must not spoil the celebration of its success over “two millennia”. Yet throughout that period its prominence is evident for at most, in total, one thousand years.

For Shaw the failure of the Carolingian reforms, at least in lettering, is because “Charlemagne’s heirs lacked both his political skills and his passion for the ancient world” (Shaw 2). Such an account ignores entirely the heterographic argument put forward in 1957 by Stanley Morison: that the roman-lettered world of western Christendom accommodated itself to the more powerful Greek-scripted world of Byzantium.

For the commemorating of Christian martyrs, Damasus, Pope in Rome from 366 to 384, had ordered a new script to replace the roman lettering associated with their persecutors: the lettering of the Church ought to differ from that of the pagan Empire (Morison 93). Damasian script, adopted by the Church throughout the West, weakened the authority of Roman capitals, and led to the devising of an alphabet designed to be “effective in the service of a Greek-speaking population dominated by a Latin-speaking administration” (Morison 102).

This is a case of letters being designed to “look like” other letters, and for two reasons. First, a Latin alphabet that does not look Classical begins to look distinctively Christian, as uncials still do. Second, a Latin alphabet intended for Byzantine readers must assimilate its letter-forms to those of the Greek alphabet. Damasian script was designed to mediate between Greek and Latin. The letters might be phonetic, but their overall type or set would be visually distinct, iconic in the sense not of a letter resembling an object, but of one letter-form resembling another.

Thus features of form, design and layout can be exploited to indicate political and ideological allegiances:

The design of the letters was evidently formed after careful consideration … The script [was] devised within two generations of Constantine’s proclamation of Byzantium as his new capital … It is not surprising that the lettering devised to eulogize the martyrs who had perished only fifty years before Damasus was born [c. 305] should be in contrast with the persecutors.

Morison 93

The eventual consequence of Damasus’s innovation was Uncial or what Morison terms “Graeco-Latin script”, legible whether one’s native letters are Greek or Roman.

Although Gregory the Great (590–604) reverted to Classical or “Square Latin” inscriptions, the revival of this lettering hardly extended outside the city of Rome: the “principal scriptoria in the Western patriarchate are to be found … using Greek symbols, ligatures or conventions.” (Morison 126) Such lettering, devised for “theological comprehensiveness,” must not be thought merely a decorative or provincial variant of roman script. These are roman letters heterographically designed for the benefit of Greek readers.

This conciliatory script became over time standard for titles, headings and inscriptions—display contexts—while continuous text reverted largely to roman. Thus, perhaps, the Greek-speaking population could read the headlines while Byzantium’s Latin-speaking administrators could also read the fine print.

Much about scripts must remain speculative, given that so little evidence survives from before the ninth century. Most of what we have from the ancient world, in Latin or Greek, is copied from the ninth century onwards, and what copyists very seldom do is to respect the exact form of the letters. No more would a scholar today, in citing a passage from Shaw or Morison, set the word-processor’s font to accord with that in which the passage had been set in print. Quotations are permitted in other scripts (Greek, Chinese), though almost never (unless by way of illustration) in a variant font of the one (roman) script. There are clearly demarcated conditions in which italic is required, but otherwise the look of the page, the homogeneity of the lettered text, takes precedence over any scholarly request for a diversity of fonts.

Accuracy is reckoned a virtue in scholarship; we strive for a precise transcribing of words and punctuation. However, palaeographers apart, we are not expected accurately to reproduce either the typography or the hand. Editing might be defined as the pursuit of accuracy in the reproduction of texts, made possible only by eschewing the accurate representation of the forms of letters.

This may account for a general indifference to script in western scholarship and, outside of palaeography, an ignorance of its history. It is hard to reflect in a scholarly manner on what cannot be displayed according to the conventions of scholarly presentation. The founding figures of palaeography are Jean Mabillon (1632–1707), Bernard Montfaucon (1655–1741), who coined the discipline’s name in 1708, and Humfrey Wanley (1672–1726). Each understood, independently, that copying a “text” is not the same as copying the “hand”. Medieval scribes aimed to record the words but not their shapes. Scribes pay no attention to letter-forms and not much to orthography: what they provide is not a visual representation of a document but a phonetic account of a text whose accuracy need be judged by the ear alone. This dereliction of duty to accuracy—copying, but not tracing—might be ascribed to the presumption of the “exclusively phonetic” nature of roman letters: what really matters is what cannot be seen.

The short duration of the Carolingian revival of roman lettering can be attributed to the general abandonment of upper-case or majuscule forms when cursive emerged, during the ninth century, independently in three distinct writing-systems: Greek, Latin and Arabic (Louth 96). The use of minuscule increased the speed of writing as well as the ease of reading, presumably by encouraging silent modes of textual ingestion (Parkes “Set in Their Own” 87–93; Saenger). The Arabic term for cursive script, distinct from the more formal kufic, is naskhi, “copying”, suggesting a rate of textual production more rapid than that of composition. By “composition” may be implied either the discursive sense of making a text or the display, layout and arrangement of letters. Because scribes attended almost exclusively to the discursive sense, we have very little evidence of letter-forms before the ninth century; we remain in the early days of scholarship, in literal and scribal terms.

Greek and Latin were not the only languages of Christianity, nor Greek and roman the only scripts. Other letters were used in the non-Chalcedonian or “Miaphysite” churches: Coptic, Syriac, Ethiopian, Georgian and Armenian, and the Malankara church of India. In contrast to the use of roman across the west—and of Greek across the Byzantine Empire—each of these churches preserves its own script. It may be the determination to retain a script that contributed to the survival of each as an autonomous church. In rejecting the Council of Chalcedon (451), they resisted not only Byzantine authority but also Greek lettering; preserving their scripts, each used for one language only, the Miaphysite churches have remained outside the “Graeco-Latin” sphere.

Yet very little attention has been paid to these scripts in their graphic and heterographic contexts. J.F. Coakley, a historian of Syriac typography, asks about this neglect:

considering that most Syriac scholars read printed texts, not manuscripts, most of the time, it is surprising that there has never been a study of Syriac typography by someone in our field. I suppose it is simply not part of the western scholarly tradition to comment on the design or the source of the types in which one’s own or others’ publications are set.

Coakley [xiii]

Among the very earliest books to be printed in Syriac was the New Testament (Vienna, 1555). The title page states that this book is printed “characteribus & lingua Syra, IESU CHRISTO vernacula” (Coakley 1, n. 4). That Syriac was the vernacular spoken by Jesus may be disputed, but not that Syriac is close to Aramaic, and that its script would have been legible to Jesus. It is perplexing that so few Christians have thought it worth studying the “native” or “mother-script” of Jesus. The triumph of roman lettering has occluded even such a literal salience as this.

The imposture of the supremacy of roman across two millennia has also obscured the influence of Syriac on the development of various scripts in India and China. It was another Pope from Syria, Gregory III (731–41), who gave new purpose to the Graeco-Latin script; he might well have been tri-scribal, that is to say, fluent in three scripts and thus attentive to the politics of lettering.

It might be supposed that churches, language and writing systems excluded from the Roman empire (Eastern and Western) would have little to do with Greek or Latin. Yet scripts can survive and flourish in heterographic rivalry and distinctiveness. Syriac was particularly important in the rise and spread of Christianity. The concern of Gregory III to maintain the “Graeco-Latin script” is attributed to the weakness of Rome, then under threat from the Franks. (Morison 126) This threat would be realized in 774 when Charlemagne conquered Italy and restored roman capitals. As a Frank, Charlemagne rejected all Rome’s accommodations with Byzantium, including those involving alphabets and scripts. After Charlemagne the Western Church ignored the Eastern (formally so after 1054), and barely acknowledged the non-Chalcedonian churches. Latin alone would become the language of the west, and “roman” its single alphabet and writing system.

The lettering and the language of three of the Miaphysite churches—Ethiopian/Ge’ez, Coptic, Syriac—is Semitic, related to Hebrew and Arabic. A language and a script marginalized in the history of the West may yet be central in other figurations. Syriac has a writing-system derived (like Hebrew) from Aramaic; its influence may be detected in the development of scripts across Asia. Today, apart from the diaspora, the regular use of Syriac script is largely confined to the monastic communities of Tur Abdin and the surrounding area, where it is now in grave peril. Yet Syriac lettering once extended its reach to Rome and further west (to Gaul, perhaps Britain) and eastwards across India to China. In the early years of the Tang Dynasty (c. 640 CE) Nestorian texts were translated from Syriac; this event was recorded on the Xi’an stele in c. 781; the stele was buried for safety c. 845 and discovered c. 1625 (Jingyi Ji 41–42). The letters and marks incised on the Xi’an stele were first published by Athanasius Kircher in 1667. Kircher was careful to represent accurately the Chinese characters as well as the Syriac. This is therefore the first substantial text in Chinese to be published in Europe.

How are we to measure the significance of this: that these, among the first Chinese characters printed in Europe, were glossed by Syriac? The Xi’an stele has often been compared to the Rosetta Stone, though only a few of the nineteen hundred Chinese characters are glossed in Syriac, of which there are some fifty words (Keevak 23–24). One factor to be considered is that Sogdian, the lingua franca of central Asia during the Tang Dynasty (c. 600–900 CE), was written in a script derived, like Syriac, from Aramaic. Thus through the Nestorian missionaries, the Christian message may have been welcomed in China on account of its letters, recognizable to readers of Sogdian. Such a claim must be hypothetical, but the Xi’an stele deserves study in a wider context.

A further hypothesis suggests that the discipline of palaeography in Europe owes its founding to the Xi’an stele: that it was Athanasius Kircher’s careful representation of Chinese and Syriac characters in 1667 that led Mabillon, Montfaucon and Wanley, in the years following, to reckon roman characters as no less deserving of meticulous accuracy in the copying.

On the Xi’an stele many of the proper names in Syriac (and Persian and Sanskrit) are “left untranslated”. This has scribal consequences. There are always difficulties in transliterating proper names from one alphabet to another, and those difficulties can leave their marks, just as French proper names—or terms taken rather than translated—“taint” English letters with their diacriticals: is François in the café? It cannot be said that there is transliteration between Syriac and Chinese, but a sort of “phonetic transposition” between two different orders of signs. The Chinese writing-system was presumably inflected, to some degree, by the representing of such proper names as Allaha, Mshiha and Satana.1

The occlusion of Syriac is but one instance of what is lost when the story of roman lettering is told as a triumph of visual clarity (“transparency”) and phonetic efficiency. Stanley Morison, himself the designer of Times New Roman, is one of the very few historians of writing and printing to have understood the part played by ideological considerations. Writing systems do not develop only with reference to phonetics and technology, though that is the assumption of most general accounts (Diringer, Coulmas, Fischer). Morison apprehends the ideological force of script, where political considerations may be allowed to interfere with communication. Morison presents some cases of letters whose clarity and efficiency in the reading were deemed less important than their distinctive appearance, and some in which ease of “native reading” was sacrificed so as to make the letters more accessible to non-native readers.

We have seen that each of the Miaphysite churches has its own language and its own script. Though we may think of it as “normal”, the roman script is unusual in representing many different languages, in serving a diversity of phonetic and philological needs. It was only on account of its political and cultural prestige, its geographical extensiveness and the scribal authority of the western Church, that the roman alphabet was adapted to give scribal representation to the vernacular languages of western Europe. The suppression of runes, the excision of now archaic letters, would inconvenience native readers while making each language of Western Europe at least literally accessible to readers of all of them.

Paul Saenger argues that English was the first vernacular to use the roman alphabet for its writing system, c. 800. (Saenger 96–97) The priority accorded thus to English, followed by German, may be due to the fact that among the vernacular languages spoken within and around scriptoria, only these were considered distinct from Latin: “there is no evidence before about 950 of a conscious awareness that the spoken Romance dialects were languages different from Latin” (Saenger 101).

While roman lettering may appear universal and normative, its very reach and versatility make it truly exceptional. Most writing-systems serve one language only. The Greek alphabet has served diverse texts, from Homer and Plato to the Gospels, and is now the official script of a nation within the European Union. Yet only on very few occasions has Greek been used for the writing of another language. Greek typography has changed little since its earliest printing in Venice by Aldus Manutius, c. 1490; the consistent look of Greek typography over five hundred years—when contrasted with all the changes in roman letters—may be due to the limited demands made on any one script by a single language.

Cyrillic, by contrast, was first created, for missionary purposes, in the ninth century, just as the Greek, Latin and Arabic systems were developing cursive forms, and when English was first being written in roman letters. Cyrillic was a missionary script, based on Greek letters modified to accomplish what the Greek alphabet habitually fails in, the representing of another language.2

Thus Cyrillic became the writing system for a number of Slavic languages and is today the native script of some 250 million people, the fifth most widely used writing-system in the world; it has also been subjected to national modifications, notably in Russian after the Revolution of 1917. Yet Cyrillic is not used in those Slavic nations such as Poland whose church is not Greek but Roman: letters follow the faith. The look of certain Polish or Czech names, and the profusion of diacritical marks in those Slavic languages that use roman letters, suggest that Cyril and Methodius (themselves of the Latin Church, not the Greek, to whose devising the script is traditionally assigned) understood letters and their proper limits. Though Christianity must be preached to the ends of the earth, even the phonetic versatility of roman could not adequately represent Slavic tongues.

It is political and economic considerations, along with cultural prestige—rather than any philological principles—that continue to drive the global extension of the roman alphabet, sometimes controversially. As in Polish or Hungarian there can be an obtrusiveness of names created by the squeezing of unfamiliar sounds into familiar letter-forms, and those familiar letters into unfamiliar combinations.

For those tentatively reading a text in a foreign language that uses the roman alphabet, proper nouns can afford security, an immediacy of recognition and understanding: this is the assimilative advantage of the roman letters for non-native readers (Lock “Conveying”). Yet in a translation of a text from a language that uses another writing-system, proper nouns create serious problems. We can hardly speak of “transliteration” where there’s no equivalence in the status of “letters,” as in the “phonetic transposing” of Syriac names into Chinese. Its challenges leave a permanent mark on the writing system into which the transposing is made. This need not be a diacritical mark but, between alphabets, merely an unprecedented sequence of letters, such as, from Cyrillic, initial “Tch” for Tchaikovsky—though not for Chekhov: an index can be a transliteral vexation. An alternative is to “translate” a proper noun according to its common meaning, as the name Cephas (meaning “stone” in Aramaic) becomes Peter in Greek (and later, in Danish, Steen). But when no such semantic equivalent is available the writing system needs to incorporate a name as strange as the letters of which it’s formed. And where there is “transcription between writing systems”, or “phonetic transposition,” there is a possibility, even an inevitability, of interference in the receiving system.

Nowhere has “phonetic transposition” been more tried and tested than in the most translated of all the world’s texts, the Bible, one part of which uses Hebrew letters, the other, Greek. As Greek letters correspond quite closely to Roman letters, transliteration poses relatively few difficulties: should phi be transliterated as f or ph; kappa as k or c? The problem becomes serious, however, with the transcription into Greek, and then into Latin, of Hebrew names. As Jesus does not appear in the Hebrew scriptures, nor in Hebrew script, but only in Greek, we know him (in English) as Jesus; he does however have a namesake in the Old Testament, though we are unlikely to identify him as such, for there the Hebrew form of the name is transliterated as Joshua. It is as though, rather than think about what it might mean to call Joshua by the name of Jesus or to address the son of God as “Joshua Christ”, we would prefer to remain graphically indifferent.

Difficulties are more likely to become evident elsewhere than in a canonical text venerated into a semblance of stability. The pursuit of scientific systems of transliteration engaged some of the most brilliant minds among the scholars and the administrators of the British Empire, notably in India, with effects sometimes comic, usually cumbersome, and always visually disruptive. Few were of much value to those who for whom they were intended (Majeed; Lock “Heterographics” 105). We are inclined to suppose that the roman alphabet in all its logic and phonetic purity has no problems of its own. The problems arise when one must transliterate into roman from writing-systems that are supposedly “less efficient” than the roman: either because they lack vowels or because (as with Chinese) it is thought by roman readers to use “pictures” or “ideas”: not quite “proper writing” as the roman alphabetical triumphalists would have it. Haun Saussy warns against speculations over Chinese script by those who see it as an example of “what can go wrong with language when it is not adequately released from its debt to materiality” (Bachner 20).

The alphabet devised within Latin has extended its reach to dozens of languages, often with a conspicuously ill-fitting appearance. Even if we consider only the languages of western Europe we may notice instances of frustration with the roman alphabet. The familiarity of these writing systems conceals from us a series of literal constraints and diacritical rebellions. These ought properly to be understood as difficulties of transliteration.

English, first among all the Germanic languages to adopt roman letters for its writing system, never thereafter found a more precise way of representing either þ (thorn) or ð (eth) (or the difference between the voiced and unvoiced dental fricative) than “th”. Icelandic was among the last to adopt roman letters, and it preserves these “archaic” characters in all its fonts. English has been remarkably restrained in its workings with the roman alphabet; for centuries it has found no reason to add other letters than those (J, U, W) in excess of the twenty-three characters of the Latin alphabet. Even more strikingly, English has entirely eschewed diacritical marks. They certainly order these matters differently à la française, and in almost every other adaptation of roman letters to a western European language. Every time a change is made to a national writing-system there is a new challenge for transliteration. That challenge is particularly acute for those transliterating into English not only because it is the “global target language” but also because, thanks to its resistance to diacriticals, English remains closest to Latin in its graphic resources, in both their economy and their versatility. Almost every reform of another language introduces a distinction therein, and yet another sign has to be rendered into one of the constant twenty-six of English.

The dominance of English has consequences for digital writing systems; though many accents are optional on keyboards, there are extra keys available, in Denmark, for Ø, Å and Æ. Those keys should be avoided in writing by email to an outsider, for there’s no transliteration at the other end. And so transliteration becomes a daily concern, even between and among writers of languages all of whom use the roman alphabet.

Or do they? One grows wary of claims for the global validity of the roman alphabet. The Universal Postal Union formed in Berne in 1876 was among the first institutions to give the weight of international mandate for the use of roman letters: even today the U.P.U. instructs the world that all addresses, regardless of nation, must be written “in roman letters and Arabic numerals”. (Note the respective upper- and lower-case initials, here carefully replicated.) Would it be quibbling to ask how many letters the U.P.U. considers “roman”? And what of accents, supplied in each language (English excepted) to make good some of the phonetic deficiencies of roman letters registered by speakers and listeners in diverse vernaculars?

Though so many European languages seem to share the roman alphabet, how many of them can be said actually to share a writing-system? Are there any two languages that use the roman alphabet and share a single set of characters with one set of accents, diacritical marks, pointings and other extra-alphabetical signs? There is no major European language today that confines itself to the 23 letters of the Latin alphabet, and all except English use diacritical marks as well. And there are no two languages using the roman alphabet that have identical graphic sets. The heterographic is not only present between roman and Greek, Arabic and Chinese, Cyrillic and Syriac; the heterographic mediates English and French, French and German, Latin and all its “successor alphabets” (Lock “Heterographics” 110–12).

The overlooking of the graphic has endured far too long, and has obscured a major instrument of ideological power and control, as we have seen in the elision of Damasian and its uncial and Byzantine consequences, the better to tell the story of the continuity of Roman Capitals. Attentiveness to the graphic can also disclose important matters of a strictly linguistic nature: it may be that “a language” is constituted, and can best be defined, not by any distinctive set of phonetic or even philological features, but rather by a unique set of graphic marks. One might argue that among the users of roman letters there are precisely as many writing-systems as there are languages. This would make clear the distinction between letters as independent characters and letters as members of an alphabetical set. Instead of supposing that so many languages use the roman alphabet, we might more accurately say that while most of the characters used are to be found in the roman alphabet, each language has its own.

The letters of the roman alphabet, intended for Latin, now offer graphic hospitality to many languages. The standard modern western histories of writing and script—e.g. Diringer, Coulmas, Fischer—invariably invoke a modifier such as “writing proper”, or “true” or “complete” writing, by which to distinguish what is written in a “phonetic alphabet” from what is termed “incomplete writing” or “pre-writing”.

Yet there is an indispensable aspect of roman letters so obvious as to have blinded all enquiry: what is punctuation but the non-phonetic supplement to phonetic signs? Does any other writing system depend so heavily as roman on punctuation? Many others do without it entirely. Punctuation asserts itself in the West, in the early Middle Ages, precisely when roman letters are being deployed to represent languages other than Latin. The development of cursive in the ninth century comes with punctuation, and with the emergence in writing of languages identifiably not Latin, first among them English.

Any writing-system must combine phonetic with non-phonetic features. There will always be signs—including spaces—that cannot be subsumed in voice, and which form a “visual residue”. Simultaneously with the development of cursive came a visual system of pointing that is now a fundamental part of all writing-systems that use roman letters. It is easy to celebrate the purely phonetic nature of the roman alphabet as long as one avoids the term “writing system”. But in considering the system of writing of which roman letters are a part, there is no hiding the importance of punctuation. Nor that of those white spaces that also emerged in the ninth century, opening up gaps between words to ease their reading in silence (Parkes Pause; Saenger).

How do we distinguish marks of punctuation from the letters? Scholars for whom the letters of the roman alphabet form their native script devised the term “writing-system” (in the mid twentieth century) to describe all those scripts less phonetic than their own. Yet punctuation is integral to all the writing-systems that use the characters of the roman alphabet. The power of the west and the hegemonic force that roman letters have exercised over international scholarship seem to have inhibited a due degree of criticism and analysis. Scribes and readers of Chinese, Korean, Japanese, Pali, Devanagari, Armenian, Syriac, Arabic, Hebrew, Cyrillic, Ge’ez, suffer no phonetic deficiency with regard to roman readers and writers. To the contrary, it might be asked whether there is another writing system, apart from roman, that has so many signs that cannot be voiced—including spaces. Punctuation holds words apart, and it holds the page together; it is a visual code that guides the voice but cannot be reduced to the voice or subsumed within it. One does not ask how to pronounce a comma.

That we have so long believed ours to be a phonetic alphabet is a vast ideological imposture. The suppression of the graphic and the non-phonetic has enabled native readers of roman to argue for the supremacy of their letters over all others, and (since “the Phoenician miracle” so often invoked by western historians of roman writing) to present the roman alphabet as having developed only in the service of technical efficiency and phonetic accuracy. Stanley Morison argues that there is no writing system free from ideological purpose, nor one whose developments are not motivated by political considerations. Morison has been largely neglected; instead we continue to suppress our sense of how writing looks. We have thus blinded ourselves to their implication—through their very forms—in the ideological and the material. Those who look at scripts, instead of “merely reading” them, will discover that there is nothing neutral in their constitution and that their use can never be wholly innocent.

My interest in Syriac has been stirred by the plight of other-lettered refugees, and informed (however inadequately) by Coakley on Syriac typography. That one working in Syriac should observe that it is “not part of the western scholarly tradition” to attend to the font or typeface of what we study, nor of what we publish, is a telling lament. For this is no slight dereliction. Its effect is to occlude the heterographic as a historical force, and to render roman letters almost impervious to analysis or challenge. Yet on account of our letters—first called “roman” just five hundred years ago—we still owe a debt to materiality; they remain immersed in the contestings and displacings of this world and its empires.

Works Cited

Bachner, Andrea. Beyond Sinology: Chinese Writing and the Scripts of Culture. New York: Columbia UP, 2014.

Coakley, J.F. The Typography of Syriac: A Historical Catalogue of Printing Types, 1537–1958. Delaware MD: Oak Knoll/London: British Library, 2006.

Coulmas, Florian. The Writing-Systems of the World. Oxford: Blackwell, 1989.

Diringer, David. Writing. London: Thames and Hudson, 1962.

Fischer, Steven Roger. A History of Writing. London: Reaktion, 2001.

Heyworth, P.L., ed. Letters of Humfrey Wanley: Palaeographer, Anglo-Saxonist, Librarian, 1672–1726. Oxford: Clarendon Press, 1989.

Jingyi Ji. Encounters between Chinese Culture and Christianity: A Hermeneutical Perspective. Münster: LIT Verlag, 2007.

Keevak, Michael. The Story of a Stele: China’s Nestorian Monument and its Reception in the West, 1625–1916. Aberdeen, HK: Hong Kong UP, 2008

Lock, Charles. “Conveying the Silence: Towards a Grammatological Theory of Translation.” In Literary Research/Recherche Littéraire, 16: 32 (1999), 182–201.

Lock, Charles. “Heterographics: Towards a History and a Theory of Other Lettering.” In Angles on the English-speaking World, vol. 6 (2006), 97–112.

Louth, Andrew. Greek East and Latin West: the Church A.D. 681–1071. Crestwood NY: St. Vladimir’s Seminary Press, 2007.

Majeed, Javed. “Modernity’s Script and a Tom Thumb Performance: English Linguistic Modernity and Persian/Urdu Lexicography in Nineteenth-Century India.” In Trans-colonial Modernities in South Asia, eds. Michael S. Dodson and Brian A. Hatcher. London & New York: Routledge, 2012, 95–115.

Morison, Stanley. Politics and Script: Aspects of Authority and Freedom in the Development of Graeco-Latin Script from the Sixth Century B.C. to the Twentieth Century A.D. Oxford: Clarendon Press, 1972.

Parkes, M.B. Pause and Effect: Towards a History of Punctuation in the West. Aldershot: Scolar, 1992.

Parkes, M.B. “Set in Their Own Ways: Scribes and Book Hands c. 800–1200.” In Their Hands Before Our Eyes: A Closer Look at Scribes. Aldershot: Ashgate, 2008.

Saenger, Paul B. Space between Words: The Origins of Silent Reading. Stanford CA: Stanford UP, 1997.

Sampson, Geoffrey. Writing Systems: A Linguistic Introduction. Stanford CA: Stanford UP, 1985.

Shaw, Paul, ed. The Eternal Letter: Two Millennia of the Classical Roman Capital. Cambridge MA: MIT Press, 2015.

The uncapitalized initials in the title are deliberate: roman graphic conventions may best be observed when resisted.

“Mshiha” is the roman transliteration of the Syriac “Messiah”; the influence of Syriac on Chinese may also be registered through the 9th–14th centuries by the proximity of Old Uyghur, an alphabet derived ultimately, through Sogdian, from Syriac. The present status of the Uighur in Xinjiang is, not least, a heterographic matter.

It may be because the Greek alphabet is universally familiar yet has so seldom been used to represent another language that its characters can function as symbols in mathematics and other non-vocalic codes.



Content Metrics

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 29 29 11
PDF Downloads 6 6 2
EPUB Downloads 3 3 0