Pause and Effect: An Introduction to the History of Punctuation in the West 
by M.B. Parkes.
Scolar, 327 pp., £55, September 1992, 0 85967 742 7
Show More
Show More

The history of punctuation is bound up with the most important shift in the theory of writing to have taken place in our culture. The written word began as a record of speech, a priority of voice over text which held sway in the ancient world and was literally (i.e. graphically) enforced. Reading meant reading aloud; texts were the libretti of performances so there was no need for elaborate pointing. Indeed there is no need even for the most minimal punctuation of all, word division. Classical texts were copied in scriptio continua (joined-up writing with a vengeance) so that the opening of, say, ‘Resolution and Independence’ would have looked like this (only more so, because there would have been quirks of orthography, such as contractions and elisions, to contend with as well):


It may be asked how readers were supposed to tell that ‘allnight’ at the end of the first line belonged to ‘aroaringinthewind’ rather than ‘theraincameheavily’, or that ‘inthedistantwoods’ belonged to ‘thebirdsaresinging’ rather than ‘overhisownsweetvoicethestockdovebroods’. The answer is that readers were supposed to use their judgment. If the rhythm of the poem suggested to them a series of end-stopped lines, then that is how they would deliver them; if they wished to introduce enjambment where the sense did not disallow it, they were free to do so. The exercise of this judgment has social as well as literary implications. As Parkes says,

the merit of scriptio continua was that it presented the reader with a neutral text. To introduce graded pauses while reading involved an interpretation of the text, an activity requiring literary judgment and therefore one properly reserved to the reader. In ancient Rome readers of literary texts were mostly a social élite, whereas full-time scribes were usually freedmen or slaves.

In this hierarchy of readers and scribes, what has happened to the figure of the author? The inevitable consequence of scriptio continua is that authors must share interpretative power with their readers. Socrates complains of this surrender of authority in Plato’s Phaedrus: a writing cannot correct a mistaken reading of itself, cannot impose its authority on the reader, cannot answer back. The history of punctuation is in part the history of authors’ attempts to wrest power from readers, and of scribes’ attempts to wrest power from both readers and authors, and of the attempt by institutions such as the Church to control every aspect of the production, transmission and interpretation of texts.

The ‘moment of writing’ came about (and is only now, and only perhaps, beginning to pass away) because of a revolution in the way writing was perceived. Parkes compares Augustine, for whom ‘letters were invented that we might be able to converse even with the absent,’ and who followed Aristotle in seeing written words as the ‘signs of sounds’, with Isidore of Seville, three centuries later, who, saw letters as having ‘the power to convey to us silently the sayings of those who are absent’. Isidore’s view of writing silences both author and reader, and makes the text meaningful in itself. But if writing is not the medium of speech, Plato is right to be suspicious of its impersonality. Writing opens up a space of hermeneutic freedom, in which the reader’s understanding (not to mention his imagination) is licensed to play. Such spaces are dangerous if the text itself is held to articulate an authoritative meaning.

Parkes convincingly argues the decisive role of Christianity in the evolution of textual theory and, therefore, of punctuation. ‘Christianity was a religion of the book, and grammatical culture was fundamental to the study of Scripture.’ A sacred text can tolerate many kinds of interpretation, but it cannot afford to leave its readers in doubt about its basic syntax. Theologians soon realised that unpunctuated texts of Scripture did just this. For Augustine the problem was one of how to read aloud: where, for example, to pause in the correct (that is, orthodox) delivery of the opening verse of St John’s Gospel: in principio erat verbum et verbum erat apud deum et deus erat verbum hoc erat in principio apud deum. It was heretical to pause after et deus erat, Augustine pointed out: it denied the divinity of the Word. How was the reader to know this? By knowledge of Christian doctrine derived from other parts of Scripture, and by the authority of the Church; but the former argument is circular, and the latter merely dogmatic. Augustine’s difficulty would have been (and eventually was) resolved by punctuation, a means of putting the doctrinal fix in texts.

When God is the author his text must be stabilised, since such stability is essential to the transmission of meaning. The editors, scribes and commentators who produced the text of the Bible over many centuries were engaged in a collective project of incalculable significance for our reading of all texts. For the idea of the divine author and his unvarying text migrated into the non-sacred literature, and still dominates our thinking about the transmission of texts. A modern edition of Wordsworth gives the following text of the first stanza of ‘Resolution and Independence’:

There was a roaring in the wind all night;
The rain came heavily and fell in floods;
But not the sun is rising calm and bright;
The birds are singing in the distant woods;
Over his own sweet voice the Stock-dove broods;
The Jay makes answer as the Magpie chatters;
And all the air is filled with pleasant noise of waters.

A roman numeral designates the stanza and initial capitals mark the beginnings of each line; other capitals (litterae notabiliores) highlight and personalise the names of the birds, in contrast to the other nouns, which describe the weather or organic nature (wind, rain, sun, woods). The feature which stands out, however, is the repetitive semi-colon, which gives the stanza a heavily end-stopped rhythm and draws attention to the preponderance of half-rhymes. (The lines themselves are remarkably integral, with almost no scope for internal punctuation.) The substitution of commas for semi-colons in lines 2, 5 and 6 would produce a lighter and more rapid sequence; a colon in line 4 might make the grammatical structure clearer. An exclamation mark at the end would indicate a greater degree of response to the fresh morning landscape. But the reader is at liberty to supply none of these variations.

Who, then, is in charge of the ‘effects’ of punctuation? Not Wordsworth: he was a notoriously poor speller and an erratic and inconsistent punctuator, a lamentable product, no doubt, of declining educational standards in his day. His texts were subject to the intervention of others (whether at home or in the printing-house) on whom he relied to produce an acceptable text. It is difficult to tell without Wordsworth’s own manuscript, but I am willing to bet that the semi-colon was as foreign to him as it is to many of my students. Most authors in the period used commas and dashes, singly or in combination: the dash as an all-purpose punctuation mark appears in Byron’s letters, for example, and in Browning’s early manuscripts. The semi-colon was an invention of humanist printing at the end of the 15th century, and is unusual both for that reason (since most punctuation marks derive from handwritten forms) and, Parkes suggests, in being a deliberate invention by Venetian printers. It has always been a mark more favoured by printers than writers. It does not appear in the earliest extant manuscript of the poem, which is in Sarah Hutchinson’s hand:

There was a roaring in the wind all night
The rain came heavily and fell in floods
But now the sun is rising calm and bright
The Birds are singing in the distant woods
Over his own sweet voice the stock-dove broods
The Jay makes answer as the magpie chatters
And all the air is fill’d with pleasant noise of waters

If this is anything like an accurate transcript of what Wordsworth wrote, it shows a poet with a primitive, not to say Classical approach to punctuation (especially in the lack of a stop at the end of the stanza). The next earliest manuscript of the poem reads like this:

There was a roaring in the wind all night;
The rain came heavily, – fell in floods;
But now the sun is rising calm and bright,
The birds are singing in the distant woods;
Over his own sweet voice the stock dove broods,
The jay makes answer as the magpie chatters;
And all the air is fill’d with pleasant noise of waters.

This is a copy by Coleridge of a transcript made by Dorothy Wordsworth. On the assumption that the punctuation is Coleridge’s, he may be said to be the poem’s first editor (and not a bad one, either). But the one thing that is clear is that Wordsworth had already surrendered control over his text, a surrender which the process of printing would formalise and legitimate.

Wordsworth is an extreme but not unusual case. Before the mid-19th century the number of authors who exercised close – or even any – control over the punctuation of their texts was tiny: punctuation was the province of the publisher’s reader and the compositor. So the divinity of the authorial word has a crack in it; authority over meaning is devolved to the producers of the physical text, who become its primary interpreters if not actually its collaborative authors. Not surprisingly, this process generated errors which have vexed literary critics jealous of their authors’ creative and interpretative autonomy. In their recent edition of Sons and Lovers, Helen and Carl Baron point out that Lawrence’s punctuation was extensively revised by the printers of the first edition, and that their revisions occasionally resulted in blatant changes of meaning. Lawrence wrote: ‘Perhaps it was essential to him – as to some men to sow wild oats.’ The printers set: ‘Perhaps it was essential to him, as to some men, to sow wild oats.’ The Barons comment: ‘Lawrence’s dash indicates that “wild oats” are not the experience Paul wanted.’ But there is the awkward fact that, to quote the Barons again, ‘With regard to punctuation, Lawrence largely left the printers’ changes intact’ – including the above example. We may not like the result in this case, but we should recognise it as of the essence of print culture, in which no text can claim an unmediated authorial origin.

Many authors did more than accept the normalisation of their punctuation: they demanded it. ‘There are innumerable faults in the punctuation,’ Rousseau complained when he saw the first proofs of his essay ‘On the Origin of Inequality’; he was complaining about his own errors. ‘When I said that I wanted the manuscript to be followed exactly, I did not mean this to apply to the punctuation, which is thoroughly defective. Ask the Abbé Yvon to be so good as to put it right in the proofs which are to follow.’

Parkes fails to take account of this phenomenon in his discussion of the development of punctuation as an interpretative system: he persistently attributes its effects to authorial intention where the evidence is at best problematic. ‘Since by the mid-19th century writer and reader had come to share an extended repertory of graphic conventions, novelists deployed punctuation as a feature of the pragmatics of the written medium to assist their mimetic ambitions,’ he writes at one point. But this deployment was as much the result of a normalising and standardising process in the book trade as of individual authors consciously making use of ‘graphic conventions’. Of course this process was not one-sided, either; the point is that authors and printers collaborated to restrict the interpretative scope of the reader.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 15 No. 3 · 11 February 1993

Reviewing M.B. Parkes on Punctuation in the West (LRB, 7 January), Danny Karlin earns our gratitude by opening the subject in arresting ways. He also leaves me feeling (such is readerly ingratitude) that his apertures need enlarging. He quotes with approval Parkes’s remark: ‘the merit of scriptio continua was that it presented the reader with a neutral text.’ After it has left the author, every text is neutral: in the strict sense of being unknown until known, interpreted by a reader. The rest is a matter of codings that can or cannot be known (Linear A, B; a ‘foreign’ language etc). The manner of knowing may be trivial or important to the author, the reader and those between: oral v. written transmission, scroll v. codex, manuscript v. printing, recited (holy writ, hymns) or played (stage) from memory v. read, etc. Karlin concludes, pace Parkes: ‘the point is that authors and printers’ – presumably also other intermediaries – ‘collaborated to restrict the interpretative scope of the reader’. In a sense that is true, but there is no meaning anything without the implication that all else is not being known. And it is a prior ‘point’ that those ‘authors and printers’ enable the suitably equipped reader to know something in the first place. Would his worship banish and choke all the authors and printers of the town?

The subject of ‘punctuation in the West’ is far more complex, which is not to say always earth-shaking. How much difference is made by the fact that the blind Milton ensured that his three greatest poems were the first in English numbered by lines and printed on ruled pages? What do people mean practically by crediting Seneca with invention of the paragraph? Is there really a conspiracy involved in the writing (or printing) of Hebrew and Arabic unpointed for vowels? Or the addition of such pointing for valued texts such as the Bible and the Quran?

The whole subject – ‘in the West’ – is parochial. Chinese and Japanese were written for centuries without punctuation, without word breaks, and of course with no capitals. (Korea is its own special case.) For long, Chinese and Japanese verse was printed ‘like prose’; that is, without breaking for lines as in all Karlin’s quotations of that stanza from ‘Resolution and Independence’. The earliest mss of the Genji Monogatari (Tale of Genji) are the harder to read for being totally unpunctuated and without paragraphs in any familiar sense: the only indentations signal the beginning of a poem. The harder also for being written in that kana syllabary (and not modern kana although reproducible in that) which women used: i.e. with few Chinese characters. There were also none of today’s familiar diacritics indicating voicing of consonants (zu from su, pa from ha etc). Yet young women (and old men) once read the graphs with ease, and probably aloud, whether to themselves or others. Only a very small number today can do so. In fact those lengthy, complex sentences (if they are sentences in our sense) omitting topics and subjects of verbs, require glossing as to who is speaking or thinking (an alternative not always clear) to or about whom. We rely on editors for signals: addition of abbreviated names with particles along with punctuation. Each of us reading a modern text knows that the editors have intervened with their interpretations, ruling out others when they rule in their own, although not thereby wholly controlling what individual readers will think, something impossible. Some editions of the Genji (but not solely it) even print ‘modern language translations’ (gendaigo yaku) as aid to the serious and no doubt as cribs for the lazy. How regrettable the interference – in a puristic sense such as applies to no practical end! The assisted (and thereby limited) understanding provided by a scholarly specialist of tenth and 11th-century Japanese is unquestionably a limiting (and a crucial) assistance.

The subject is often truly important and, even when not truly important, interesting. But there is no need to demonise ‘authors and printers’ – or to restrict the matter to that ‘in the West’. Surely a more reasonable place to begin is with scepticism about our own powers unaided by those breaks, marks, larger letters, initials, numbers, and much else that we take for granted solely because we think they can always be assumed?

Earl Miner
Princeton University

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences