The nine basic marks of punctuation—comma, dash, hyphen, period, parenthesis, semi-colon, colon, space, and capital letter—seem so apt to us now, so pipe-smokingly Indo-European, so naturally suited in their disjunctive charge and mass to their given sentential offices, that we may forgivably assume that commas have been around for at least as long as electrons, and that while dialects, cursive styles, and typefaces have come and gone, the semi-colon, that supremely self-possessed valet of phraseology, is immutable.

But in fact the semi-colon is relatively modern. Something medieval called a punctus versus, which strongly resembled a semi-colon, though it was often encountered dangling below the written line, had roughly the force of a modern period; another sign that looked (in some scribal hands) exactly like a semi-colon was a widely used abbreviation for several Latin word endings—atque could appear as atq;, and omnibus as omnib;. But the semi-colon that we resort to daily, hourly, entered the picture with the first edition of Pietro Bembo’s De Aetna two years after Columbus reached America, the handiwork of Aldus Manutius the Elder (or someone close to him) and his tasteful punch-cutter, Francesco Griffo. The mark, we are told by Dr. Malcolm Parkes, its historian, took much longer than the parenthesis did to earn the trust of typesetters: shockingly, its use was apparently not fully understood by some of those assigned to work on the first folio of Shakespeare.

And it is of course even now subject to episodes of neglect and derision. Joyce much preferred the more Attic colon, at least in Ulysses, and Beckett, as well, gradually rid his prose of what must have seemed to him an emblem of vulgar, high-Victorian applied ornament, a cast-iron flower of mass-produced Ciceronianism: instead of semi-colons, he spliced the phrases of Malone Dies and Molloy together with one-size-fits-all commas, as commonplace as stones on a beach, to achieve that dejected sort of murmured ecphonesis so characteristic of his narrative voice—all part of the general urge, perhaps, that led him to ditch English in favor of French, “pour m’appauvir“: to impoverish himself.

Donald Barthelme, too, who said that the example of Beckett was what first “allowed him to write,” thought that the semi-colon was “ugly, ugly as a tick on a dog’s belly”—but he allowed that others might feel differently. And still the semi-colon survives, far too subtle and useful, as it turns out, to be a casualty of modernism. It even participates in those newer forms of emotional punctuation called “smileys” or “emoticons”—vaguely irritating attempts to supply a sideways facial expression at the close of an e-mail paragraph—e.g., 🙂 and >%-(. The semi-colon collaborates in the “wink” or “smirk,” thus—;).

So our familiar and highly serviceable repertoire of punctles was a long time coming; it emerged from swarms of competing and overlapping systems and theories, many of them misapplied or half-forgotten. Petrarch, for example, used a slash with a dot in the middle of it to signal the onset of a parenthetical phrase. A percontativus, or backward question mark, occasionally marked the close of a rhetorical question even into the seventeenth century—Robert Herrick wrote with it. A punctus elevatus, resembling an upside-down semi-colon or, later, a fancy, black-letter s, performed the function of a colon in many medieval texts; when used at the end of a line of poetry, however, it could signal the presence of an enjambment. A nameless figure shaped like a tilted candycane served to terminate paragraphs of Augustus’ autobiography (14 AD), inscribed on his tomb. Around 600 AD, Isidore of Seville recommended ending a paragraph with a 7, which he called the positura. He also advocated the placing of a horizontal dash next to a corrupted or questionable text (“so that a kind of arrow may slit the throat of what is superfluous and penetrate to the vitals of what is false”), and he relied on the ancient cryphia, a C turned on its side with a dot in the middle to be used next to those places in a text where “a hard and obscure question cannot be opened up or solved.”

The upright letter C, for capitulum, developed into the popular medieval paragraph symbol, ¶, called at times a pilcrow or a paraph. Seventh-century Irish scribes were in the habit of using more points when they wanted a longer pause; thus a sentence might end with a colon and a comma (:,), or two periods and a comma (..,), or three commas together (,,,). At the close of the twelfth century, one of the dictaminists, a man named Buoncompagno, troubled by so much irreconcilable complexity, proposed a pareddown slash-and-dash method: a dash would mark all final pauses, and a slash would mark all lesser pauses. It didn’t take, although the “double virgula” (//) was used to separate sentences in the fifteenth century, and Edmund Spenser and Walter Ralegh sometimes hand-wrote with single slashes, rather than commas. A plus sign (+) stood for a period in a few early printed books; in others, it could set off a quotation.


Printing eventually slowed the pace of makeshift invention, forcing out many quaint superfluities, but novel marks, and surprising adaptations of old marks, may appear at any time. Besides smileys, on-line services have lately given rise to the ecstatic bracket hug of greeting: {{{{{{{{Shana!!!}}}}}}}}. Legal punctuation continues to thrive—the TM, the ®, and the ©, are everywhere. (The title of Jurassic Park is not Jurassic Park, but Jurassic ParkTM; likewise David Feldman’s Why Do Clocks Run Clockwise and Other ImponderablesTM.)

Especially fashionable now is the SM, as in “Forget Anything?SM“—observed not long ago on a triangular piece of folded cardboard beside the bathroom sink in a room at a Holiday Inn: a mark that modifies the phrase it follows to mean, “This is not merely a polite question regarding whether you have successfully packed everything you require during your stay, this utterance is part of our current chain-wide marketing campaign, and we are so serious about asking it of you that we hereby offer fair warning that if you or anyone else attempts to extend such a courtesy to another guest anywhere in the hotel industry in printed or published form, either on flyers, placards, signs, pins, or pieces of folded cardboard positioned at or beside a sink, vanity, or other bathroom fixture, we, the owner of this service mark, will torment and tease you with legal remedies.” Even the good old comma continues to evolve: it was flipped upside down and turned into the quotation mark circa 1714, and a woman I knew in college punctuated her letters to her high-school friends with home-made comma-shapes made out of photographs of side-flopping male genitals that she had cut out of Playgirl.

Until now, readers have had to fulfill their need for the historical particulars of this engrossingly prosaic subject with narrow-gauge works of erudition such as E. Otha Wingo’s sober Latin Punctuation in the Classical Age, or John Lennard’s extraordinary recent monograph on the history of the parenthesis, But I Digress (1991)—a jewel of Oxford University Press scholarship, by the way, gracefully written and full of intelligence, decked out with a complete scholarly apparatus of multiple indices, bibliographies, and notes, whose author, to judge by the startling jacket photo shaved head with up-sticking central proto-Mohawk tuft, earring on left ear, wilted corduroy jacket, and over-laundered T-shirt bearing some enigmatic insignia underneath), put himself through graduate school by working as a ticket scalper at Elvis Costello concerts. (A discussion of Elvis Costello’s use of the parenthesis in “Let Him Dangle” figures in a late chapter.)

At last, however, we have Pause and Effect, Dr. Malcolm Parkes’s brave overview: “an introduction,” so he unassumingly subtitles it, though it is much more than introductory, “to the history of punctuation in the West.” Not in the East, mind, or elsewhere—Arabic, Greek, and Sanskrit customs await a final fuse-blowing collation. (And according to the MLA index, there is Nanette Twine’s 1984 article on “The Adoption of Punctuation in Japanese Script,” in Visible Language, a journal that has recently done exciting things for the study of the punctuational past, to be assimilated; and, for canon-stretchers, John Duitsman’s “Punctuation in Thirteen West African Languages” and Carol F. Justus’s “Visible Sentences in Cuneiform Hittite.”) Though his punning title promises sprightliness, Dr. Parkes—Fellow of Keble College and Lecturer in Paleography at the University of Oxford—has produced a rich, complex, and decidedly unsprightly book of coffee-table dimensions, with seventy-four illustrative plates, a glossary, and, regrettably, no index rerum.

It is not an easy book to read in bed. Because of the oversized folio format, each line on the page extends an inch or so longer than usual, resulting in eye-sweeps that must take in fourteen words at a time, rather than the more comfortable ten or eleven. As his shoulder muscles tire of supporting the full weight of the open book, the reader, lying on his left side, finally allows it to slump to the mattress and assume an L-position, and he then attempts to process the text with one open eye, which, instead of scanning left to right, reads by focusing outward along a radically foreshortened line of type that is almost parallel with his line of sight, skipping or supplying by guesswork those words that disappear beyond the gentle rise of the page. The gaps between each word narrow, hindering comprehension, although they never achieve that incomprehensible Greek ideal of page-layout called scriptio continua, in which the text is recorded unspaced as solid lines of letters.

And why, in fact, did the Greeks relinquish so sensible a practice as word-spacing, which even the cuneiformists of Minoan Crete apparently used? Lejeune, for one, finds this development “remarquable“; but even more remarquable is the fact that the pragmatic Romans had wordspacing available to them (via the Etruscans), in the form of “interpuncts,” or hovering dots between each word (a practice successfully revived by Wang word-processing software in the 1980s), which they too abandoned in early Christian times. “For this amazing and deplorable regression one can conjecture no reason other than an inept desire to imitate even the worst characteristic of Greek books,” scolds Revilo P. Oliver. Dr. Parkes, on the other hand, theorizes that class differences between readers and scribes may have had something to do with the perseverance of scriptio continua—a scribal slave must not presume to word-space, or otherwise punctuate, because he would thereby be imposing his personal reading of the constitutive letters on his employer. There were also, in monkish contexts, quasi-mystical arguments to be made for unspaced impenetrability: a resistant text, slow to offer up its literal meaning, encouraged meditation and memorization, suggested Cassian (a prominent fifth-century recluse); and the moment when, after much futile staring, the daunting word-search-puzzle of the sacred page finally spaced itself out, coalescing into comprehensible units of the Psalter, might serve to remind the swooning lector of the miracle of the act of reading, which is impossible without God’s loving condescension into human language and human form.


Amid all this phylogeny, Parkes does not mention, nor should he necessarily mention, the more mundane developmental fact that scriptio continua comes naturally to children:





Children aren’t taught to forgo spacing; all their written models are properly spaced. Occasionally, as a concession to the recipient (or adult onlooker), they will go back and insert a virgule here and there between words for clarity. There is something so exciting about writing, perhaps, that, like barely literate five-year-olds, civilizations in the midst of discovering or rediscovering its pleasures and traditions take a while before they begin to care about casual readability—and consequently their scholars are said to study litterae, “letters,” not words.

In part as a result of the unspaced line, pointing was viewed from the beginning as a form of ornament, as well as a means of what Parkes calls “disambiguation.” Cassiodorus, the first great biblical pointillist, advised sixth-century monks to add punctuation “in order that you may be seen to be adding embellishment.” Alcuin wrote Charlemagne that “Distinctiones or subdistinctiones by points can make embellishment in sentences most beautiful.” Early medieval readers like Dulcitius of Aquino would decorate a work with dots and diples and paragraph marks as they read it and then proudly sign their name on the page: “I, Dulcitius, read this.” Punctuation, like marginal and interlinear commentary, seems at times to have been a ritual of reciprocation, a way of returning something to the text in grateful tribute after it had released its meaning in the reader’s mind.

Somewhat surprisingly, scholastic philosophy of the thirteenth and fourteenth centuries, which is, as Francis Bacon uncharitably observed, a vast and intricate cobweb spun from Aristotle, “admirable for the fineness of thread and work, but of no substance and profit,” and thus ideally decorative and mannerist rather than functional, pushed by logical and disputational energy rather than pulled by truth—the sort of era, then, in which you might expect punctuation to thrive—turns out in fact to be a dark, sad time for subdistinctiones. Parkes explains that the paradigmatic nature of the scholastic manuscript, with its repetitive queriturs and quaestios signaling to the reader precisely where he was in the formal structure of the argument, made a sophisticated punctuational tool-set unnecessary.

On the other hand, it may just be that the schoolmen, spending their days reading awful Latin translations from the Arabic of translations from the Greek, had no ear. Cicero himself disdained punctuation, insisting that the well-cadenced sentence would audibly manifest its own terminus, without the need of any mere “stroke interposed by a copyist”; but those who afterward took punctuation, and took Cicero, seriously—Cassiodorus, Isidore, Bembo, Petrarch—proved their allegiance by their virgulae: like archaizing composers who want to ensure a certain once-standard performance practice and therefore spell out every trill and every ritenuto, though their historical models offer only unadorned notes, these admirers could hear the implied punctuation of Ciceronian rhythm, and could in some cases duplicate his rollaway effects in their own writing, but they didn’t trust their contemporaries to detect a classicizing clausula without the help of visual aids.

Dr. Parkes’s own prose is serviceable and unprecious, if non-passerine. For those of us whose Latin never quite took flight, he has provided translations of every passage he quotes. He takes care from time to time to mention political developments as they impinge on the punctuational sphere: if some depradation or upheaval happens to have brought on “a situation hostile to grammatical culture,” he says so. The puzzling thing, though, is how casual Parkes is—this eagle-eyed paleographer, who has worked so hard to “raise a reader’s consciousness of what punctuation is and does”!—about his commas. Where are they? “Pausing therefore was part of the process of reading not copying….” “Before the advent of printing a text left its author and fell among scribes.” “The printing process not only stabilized the shapes and functions of the symbols it also sustained existing conventions that governed the ways in which they were employed.” And: “This increase in the range of distinctive symbols also promoted new developments in usage since the symbols not only enabled readers to identify more easily the functions of grammatical constituents within a sentence but also made possible more subtle refinements in the communication of the message of a text.”

Were it not for Dr. Parkes’s surefooted employment of the comma elsewhere, one might almost suspect that his was a case reminiscent of those psychotherapists who enter their profession because they sense something deeply amiss within themselves, or of those humorless people who buy joke books and go to comedy clubs to correct internal deficiencies. In a commentary accompanying a fascinating page of Richard Hooker, Parkes, or someone with whom he has shared (as he nicely says in his preface) “the burden of proofs,” flouts even the sacred law of the serial comma: “The notation series for indicating glosses notes and citations in the margins, based on letters of the alphabet in sequence, was also used in the Geneva Bible of 1560.”

Once, however, Parkes surprises us by unconsciously using the old-fashioned, eighteenth-century that-comma. It is the comma of Gibbon—

It has been calculated by the ablest politicians, that no state, without being soon exhausted, can maintain above the hundredth part of its members in arms and idleness

and of Gibbon’s model Montesquieu, in Nugent’s 1750 translation (twice)—

Plato thanked the Gods, that he was born in the same age with Socrates: and for my part, I give thanks to the Almighty, that I was born a subject of that government under which I live; and that it is his pleasure I should obey those, whom he has made me love

and of Burke (twice)—

Mr. Hume told me, that he had from Rousseau himself the secret of his principles of composition. That acute, though eccentric, observer had perceived, that to strike and interest the public, the marvellous must be produced…

and of Burke’s brilliant adversary, Thomas Paine—

Admitting that Government is a contrivance of human wisdom, it must necessarily follow, that hereditary succession, and hereditary rights (as they are called), can make no part of it, because it is impossible to make wisdom hereditary….

Parkes writes, “The punctuation of the manuscript has been so freely corrected and adapted by later scribes, that it is not easy to determine whether any of the other ecphonetic signs are also by the original scribe or whether they have been added.” The only other person I can think of who uses old-style that-commas with any consistency is Peter Brown, who, like Parkes, spends his time with Latin quod-clauses that have been punctuated by old German commentators. (A comma is still regularly used before a daß-clause in German.)

Another rarity in Parkes’s book, perhaps the very first of its kind, is the occurrence of the two halves of semicolon linked, not by a hyphen, but by a full-scale em-dash: semi—colon. This elongation could be Parkes’s secret way of protesting American trends in copy-editing, which would have the noun-unit spelled without any divisive internal rule at all: semicolon. Truly, American copy-editing has fallen into a state of demoralized confusion over hyphenated and unhyphenated compounds—or at least, I am demoralized and confused, having just gone through the manuscript of a novel in which a very smart and careful and goodnatured copy-editor has deleted about two hundred of my innocent tinkertoy hyphens. I wrote “stet hyphen” in the margin so many times that I finally abbreviated it to “SH”—but there was no wicked glee in my intransigence: I didn’t want to be the typical prose prima donna who made her life difficult.

On the other hand, I remembered an earlier manuscript of mine in which an event took place in the back seat of a car: in the bound galleys, the same event occurred in the “backseat.” The backseat. Grateful for hundreds of other fixes, unwilling to seem stubborn, I had agreed without protest to the closing-up, but I stewed about it afterward and finally reinserted a space before publication. (“Backseat” wants to be read as a trochee, BACKseat, like “baseball,” when in reality we habitually give both halves of the compound equal spoken weight.) Therefore, mindful of my near miss with “back seat,” I stetted myself sick over the new manuscript. I stetted re-enter (rather than reenter), post-doc (rather than postdoc), foot-pedal (rather than foot pedal), second-hand (rather than secondhand), twist-tie (rather than twist tie), and pleasure-nubbins (rather than pleasure nubbins).

The copy-editor, because her talents permit her to be undoctrinaire, and because it is, after all, my book, indulged me, for better or worse. In passing, we had a stimulating discussion of the word pantyhose, which she had emended to read panty hose. My feeling was that the word hose is unused now in reference to footwear, and that panty, too, in its singular form, is imaginable only as part of pantywaist or in some hypothetical L. L. Bean catalog: Bean’s finest chamois-paneled trail panty.” Pantyhose thus constitutes a single, interfused unit of sense, greater than the sum of its parts, which ought to be the criterion for jointure. And yet, though the suggested space seemed to me mistaken, I could just as easily have gone for panty-hose as pantyhose—in fact, normally I would have campaigned for a hyphen in this sort of setting, since the power-crazed policy-makers at Merriam-Webster and Words Into Type have been reading too much Joyce in recent years and making condominiums out of terms (especially -like compounds, which can look like transliterated Japanese when closed up) that deserve semi-detachment. (Joyce, one feels, wanted his prose to look different, Irish, strange, not tricked out with fastidious Oxford hyphens that handled uncouth nounclumps with gloved fingertips: he would have been embarrassed to see his idiosyncratic cuffedge and watchchain and famous scrotumtightening acting to sway US style-shepherds.) A tasteful spandex hyphen would have been, so my confusion whispers to me now, perfectly all right in panty-hose, pulling the phrase together scrotumtighteningly at its crotch.

I offer this personal note merely to illustrate how small the moments are that cumulatively result in punctuational thigmotaxis. Evolution proceeds hyphen by hyphen, and manuscript by manuscript—impelled by the tension between working writers and their copy-editors, and between working copy-editors and their works of reference (“I’ll just go check the Big Web,” a magazine editor once said to me cheerfully); by the admiration of ancestors, and by the ever-imminent possibility of paralysis through boredom. Are the marks that we have right now really enough? Don’t you sometimes feel a sudden abdominal cramp of revulsion when you scan down a column of type and see several nice little clauses (only one per sentence, of course: Chic. Man. St. § 5.91) set off by cute little pairs of unadorned dashes?

The nineteenth century didn’t think the dash on its own was nearly enough. Dr. Parkes ends his brief discussion of “The Mimetic Ambitions of the Novelist and the Exploitation of the Pragmatics of the Written Medium” with Virginia Woolf, so he (pardonably) avoids treating the single most momentous change in twentieth-century punctuation, namely the disappearance of the great dash-hybrids. All three of them—the commash ,—, the semi-colash;—, and the colash :—(so I name them, because naming makes analysis possible)—are of profound importance to Victorian prose, and all three are now (except for certain revivalist zoo specimens to be mentioned later) extinct.

Everyone used dash-hybrids. They are in Dickens, Wilkie Collins, Charlotte Brontë, and George Meredith. They are on practically every page of Trollope—

He was nominally, not only the heir to, but actually the possessor of, a large property;—but he could not touch the principal, and of the income only so much as certain legal curmudgeons would allow him. As Greystock had said, everybody was at law with him,—so successful had been his father, in mismanaging, and miscontrolling, and misappropriating the property.

Rapid writing will no doubt give rise to inaccuracy,—chiefly because the ear, quick and true as may be its operation, will occasionally break down under pressure, and, before a sentence be closed, will forget the nature of the composition with which it commenced.

The novels of a man possessed of so singular a mind must themselves be very strange,—and they are strange.

They are in Thackeray—

[…] the Captain was not only accustomed to tell the truth,—he was unable even to think it—and fact and fiction reeled together in his muzzy, whiskified brain

and in George Eliot—

The general expectation now was that the “much” would fall to Fred Vincy, but the Vincys themselves were surprised when ten thousand pounds in specified investments were declared to be bequeathed to him:—was the land coming too?

The toniest nonfictional Prosicrucians—De Quincey, Carlyle, Ruskin, Newman, Doughty—also make constant use of dashtards, often at rhetorical peaks:

It is well to be a gentleman, it is well to have a cultivated intellect, a delicate taste, a candid, equitable, dispassionate mind, a noble and courteous bearing in the conduct of life;—these are the connatural qualities of a large knowledge; they are the objects of a University: I am advocating, I shall illustrate and insist upon them; but still, I repeat, they are no guarantee for sanctity or even for conscientiousness, they may attach to the man of the world, to the profligate, to the heartless,—pleasant, alas, and attractive as he shows when decked out in them.

Pater, though he has been charged with over-sonorous purism, and is unquestionably at times a little light in his Capezios (to steal a phrase from Arsenio Hall), depends on punctuational pair-bonding to help him wrap up his terrific essay on style and his “Conclusion” to The Renaissance. Sydney Smith wrote that if Francis Jeffrey were given the solar system to review—Francis Jeffrey being the sour critic who said “This will never do” of Wordsworth’s Excursion—he would pan it: “Bad light—planets too distant—pestered with comets—feeble contrivance;—could make a better with great ease.” Emerson was a huge user of the semi-colash; in fact, of the fifty-two dashes in The American Scholar, only four, by my count, appear unaccompanied by either a semi-colon or a comma.

Hybrids become somewhat less common, though they are still easily found, after the turn of the century. Henry James employed a few in his early writing, but revised them out in the édition de luxe that began appearing in 1907. The 1911 Encyclopedia Britannica, a good test bed of Edwardian norms, resorts to them fairly frequently:

A large screw can, however, be roughly examined in the following manner:—(1) See whether the surface of the threads has a perfect polish…. (2) mount it between the centres of a lathe … (4) Observe whether the short nut runs from end to end of the screw without a wabbling motion when the screw is turned and the nut kept from revolving. If it wabbles the screw is said to be drunk.

And Edmund Gosse’s 1907 Father and Son has a lovely comma-softened dash that can be read as a wistful farewell to a form of punctuation in its twilight:

These rock-basins, fringed by corallines, filled with still water almost as pellucid as the upper air itself, thronged with beautiful sensitive forms of life,—they exist no longer, they are all profaned, and emptied, and vulgarised.

They pop up here and there in Norman Douglas, early J. B. Priestley, and Cyril Connolly. J.M. Keynes used a scattering of all three forms in his 1920 Economic Consequences of the Peace. For instance:

The policy of reducing Germany to servitude for a generation, of degrading the lives of millions of human beings, and of depriving a whole nation of happiness should be abhorrent and detestable,—abhorrent and detestable, even if it were possible, even if it enriched ourselves, even if it did not sow the decay of the whole civilized life of Europe.

But hybrid punctuation was doomed by then. Proust used two pre-war semi-colashes in the enormous “bed-rooms I have known” sentence in the opening of A la rehash; Scott Moncrieff removed them in his post-war 1922 translation. (Terence Kilmartin, good man, restored the original punctuation in 1981.) The dandiest dandy of them all, Vladimir Nabokov (who, I think, read Father and Son just as closely as he read Proust, drawn to its engaging combination of literature and amateur naturalism), used over sixty excellent comma-dash pairings in his first and quite Edwardian English-language novel, The Real Life of Sebastian Knight (1941). (For example, “‘A title,’ said Clare, ‘must convey the colour of the book,—not its subject.”‘) He used none at all in Speak, Memory: The New Yorker had sweated it out of him. In all of his later work I have noticed only one precious semi-colash: Humbert writes, “I remember the operation was over, all over, and she was weeping in my arms;—a salutory storm of sobs after one of the fits of moodiness that had become so frequent with her in the course of that otherwise admirable year!”

More precious still, to the punctuational historian, are the two instances of reversed commashes in Updike’s early novels—one in the Fawcett edition of The Centaur, and one of page 22 of the Fawcett Of the Farm:

As Joan comforted him, my mother, still holding the yard-stick—an orange one stamped with the name of an Alton hardware store—, explained that the boy had been “giving her the eye” all morning, and for some time had been planning to “put her to the test.”

This extremely rare variant form hearkens back to Forties Mencken—

My father put in a steam-heating plant toward the end of the eighties—the first ever seen in Hollins Street—, but such things were rare until well into the new century

and again to Proust: “Que nous l’aimons—comme en ce moment j’aimais Françoise—, l’intermédiaire bien intentionné qui” etc.

But Updike, our standard-bearer, never stands up for dashtards now. Even John Barth’s eighteenth-century pastiche, The Sot-Weed Factor, where they would have been right at home, doesn’t use them. (They were everywhere in the eighteenth century, too.) What comet or glacier made them die out? This may be the great literary question of our time. I timidly tried to use a semi-colash in a philosophical essay for The Atlantic Monthly in 1983: the associate editor made a strange whirring sound in her throat, denoting inconceivability, and I immediately backed down. Why, why are they gone? Was it—and one always gropes for the McLuhanesque explanation first—the increasing use of the typewriter for final drafts, whose arrangement of comma, colon, and semi-colon keys made a quick reach up to the hyphen key immediately after another punctuation mark physically awkward? Or was it—for one always gropes for the pseudo-scientific explanation just after McLuhan—the triumphant success of quantum mechanics? A comma is indisputably more of a quantum than a commash. Did the point-play of the Dadaists and E. E. Cummings, and the unpunctled last chapter of Ulysses, force a scramble for a simpler hegemony against which revolt could be measured?

The style manuals had been somewhat uncomfortable with hybrid punctuation all along—understandably so, since it interferes with systematization. The most influential Victorian antibarbarus, John Wilson’s A Treatise on English Punctuation, which went through something like thirty editions in England and America, tolerated mixed points; indeed, later editions offered pages of exercises, written and oral, intended to help the student refine his dexterity with the commash. But “the unnecessary profusion of straight lines,” Wilson warned, as others had warned before him, “particularly on a printed page, is offensive to good taste, is an index of the dasher’s profound ignorance of the art of punctuation …” In “Stops” or, How to Punctuate, Paul Allardyce, the Edwardian successor to Wilson, was more severe: “There is seldom any reason for the use of double points.” G. V. Carey, in Mind the Stop (1939), was unequivocal: “The combination of other stops with dashes is even less admissible than with brackets.” There was a glimmer of hope in Eric Partridge’s You Have a Point There (1953)—he advised that, yes, compound points should be used with “caution and moderation,” but he had the courage to admit that “occasionally [they] are, in fact, unavoidable.”

But that was 1953, in fault-tolerant England. According to the Chicago Manual of Style (§ 5.5), dash-hybrids are currently illegal in the US. In the name of biodiversity, however, I stuck a few of them in out-of-the-way places in my first novel, over the objections of the copy-editor, in 1988. I thought I was making history. But Salman Rushdie had beaten me to it, as it turned out: The Satanic Verses, which appeared a few months before my book, uses dozens and dozens of dashtards, and uses them aggressively, flauntingly, more in the tradition of Laurence Sterne than of Trollope. Brad Leithauser’s arch, Sebastian Knight–like frame-narrator, in Hence (1989), uses many commashes; and Leithauser discussed Rushdie’s “Emily Dickinsonian onslaught of dashes” in a New Yorker review that same year—although somehow Emily Dickinson doesn’t seem quite right. But we’re just playing at it now, the three of us—we aren’t sincere in our dashtardy;—we can’t be.

It would be nice to see Dr. Parkes or Dr. Lennard (of parenthetical fame) attempt a carefully researched sociohistorical explanation of the passing of mixed punctuation. Unfortunately, a full explanation would have to include everything—Gustav Stickley, Henry Ford, Herbert Read, Gertrude Stein, Norbert Weiner, Harold Geneen, James Watson, Saint Strunk, and especially The New Yorker’s Miss Eleanor Gould, whose faint, gray, normative pencil-point still floats above us all. And even then the real micro-structure of the shift would elude us. We should give daily thanks, in any case, to Malcolm Parkes, for offering us some sense of the flourishing coralline tidepools of punctuational pluralism that preceded our own purer, more consistent, more teachably codified, and perhaps more arid century.

This Issue

November 4, 1993