The medium variously known as comic books, graphic novels, bandes-dessinés, manga, manga, sequential art, and sequart has been defined as the juxtaposition of text and image on the static page. Once can here recall Words and Pictures, the name of a comics museum, or imageTexT, the title of one online comics journal. But image and text exist in tension to one another in various ways, and our understanding of these ways informs our understanding of the medium itself.

Division of Labor

In mainstream corporate comics, the separation between illustration and text is generally a formal one. While a writer describes the panels on the page, sometimes in as little as a sentence and sometimes in as much as multiple pages describing the composition and layout of panels with great precision, the lettering of the writer’s captions and dialogue onto the page is generally done not by the artist but by a letterer. This used to be done by hand, with letterers painstakingly trying to produce a typewriter-like consistency despite their different styles. Today, lettering (like coloring) is almost entirely done with computer programs after the artwork — still produced by hand — is scanned into a computer. While their credits are today printed in comics alongside all other contributors, letters, while certainly known by editors, are almost entirely unknown by the reading public. In a very real sense, text and image are separated through division of labor — and not only do different people perform the two tasks, but they do so using different media.

This produces the odd effect — which still happens often enough — in which word balloons are assigned to the wrong people. Needless to say, this can be very distracting. Sometimes, the effects are comic (by which I mean humorous, not the singular of “comics”): a man with a gun might scream “Help!” while chasing a potential victim. Even more frustrating is the fact that these errors are frequently not corrected in subsequent printings, including collection of comic books into trade paperbacks or even expensive hardcover editions.

Other times, the errors that result from this division of labor have been more serious. One issue of The Invisibles, Volume Three was printed with several pages having the wrong dialogue. It is easy to understand why the mistake was not caught: the series is notoriously difficult at times, and its writer Grant Morrison regarded as something of an enfant terrible. In that case, the trade paperback did correct the error.

It is worth pointing out that such errors are not limited to letterers. Most problematically, pages are sometimes sent in the wrong order to printers or printers somehow invert the pages’ order. I vividly remember reading an issue of Marvel’s Tales of G.I.Joe as a child that had this problem, and it wasn’t until I got to the earlier pages that had been printed later that I figured out why I had been confused: I simply assumed the error to be an artistic choice of some kind. More recently, Dark Horse’s collection of Matt Wagner’s first Grendel arc saw several pages printed twice — the second time, replacing a series of pages that should have been there. So in defense of this division of labor, such problems in production are not limited to the juxtaposition of text and lettering.

Some artists — particularly independent ones — prefer to letter their own panels. This gives them creative control over the full appearance of the page — unless, of course, they also have a colorist or an inker working over their work (though generally those who do their own lettering work in black and white and, if they use ink at all, ink their own work).

Artistic Unity

For some, another problem that arises out of this division of labor is an artistic one: the illustrations are intended to be idiosyncratic. This is perhaps understandable: text is generally judged by its consistency, not its artistic merit. The ideal lettering is generally thought to be lettering that is invisible: most readers want to read the comic without thinking of the process of lettering. One expects, on the other hand, individual artists to have their own style — derivative artists are often condemned on those grounds. In stark contrast to the lettering, the reader wants to be impressed by the images, to pause and take them in, to be impressed with the artistic panache displayed — although, it is important to note, generally not so much that one is distracted from the flow of the narrative.

There are, of course, exceptions to this “invisibility” of the lettering. The clearest example is when someone shouts or whispers, in which case the size of the lettering is frequently increased or decreased accordingly. Sometimes a shout may find a double border around the corresponding word balloon, as if to emphasize it or make it stand off the page.

But there are other reasons that lettering sometimes draws attention to itself. Sometimes, particularly with corporate super-heroes, characters names are replaced with their logos — a form of branding that distracts from the reading experience to celebrate the corporation and its beloved characters. The Sandman was notable for having its protagonist speak in inverted balloons, reflecting his depressed personality — while some other characters had their own typefonts that similarly expressed themselves. This has become increasingly mainstream: for a time, Marvel experimented with this, having Iron Man speak in a mechanical font with its own, gear-like border and Thor speak in a font reminiscent of Norse runes. Some comics have personalized captions, serving as shorthand to let the reader know whose language or internal monologue is being expressed: the character’s symbol or face might appear beside the caption and / or that caption might be given its background color or colored shadow.

Most of the artists who letter their own artwork prefer the same sort of invisible style preferred by mainstream comics. A few others, however, have preferred a style as individualistic as their artwork. In this, Eddie Campbell is the most noteworthy: he has employed this style in his own Bacchus and Alec series, as well as his collaborations with Alan Moore (most notably, From Hell). The result is mixed: Campbell’s script can be hard to read, creating confusion, but sometimes its idiosyncrasy produces an enjoyable effect. One might argue that such idiosyncratic lettering is best suited for autobiographical work or accompanying an artist with a particularly idiosyncratic style.

Text Other than Word Balloons and Captions

On the matter of text other than word balloons and captions, other forms exist. Traditionally, the title and credits of a graphic story, whether short or long, is given within the story itself. This allows for a sort of organic whole, allowing a prologue or sequence before the title is given much as a film might. The title is sometimes drawn by the artist and sometimes lettered by the letterer, and the level of graphical interaction between the title and the image is usually dependent either upon the artist’s preference or corporate style. Will Eisner famously incorporated titles into the page itself, particularly in The Spirit: a title might be composed of skyscrapers or of papers adrift on water. It may be worth noting that several series have attempted this with the titles on their covers: in mainstream comics, both Planetary and Promethea have radically changed their covers’ title from issue to issue — though the practice has long been sought to distract from repeat buyers’ recognition. I can say from experience that it sometimes does, although the practice so enhances the cover that it is hard to argue against, speaking artistically.

Since the 1990s, however, there has been a trend to divorce the title from the contents.  Image Comics, in particular, began with the house style of placing the title and credits on the inside front cover. Some comics — including Marvel’s The Ultimates — now try a hybrid, using a page with a horizontal bar of an image (typically the first panel of the story) on an otherwise black page. Others have started to give a good deal of space to the title within the narrative proper — creating a space along the bottom of two pages, for example.

It is worth noting that word balloons and captions are sometimes given footnotes. This was long a common practice in mainstream comics, although it has become rarer. Typically, a character or caption would reference the events of a previous issue and a footnote would cite that issue, frequently followed by the initials of the editor.

These footnotes, however, have great artistic implications, however, in terms of their placement. As Will Eisner has argued, the page itself represents a sort of meta-panel onto which various forms of panels are placed. Does the footnote then appear at the bottom of the panel or the bottom of the page? The answer, for most corporate comics, is the bottom of the panel — although occasionally the bottom of the page has been used. This should not surprise, as most corporate comics do little with the notion of the page as meta-panel: one cannot readily imagine, were Will Eisner to employ captions, that a page in which, say, water flowing out of sewer pipes formed the panel borders would have a footnote within the panel.

Another textual element besides captions and word balloons is the indicia. This is the legally-mandated information about the publisher and the official title and number of the particular issue, exactly as magazines do. Because ad space was at a premium in mainstream comics, this generally horizontal bar of tiny print was placed within the comic narrative itself, generally at the bottom of the title page. This (and next issue blurbs) has led, when issues are collected in trade paperbacks, to some odd horizontal black spaces over the artwork. One advantage of the title page, or of an increased space for the title, is that the indicia can be placed along with the title, outside of the narrative proper. Some comics now place the indicia on a letter’s page or with the traditional corporate note to readers — which would seem a rather obvious choice.

Battling for Space:  Text and Image as Layered

One of the most inherent tensions between text and image is the fact that they are typically layered. That is to say, the text is placed over the artwork. If an artist is using a letter, this means that the artist must consciously leave room for word balloons and captions.

Probably the most common problem associated with the division of labor between artist and letterer is the printing of word balloons or captions over important parts of the artwork. This is reportedly rife at the major corporate publishers, and I have heard creators employed by them complain of it terribly. While word balloons and captions are rarely placed, say, over a character’s head, letterers sometimes do not seem to understand what portion of the artwork is most important. In a landscape shot with a castle on one side and ominous trees on the other, one sometimes has to obscure one of these elements at least partially, and the letterer may not make the most logical choice for the story.

Some solve the inherent special tension between text and image by breaking the formula of captions and word balloons, incorporating text and image more directly. The celebrated but little known artist Barron Storey generally uses neither captions nor word balloons, instead incorporating the text in other ways, such as on scrolls flowing from or around characters in a manner recalling Medieval illuminated books. The most famous example, however, is probably David Mack (of Kabuki fame), who frequently incorporates the text into the artwork itself, sometimes as a panel border but often radiating out from a character or encircling that character in various ways. Sometimes, Mack’s text is clearly not word balloons or traditional captions at all, but a sort of emotional expression of a character’s state of mind — sometimes accompanied by childlike drawings communicating the same. While all of this sounds brilliant, it often fails in terms of the narrative: a single image illustrated this way is delightful, but a 100-page narrative is heavy going. One solution would be to use this technique sparingly, such as for emotional high points within a narrative.

The independent, little-known comic book Mercedes notably solved this problem by using no word balloons, instead incorporating dialogue into captions running below the panel itself. While a pleasant idea, given that it prevents obscuring the artwork, it did not read well. Text and image were even more divorced than before, and the eye had to move away from the image to read the accompanying text.

It is worth noting that there has been a trend recently towards using images in balloons, particularly thought balloons. For example, a character might imagine strangling another character, and this image could appear within a thought balloon within a larger image.  Indeed, thought balloons have fallen out of fashion, outside of this device. DC’s Impulse, starring a young and impulsive speedster, employed this device often. But for our purposes, this is irrelevant: such thought balloons are not, after all, text — merely a layering of image.

Primacy

This leads neatly into the issue of primacy. Of text or image, which is more important? The answer, naturally, has great implications. Do we refer to comics as “texts”? If comics don’t have their own departments at universities, do we study them in literature departments or art history departments? Is comics a literary art or a visual one? After all, we don’t have the necessary third term combining the two — and while one can easily say “both,” it’s difficult to remember to call comics as litero-visual art or a visuo-literary art, or some other such configuration.

Opinions, naturally, differ. Many would argue for text. After all, the text traditionally carries forward the comics narrative. In other words, we usually would not know what was going on without captions and word balloons as indications. To some extent, writers tend to correspond to this camp. The great debate for the other side, with which many artists agree, is the existence of silent comics. It is precisely for this reason that, in his seminal Understanding Comics, Scott McCloud has defined comics as “juxtaposed pictorial and other images in deliberate sequence” (9) — omitting text.

To some extent, the answer depends less on the intrinsic qualities of the medium than the work under discussion. American and European comics tend to be word-heavy, while Japanese comics can sometimes astound Americans by taking, for example, seven pages to show a character walk down a hallway. Perhaps, then, we should avoid generalizing about the entire medium in this regard.

On the other hand, one experiences illustration and text on different levels. Text, while it may well have began as pictograms, is in most languages utterly abstract. The connection between the word “dog” and a dog, as the structuralists have reminded us, is utterly arbitrary. Illustration, however, tends to be representational: we apprehend it more immediately, more directly. In comics, this can produce problems. One generally “reads” the image quicker than the dialogue or captions, and can produce the odd effect in which one forgets the image one has just “read” as one reads its accompanying caption. Additionally, one is far more likely to misread text than an illustration — especially given the inconsistent and sometimes idiosyncratic lettering style. One might wonder what is being depicted in a certain panel, but textual misreading — or a misprinted word — sometimes creates odd misunderstandings between the work and its audience. Perhaps we may try to be satisfied with the fact that they are merely different: text superior in certain ways, in certain texts; image in others.

Narrative Possibilities

We shall end by considering some possibilities in which text and image may combine in unique ways. Alan Moore has described how such possibilities ought to be one of the unique advantages of the medium. His Watchmen is notable, among other things, for precisely this. This sort of ironic juxtaposition takes two forms: complementary and contradictory.

In the complementary formulation, the text and the image are in accord — albeit in a potentially surprising way. For example, a character seen in the previous panel could be remarking in a caption about how something mundane has been blown apart while the illustration under the caption shows someone being shot with a shotgun. The text and image are in accord, yet the manner of accordance surprises us.

Secondly, the text and image may disagree. The simplest example might be a caption in which someone, commenting on a runaway, says something like “I’m sure she’s fine.” The image under this text might be the same as above: someone being shot. In this case, rather than eschewing the tensions between text and image, the narrative embraces them.

However, we have not exhausted the possibilities. Experimental comics are still being made, and some continue to surprise. The most recent example I can think is “Templates,” appearing in Elizabeth Jacobson’s independent comic Misc., published out of Kansas City. Instead of using word balloons to communicate dialogue, it uses them to communicate the underlying emotion. A word balloon might thus read “uncomfortable small talk,” answered by another word balloon saying “automated response.” While still classically juxtaposing text and image through the device of word balloon, this creates a new way of reading their relationship: the image takes on the literal qualities, carrying forward the narrative, while the text that usually communicates what’s going on takes on the role of emotional and impressionistic commentary upon a narrative that feels silent in a way comics usually don’t — a narrative suddenly both distant and close.

The medium, after all, is still evolving.