Early in his career, René Magritte devised and published a rulebook for the rest of his life's work, "Les mots et les images" (La Révolution surréaliste, no. 12, 15 décembre 1929; in Magritte 1979: 6063). It's a Dada rulebook, as hard as that concept might be to embrace. But the piece has been republished widely throughout the world of art and literature since the 1930s, and has become "canonical" (though one should note its first entry).
Dada knew comics well (very well indeed; see Inge 1990: 4158), and Magritte's rules are hardly absurd or even irrational viewed through the same frames as newspaper comics of the period, most notably Krazy Kat. The correspondence between word and image might be literal (or very close) in many mainstream comics, but the links become increasingly Dada and surreal closer to the margins. It is tempting to speculate that Magritte invented none of these guidelines, but merely observed them.
Here, for once, the need to translate the text from French to English offers an opportunity to "enhance" Magritte's concepts. To read the rules in text form, and then together with the pictures, and to compare mental notes, is to appreciate (but not quite understand) the rich potential of the comics form. After that, all the rest is merely technique.
Words and images
An object is not so attached to its name that another cannot be found for it that suits it better: [Le canon = gun]
Some objects can do without a name:
Sometimes a word serves only to designate itself: [ciel = sky]
An object encounters its image, an object encounters its name. It may be that the image and the name of the object encounter each other: [forêt = forest]
Sometimes the name of an object does instead of an image: [canon = gun]
A word may take the place of an object in reality: [le soleil = sun]
An image may take the place of a word in a proposition: [Le ... est caché par les nuages = The ... is hidden by the clouds]
An object implies that there are other objects behind it:
Everything tends to suggest that there is little connection between an object and that which represents it: [l'objet réel = real object; l'objet représenté = represented object]
The words that serve to designate two different objects do not indicate what may separate those objects from each other: [personnage perdant le mémoir = person suffering memory loss; corps de femme = a woman's body]
In a painting, words are of the same substance as images:
We see images and words differently in a painting: [montagne = mountain]
Any shape may replace the image of an object: [le soleil = sun]
An object never performs the same function as its name or its image: [cheval = horse]
Now, the visible outlines of objects in reality touch one another as if they formed a mosaic:
Undefined figures possess as necessary and as perfect a meaning as precise ones:
Sometimes the names inscribed in a painting designate precise things and the images undefined things: [canon = gun]
or the other way around: [brouillard = fog]
(translated in Meuris 1990)
George Herriman, drawing for Krazy Kat, undated, but the same period as Magritte's Les mots et les images.
McCloud (1993: 152155) devises more practical conventions of word and image, but these retain some of their surreality in comparison to literary or even journalistic approaches. In 2000, we are almost all at least familiar with the comics if not expert, and the daily juxtapositions of blaring headlines, sensational stories, and lifestyle advertisements in close quarters has probably inured us to the shock value of words and images piled atop one another.
Scott McCloud, Understanding Comics, 1993, pp 152-155.
The most common experience of these juxtapositions from the beginning of printing to the rise of the illustrated newspaper are narrated illustrations or captioned boxes.
Action Comics, 1937 (the first Superman)
The converse of the captioned box is the rebus, or the narrative text punctuated with illustrations, whether with frame or without.
Philippe Caza, Le Métro
Chris Ware,Acme Novelty Library
The enthusiasm and inventiveness of the letterer is often encouraged to take over on splash pages or advertisements in comic books, often in response to the magazine and street advertising styles of the day.
But there are times when the letterer is absent, and the potency of the styled letter is surrendered to the deadening convenience of the lettering template.
Many recent comics use computerized typefaces that emulate as best they can the comics tradition -- often all uppercase, with extravagantly italicized and emboldened emphases. But there is also the tendency to straight typography. For instance, Mad Magazine's satirical edge is in its content for sure, but that is reinforced graphically by making the look of the words and the look of the pictures very distinct and unlike "typical" comics.
But then again, all elements of Robert Crumb's satirical pictures and words are from a single pen.