Sunday, April 07, 2013

Diagramming Illustrated Sentences

This "Figures, Story & Staging" project we're working on in WORD & IMAGE has got me thinking about a particular kind of narrative image that might be described as an "illustrated sentence." In these images there are clear characters (nouns), specific actions (verbs), helpful props (direct and indirect objects) and modifiers (adjectives, adverbs) that go with each.

We deliver a story prompt to each student, something like "Jewel Heist!" or "Family Canoe Trip", along with a set of constraints to create 3 illustrated sentences that will tell that story: multiple figures interacting, no setting save for that suggested by objects and costume, and use of a 2-color printing palette. These three illustrations by Jim Flora, excerpted from a 1954 alphabet book created for employees of CBS-TV, are perfect examples of these stipulations.
What becomes clear when developing "readable" pictures like these is how important the 2-D design is: how line and shape interact, how contrast and color create hierarchy, how negative space functions, etc. These 2-D concerns -- especially when backgrounds are removed from the equation -- may lead these kind of images to 'flatten' out into shallow graphic space, like this illustration I did a few years ago (prompt: "80's Dance Party!"):
That drawing, because of its pattern of interlocking figures, claustrophobic composition and vector style, functions a bit like the guts of a (day-glow 80's) swiss watch.  The same kind of constraints (full figure ensembles, minimal backgrounds) could lead to totally different images, like these classic drawings from Love and Rockets: The Death of Speedy by Jaime Hernandez:
The space is clearly deeper, implied mostly by scale shifts between characters. But the 2-D design of the panels is still doing most of the work: the functions of shape vs. line vs. "gray" texture, the heavy spot blacks, the angles found within the composition. Note the general vertical-ness of all the characters of all the characters in the drawing above broken by the chill-axed lean of Speedy himself. Or the contrast between the guitarists at this punk show:
Sticking with black 'n' white slice of life comics but going back a ways, here are some 1917 strips from The Gumps by Sidney Smith, reprinted in The Smithsonian Collection of Newspaper Comics, (Bill Blackbeard & Martin Williams, ed.). I like these strips because they have such simple communication goals: The Gumps try and move in. The Gumps try to fix a pipe. The Gumps try to hang a picture. Even though the poor Gumps fail at each domestic task (sort of a bizarro version of the happy household chores DB Dowd talks about here) the strips themselves achieve a schematic-like level of clarity.
The design of the full-bodied characters, their frontal staging, the repeated compositions in sequence, minimal background information, and the left-to-right axis of action all contribute to the momentum and readability of these slapstick illustrated sentences. Being part of a pre-cinematic tradition, comics from this era lean heavily on what might be called traditional theatrical modes of presentation. Here's a modern comic that popped into my mind as I was thinking about this stuff, Wally Gropius by Tim Hensley. Prompt: "Garage Band Practice."
Working in a mid-century style reminiscent of John Stanley or Dan DeCarlo (a favorite of Jaime Hernandez as well -- get a load of his 2-color gag cartoons for Men's Magazines), Hensley draws the most obvious inspiration for the style of Wally Gropius -- theatrical staging, rubbery character design, zany gags, no backgrounds --  are early Harvey Comics like Ri¢hie Ri¢h:

Looking at the construction of the jokes in these covers, and thinking about them as "illustrated sentences", I couldn't help but want to diagram them the way you would in 8th grade grammar class. If we think of that first cover as the sentence: Richie Rich floats in his comically fancy bathtub while his butler Cadbury patiently holds a giant toothbrush. It might be diagrammed like this (I think... It's been a long time since 8th grade):
It seems like an interesting experiment to try and diagram the images visually in a similar fashion:
I've focused on the major nouns (subjects and objects) of each sentence, and the axes of action that their associated verbs (predicates) take place along. I tried a few more diagrams on illustrations I've done, all of which fall generally in the same shallow space, theatrical modes. Since I've got more background info on the development of these images, I can look a bit further back into the process.

IDEA SKETCHES. These are the sketches that sold the basic premise of each. "Baby New Year cheats off Old Man Time", etc. You can see that the characters and basic idea is in place, but the design is fairly loose.
DESIGN THUMBNAILS. Even though these seem 'rougher', you can see that I'm using these to tighten up the composition:
DIAGRAMMED IMAGES. Note how I made most of these decisions during the Design thumbnails stage.
These diagrams of "illustrated sentences" each create a simple map of a constructed pictorial space. They show, at a glance at least, basic composition and hierarchy much like a diagrammed text sentence does. I'm not entirely sure of the value of port-mortem autopsies like this, but certainly considering the schematic structure -- or syntax -- of an image during early stages will provide solid footing.


tsmoreau said...

Hey Dan, love your comics, just wanted to drop my two cents in on the topic here.

You touch on a great number of points in the post and the format of the comment section here doesn't really afford me the space to give them the responses they deserve, so my apologies about brevity and lack of clarity.

As to the value of diagramming the image which you touch on in your last paragraph, I think there's probably little in the mechanistic-creation sense. Especially for the seasoned creator. I do think there's still some value in it to be sure, but more in a descriptive way than a prescriptive one.

It's simply a visual description of a reading of the image by actively highlighting and separating elements of how it operates (nouns, verbs, etc, in the nomenclature presented here). The true value comes in a connecting those highlighted and separated elements to style, theme, tool use, the continued evolution of the creator, etc, etc. It's just a helpful tool in visual criticism and one that can be given it's own relatively low level of coding and grammar should one be inclined.

I notice you didn't annotate the Gumps page (or any sequences at all, actually) and I'd be quite interested to see your take on that. I'm interested mainly due to your overriding "illustrated sentence" approach in the post and in that structure the Gumps page is more like a full paragraph or so, with complex syntax and multiple subjects and objects. In particular, I'd be interested to see some color coding and greater icon-adherence in your circling, so as to emphasize body language and individual grammatical elements. Perhaps I'm simply pushing my own approach here, but it seemed to me the logical outgrowth of this sort of "hierarchy filter" or whatever.

In another way I think there may be a bit of a clash of values happening, too. I don't think the nested yet linear nature of linguistic reading comprehension shares enough structure with the networked compositional nature of image comprehension for the analogy to hold across the entire breadth of the medium, and I'd be interested to see how far one could take it before direct one-to-one concepts stop being applicable.

Similarly, you don't mention panel structure or layout or sequence at all as relates to parts of speech. Nor is there a mention of drawing style as, perhaps in this structure, tone of voice or speaking style. I don't think the bounds of the traditional sentence diagrams can be stretched far enough to encompass all the facets of the medium. Like I say, I'd be interested to see how the things would have to be manipulated to make one-to-one concepts work in as many places as possible.

It'd probably be an interesting experiment, but I don't think trying to stretch language modes into image modes is going to be much of a rewarding approach. In my opinion the visual annotations are less about complex grammar and more about simple highlighting of visual through-lines and composition and whatnot.

Of course all this would be relatively stayed across the Gumps page. Considering, as you highlight, it's middle-ground theatrical nature. I'd be interested to see your take on not only some more cinematic material, but some more abstract material as well. I feel some diagramming might be helpful in a critical presentation of material like that, or even, honestly, any of the material you present in the post.

I think the further one gets from the traditional, relatively-flat, middle-ground line-cartooning and more moves towards cinema, expression, abstraction, or whatever, mixing the grammar of the camera, the canvas, and the word, the more applicable the diagrams then become.

Though all this may simply stem from my own critical inabilities, of course.

It's late, but I just wanted to say thank you for taking the time to do the blog post and not only explore your own perspective on some of this stuff, but also offer myself and others a chance to examine that perspective along with you!

Dan Z. said...

Thanks for the thoughtful comment, ts moreau!

You're right that I only tackled diagramming a very specific kind of spatially flat, 'schematic' (single panel) illustration. That's mostly for the benefit of my students and the particular 2-D design oriented project we're working on at the moment, not for any lack of interest in diagramming more complex sequences and kinds of spaces, especially in comics. I'd love to try some of that out!

Thanks again. Looking forward to digging into

Dan Z. said...

And yeah, I totally agree that those Gump strips are more like "illustrated stories" with each panel being a paragraph, and each character performing a "sentence"? Not to mention the fact that there are actual (text) sentences in there too!

Eden Lewis said...

Reading this post (especially the diagramming bits) and seeing your demo in class have given me a lot to think about in terms of compositional structure. In particular, considering eyeline direction and implied lines of motion (like your basketball's arc) seem like useful ways to think about structuring a composition that I have been neglecting. I also LOVE the way the red shapes function in the Flora illustrations and the complex way that black is used to add line, volume (shading?), and texture to the images.

Sophia Brown said...

I was pleased to see the rough thumbnails of your design process for the RTF/Observer covers because it is really helpful (and comforting, as a budding artist) to see how idea and design do not all come at the same time. Another thing in your post I really responded to and am happy to have pointed out is Hernandez's method of distinguishing characters in his comics through posture -- in the Death of Speedy in particular, it is so true that the reason Speedy stands out from the other characters is because of that diagonal! Those pointers are really going to help me with my current project, since I've really been trying to milk the expressive value of silhouettes. Seeing a breakdown of the design of comics like this is a great guide!

Wyeisha said...

There is an amazingly wide range of illustration in these posts. As an aspiring artist, it's both motivating and a bit intimidating to see the variety of ways of image making just within illustration alone. The vague and very suggestive "stages" that these characters are set in have been helpful visual resources for my current project. The diagrams and sketches have also proven to be very informative references for figuring out the skeletons and postures of my figures.