Winsor McCay’s Comics are Formal Innovations — and also really racist

February 12, 2024 Winsor McCay’s Comics are Formal Innovations — and also really racist

First the formalist stuff.

Bloomsbury published The Comics Form in hardcover last year and paperback last month. I hope to publish a revised edition someday, but until then, this blog is a good place to muse about revisions. Though my approach in The Comics Form is formalist rather than historicist, I worked to acknowledge scholars and artists who originate certain lines of thinking or approaches, harmonizing their ideas and terms with those that have followed. Of the many many folks I left out, Winsor McCay now feels like a significant omission.

Here’s why.

I wrote in Chapter 5 about various kinds of “juxtapostional inferences”:

“Continuous inferences provide an answer to Mikkonen’s question: ‘when can a group of images be perceived as one image?’ (2017: 12). In contrast to embedded inferences creating the perception of a single image as multiple images, continuous inferences create the perception of multiple images as one image.”

That’s because:

“Discursively, continuous inferencing produces an impression of a visual element partly obscured by a visual ellipsis. Though two lines are separated by an undrawn space, a viewer perceives them as a single line.”

What should you call that?

“This effect aligns with ‘closure’ as defined in Gestalt psychology, but because McCloud uses ‘closure’ to describe all juxtapositional relationships, using ‘closure inferences’ as a type of juxtapositional inference, while accurate, would further conflate terms. I have previously suggested omitting capitalization and using ‘gestalt’ (Gavaler and Beavers 2018: 20), but ‘continuous’ seems less likely to produce further confusion, in part because its meaning is self-evident.”

I go on to describe examples by Jessica Abel and Brecht Evens. But I should have started a century earlier with McCay. Here’s the Sunday, January 27, 1907 edition of McCay’s Little Nemo in Slumberland:

Look at panels 4, 5, and 6 across the bottom row. The domed ceiling, pillars, and circular staircase extend across all three panels, while the four recurrent figures repeat.

That perfectly illustrates my description in The Comics Form:

“When a visual field is partially subdivided, with some elements appearing to continue between images and some not, attention to the continuous elements will create the effect of a single, subdivided image, while attention to the discontinuous elements will create the effect of separate images juxtaposed. Both may be present simultaneously, and visual elements may be categorized by figure and ground or other dividing principles.”

In McCay’s case, the ground is spatially continuous across the two panel divisions, while the figures instead require temporal leaps. I don’t know if this is the first continuous inference in the history of the comics medium, but it’s the first I’ve encountered. I especially like the character Icicle’s comment in the middle panel. It reads to me as a quiet metafictional acknowledgment of the visual effect of stationary images appearing to traverse a background:

“Well, you don’t have to go any further if you don’t want to!”

I’m not tempted to rename the inference “Icicle Closure,” but I do wish I’d included McCay’s sequence in The Comics Form.

And now I need to pause.

On May 5, 1907, less than four months after the above Sunday edition, McCay introduced “Jungle Imp” to his series. The character is an example of the virulently racist blackface minstrel tradition that was common in comics in the first half of the twentieth century. Will Eisner’s Ebony White in The Spirit and Charles Nicholas Wojtkoski’s Whitewash Jones in Young Allies are two late examples from the 1940s. When I wrote about them in my forthcoming essay “Reading Race in the Comics Medium,” I included descriptions but not illustrations. To continue my analysis of McCay here, I need to include images of his “Imp.” I have conflicted opinions about that. If you believe that formal comics analysis is not a sufficient reason for sharing racist imagery, I have no counter-argument and encourage you to stop reading. If you believe that formal comics analysis can be a sufficient reason to share some of McCay’s 1907 racist images, then I welcome you to continue. Either way, I’ll reexamine the question at the bottom of the post.

McCay published this edition of Little Nemo in Slumberland on November 24, 1907:

The top two-panel row is another example of the continuous panel effect. If it were merely a repetition though, I wouldn’t share it. But the second and third rows do something further.

Both trigger continuous inferences, but where the four figures in the earlier example appear to be moving across a continuous space, these three figures appear to be stationary. The sequence appears to portray them incrementally sinking down to their hands and knees. But for the narrative to make sense, they must also travel further down the hallway to the next door at the center of each panel. Because McCay draws each door identically from the same angle (relative to each door), changing only the words in the signage, the continuous effect is disguised. Panel 4, 5, and 6 look like spatial repetitions of 3.

McCay created a similar effect (this time with no racist imagery) on March 22, 1908:

Based on the backgrounds, each of the four rows requires a continuous inference to bridge the middle gutter. And yet McCay draws Nemo’s seemingly stationary figure near the center of each panel, creating a visual impression that contradicts the narrative fact that he must have begun moving and then stopped moving again during the inferred temporal gaps.

Returning to the earlier adventure with the Imp, McCay included a related but distinct effect on November 10, 1907:

As with the columns in the November 24 edition, the column at the center of the page is divided in half by the gutter. In this case though, the column isn’t one column but the halves of two different columns framed to look like a single unified object.

I describe this in The Comics Form as a “semi-continuous” inference:

“The continuous impression of an interrupted line is a discursive effect. When the line is representational, continuous inferences create the appearance of interrupted diegetic content, but the interruption may be understood at only the discursive level if the diegetic content of the two images is not continuous. Semi-continuous inferences create discursive connections between discontinuous diegetic content.”

And that’s because:

“If the trajectory of a line within one image visually aligns with the trajectory of a line in a contiguous image, a viewer may experience the two lines as a single line – even though representationally the two lines are unrelated. At the discursive level, continuous and semi-continuous inferences are indistinguishable, but semi-continuous create discursive shapes across images that appear to share diegetic qualities even though the images do not exist in a story world together. Semi-continuous inferences can produce a unifying effect at the diegetic level.”

I go on to describe examples by Kevin C. Pyle and Charles Burns — but, once again, McCay was a century ahead of them. He does it again on December 1, 1907:

The apparent middle column at the center of the first three rows is again the halves of two different columns — which the final page-width panel makes explicit. The page resembles the continuous panels of November 24, but is instead semi-continuous.

This page also features an additional meta effect. While the words “Banquet Hall” on the door in the background remain consistent, the words of the strip’s title change as the characters knock them down and then eat them. In a 2014 essay, Roy T. Cook terms the metafictional effect “objectified”: a thought or speech balloon is objectified “when it is placed in such a way as to force the reader to interpret the balloon as part of the physical universe inhabited by the characters and objects depicted in a particular panel.” Like me, Cook cites two examples published a hundred years after McCay’s.

The effect is especially paradoxical because the title is continuous across the middle gutters. The words “Little Nemo” appear only in the left panels, and “Slumberland” only in the right. That fact is normally unremarkable because the words normally aren’t part of the storyword but exist only discursively as part of the layout — what I call a “secondary diegesis” in The Comics Form. The characters don’t seem to be aware that they are drawn characters in a comic strip observed by viewers. It just seems there’s paradoxical overlap of their world and the ambiguous world in which the title letters exist.

Now back to the more important question of racist imagery.

Three of the five above comics include blackface caricature. One of those three repeats the relevant formal qualities (continuous panels) of the non-racist strips; the other two display unique qualities (semi-continuous panels and objectified title letters) that are not otherwise illustrated. Does that justify including them here?

I suspect that if I actually were working on a new edition of The Comics Form, I wouldn’t include them. I know that I don’t use comics with blackface caricatures in my classes. It seems I hold a lower standard for my blog — a space I often use for initial drafts of work I later revise and publish in academic venues. And yet I know of at least once when I was careful not to share similarly racist images: during a “cancel culture” uproar over some Dr. Seuss books no longer being available in print. There the racist imagery was central. Here formal analysis is, and the racist imagery is peripheral to that analysis. Is that an adequate argument for including it? I suspect I would challenge the rationale if someone else made it, making me at best inconsistent.

I considered obscuring the images of McCay’s Imp, but decided against that. I also considered rewriting this post and avoiding the three racist comics entirely — but now I’m finding this question more interesting than the formal analysis that initiated it. I’m still debating:

What is the best way for comics scholars to treat innovative comics with racist imagery?

Tags: Jungle Imp, Little Nemo in Slumberland, racist, Stan Lee, Windsor McCay

• Leave a comment
• Posted under Uncategorized