Skip to content

The Patron Saint of Superheroes

Chris Gavaler Explores the Multiverse of Comics, Pop Culture, and Politics

Category Archives: Uncategorized

Cover image

Though the Western is the only genre uniquely set in North America, it has diminished in U.S. fiction. Only a few American films and novels of recent decades depict the late 19th century period of colonial expansion across the so-called frontier. But the West—or at least a fantastical version of it—continues to expand in France, where Laos-born writer Loo Hui Phang and acclaimed bande dessinée artist Frederik Peeters published The Smell of Starving Boys in 2016. London’s Self Made Hero released the first English version earlier this year.

Phang imagines a privately funded propaganda campaign into the “Comancharia” for the purpose of photographing the landscape and attracting settlers who will then mine its mineral riches. But one member of the expedition adds, “I’d have loved to believe in something impossible,” a desire seemingly shared by the authors. Though the post-Civil War setting is nominally historical, Phang’s characters quickly shift from the unusual to the fantastical: a gay photographer whose fake spiritualist photos turn real; a cross-dressing teen who speaks the secret language of horses; a cadaverous-looking bounty hunter who sucks horses dry of blood. Even the pot-bellied, colonial capitalist stops wearing pants and envisions founding a utopian civilization free of women and sexuality. A tribe of Comanche populates the periphery of the plot, with an unnamed old man following the explorers and silently delivering tokens of apparent magic: a double-exposed horse head on a photographic portrait plate, a severed horse head that serves as a soul-summoning bull-horn to herds of stampeding horses.

Oscar, who has fled sexual scandal in Manhattan, wears an aristocratic boy’s portrait inside his locket. Peeters renders Milton, the party’s farmhand servant, with nearly identical features, substituting a white wig for cropped blonde curls. Milton, however, is no boy, but a virginal young woman named Weather fleeing a father trying to sell her into marriage. Oscar, apparently transformed by her boyish looks and the whimsical west, falls in love. The two have sex for the first time while her father’s zombie-like bounty hunter searches outside their secret cavern chalk-marked with magic runes by the old man.

Peeters renders Phang’s world in a consistent, loosely naturalistic style, with the occasional cartoonish feature reserved for the racist capitalist, the caricature of Phang’s social critique. Peeters’ layouts are similarly consistent and feature a four-row base with one to three panels per row. His vertical gutters occasionally align but typically avoid rigid grids in favor of shifting patterns, including a dozen sub-columns in which the novel’s reading path briefly moves top-to-bottom rather than left-to-right. When the effect is achieved through combined panels, the image content tends to be panoramic, with expansive vertical views of the western landscapes. When Peeters divides a panel into thinner, horizontal strips, the content is accordingly cropped, usually of facial close-ups reduced to eyes or mouth. Panels are almost invariably rectangular, with the exceptions appearing near the novel’s end and coinciding with moments of violence: an arrow through a neck, a bullet into a gut. Peeters abandons his base layout for the four-page climax, including the novel’s only two-page spread and full-page bleeds. The striking visual shift evokes the simultaneous splintering of the story’s reality as physical and seemingly spiritual worlds collide. When Peeters reprises the base layout for the epilogue, the world is stable again but permanently altered as Weather and Oscar ride off into their ambiguous afterlife together.

The contributions of Edward Gauvin, the French translator, are even more ambiguous. It’s unknowable whether Phang’s original French included the equivalent of the clichés “in the blink of an eye” or “There’s the welcoming committee.” Since such phrases cluster in the dialogue of the expedition leader, the most two-dimensional and satirically broad character, the effect mostly works. But when Oscar’s speech includes the novel’s title, the intent is less clear. While giving oral sex to Weather, Oscar reminisces about the sexual delights of boys: “if you only knew the smell of starving boys … I’ve never felt that way about girls … But you. Goddamn. You awaken it.” Oscar presumably means “starving” to be erotic, perhaps intending something closer to “skinny” or even “emaciated,” but “starving” has a more distressing connotation. If he actually finds starving boys arousing, then he’s turned on by a sex partner’s desperation for food, which then implies domination and coercion. Though I doubt that was Phang’s intended meaning, it’s equally difficult to believe she and Gauvin would not have worked through the semantic nuances of the novel’s title.

Image result for the smell of starving boys

I also wonder about the amount of nudity of the novel’s only female character. Phang and Peeters find multiple opportunities to strip Weather of her clothes: a bath in a stream, a sex scene, an order at gun point, another sex scene. While Oscar is sometimes naked, and Peeters draws the capitalist’s mushroom of a penis more than once too, the eye of the novel roams longest on Weather’s adolescent body, including on the cover. The Western frontier, as romanticized by the French authors, is largely a sexual frontier, a border between genders and sexualities—what the story’s moralistic villain would eliminate. While it’s difficult not to root for the romance, it’s unclear why a female body, even a female body briefly understood to be a male body, should receive greater attention. It’s also unclear why the lone and unnamed native character should care about let alone magically aid Oscar and Weather’s romance, especially given the threat that their expedition poses. As her name implies, is Weather close to nature and so, as the cliché goes, like an Indian?

Despite these questions, the graphic novel’s mix of fantasy and Western tropes is involving, and Peeters renders Phang’s story with subtle craft. Phang also emphasizes the nature of image-making from the first panel:  an upside landscape as viewed through the inverting lens of Oscar’s camera.  The authors repeat the motif twice more, including on the novel’s final page, reminding readers that The Smell of Starving Boys is itself a set of inverted images too, ones that alter as much as reveal their subject matter.

Image result for the smell of starving boys

[A version of this post and my other recent comics reviews appear in the comics section of PopMatters.]



Tags: ,

Sandra Chevrier 01

What if an evil genius is tricking you into believing that the world around you is real when it really isn’t?

What if on an alternate Earth everything is identical but for one almost undetectable detail?

What if trying to travel to the past transported you to a different universe instead?

What if a mad scientist removed your brain and is keeping it alive in a vat of nutrients?

What if lightning struck a dead tree in a swamp and transformed it into The Swampman?

Any of these fantastical plots could be the premise of a superhero comic book. Stan Lee sometimes gave artists at Marvel little more to work with—just a note on a piece of paper or a plot point mentioned on the way to his desk. Jack Kirby or Steve Ditko would work out the details.

Except none of those scenarios comes from comics. They’re all thought experiments written by highly regarded philosophers: René Descartes (1641), Hilary Putnam (1973), David Lewis (1976), Hilary Putnam (1981), and Donald Davidson (1986), respectively. Like comics, fantastical tales are a staple of philosophy. Philosopher Peg Tittle includes 126 in her 2005 What If … Collected Thought Experiments in Philosophy. But superhero comics were well ahead of her. Marvel published its first What If? in 1977, and DC published a range of “Imaginary Tales” in the 1950s and included two “Just Suppose” tales in one of its first 1936 titles.

Philosophers could fill volumes too. David Chalmers writes about zombies, Laurence BonJour about clairvoyants, and Frank Jackson about a scientifically all-knowing woman who’s never seen color. The list of “What If?”s seems endless:

What if your body slowly transformed into rock, but no one around you noticed?

What if a god were stripped of his memories and forced to live as a crippled human?

What if a time traveler returned to his childhood and told his past self about the future?

What if you could save the world but had to sacrifice millions of people first?

What if you and all the universe were just the thoughts of a small child?

Except those scenarios don’t come from academic philosophy. They’re all from superhero comics: The Fantastic Four (1961), The Mighty Thor (1968), The Defenders (1975), Watchmen (1987), and Heroes Reborn: The Return (1997). And they are no more fantastical than scenarios philosophers have been dreaming up for centuries. Not just What If and “Imaginary Tales,” but arguably all superhero comics contain thought experiments. While philosophy’s most amazing thought experiments could be adapted into a limited series of illustrated superhero comics titled Thought Experiments, the reverse is true too. Writers and artists of Marvel and DC can be understood as philosophers and their comics as works of philosophy.

And what if they actually are read that way?

That’s the fantastical “what if” premise of What If? Philosophical Thought Experiments of Superhero Comics, a manuscript I’m co-authoring with my W&L colleague Nathaniel Goldberg. I’m happy to report that a university press has given us a green light, and so Nathaniel and I are spending part of our summer revising for an August deadline. You’re currently reading a condensed draft of the introduction.

Each chapter presents puzzles, or philosophical thought experiments, derived from superhero comics. We then select tools from philosophers—Kant’s Categorical Imperative, Descartes’ evil genius, Dennett’s intentional stance, and others—to help solve that puzzle by helping to understand the thought experiments themselves. Our goal isn’t necessarily to explain philosophy. It’s to use superhero comics to illustrate philosophical thought experiments, and then in turn to use philosophy to explain superhero comics. So unless you’re already well-versed in philosophy, you’ll learn about philosophy too.

Philosophers who identify as analytic—which the majority of English-speaking philosophers do—spend a great deal of time analyzing concepts and defining terms. Though literary critics’ attempts at analysis and definition tend to be limited to literary concerns—including in recent years comics—that’s where philosophy and literary criticism happily collide. For the authors of this book, the collision took place in front of Washington and Lee University’s English department photocopier. Nathaniel Goldberg had descended from the philosophy floor because their machine was on the fritz. Chris Gavaler, himself in the English department, was doing some copying of his own. Nathaniel struck up a conversation about Chris’s superhero blog. Superheroes are not the most typical focus for literary criticism, but Nathaniel assured Chris that philosophers wrote about weird things too. In fact, Nathaniel was an expert on Donald Davidson’s The Swampman, a thought experiment Chris noticed resembled Alan Moore’s own Swamp Thing. A conference paper in Iceland soon followed and now this book. In the process, Nathaniel learned MLA citation norms and Chris learned what is now one of his favorite phrases, “necessary and sufficient,” as in “What are the necessary and sufficient conditions of being a comic?”

Chris answers this question in his recent essay “Refining the Comics Form” where he defines a comic as

a static, spatial field with recurrent elements perceived as conceptually discrete images in juxtaposition with other conceptually discrete images, in which the images are pictorial, abstract, typographic, and/or linguistic, but not linguistic and typographic only.

If you prefer a shorter answer, we recommend Scott McCloud’s pioneering 1993 definition: “juxtaposed pictorial and other images in deliberate sequence, intended to convey information and/or produce an aesthetic response in the viewer.” And if you prefer a really short answer, Will Eisner gets it done in two words: “sequential art.” Many comics scholars, including Chris and several philosophers, take issue with McCloud and Eisner for a range of reasons: what about one-panel “comics” like The Far Side and The Family Circle? What about the moving images juxtaposed in film and TV? What about physical panels displayed on a gallery wall? What about juxtaposed images in mediums that pre-date the 20th-century and so the term “comics”? While these questions are good ones, they and others like them are not the focus of this book. This volume includes “Superhero Comics” in its subtitle, and the superhero genre squats near the center of the definitional zone. Our reading list includes only multi-paneled works printed on paper, bound in units of typically twenty-two pages, and published after 1937.

“Superhero,” as naming both a genre and a character type, also presents a range of definitions, which variously include and exclude marginal cases such as Harry Potter, Buffy the Vampire Slayer, and Nick Fury of Avengers fame. Chris takes a different, “no-common-denominator approach” in On the Origin of Superheroes, arguing instead that the category “superhero” has no single necessary or sufficient condition but only a list of potential ones, with different characters demonstrating different combinations with potentially no overlap (3). Twentieth-century philosopher Ludwig Wittgenstein might argue that examples of different superheroes share a “family resemblance.” Just as there many be no single necessary or sufficient physical condition of all members of a family, individual members do share some with at least some others, and through a series of overlaps the family can be picked out as a whole. No matter, since instead of exploring border cases to further test and refine definitions, we again stake our analysis at the genre’s and medium’s centers.

Defining “philosophy” presents challenges too. But enough philosophers think that some sort of analyzing concepts and defining words is what philosophy amounts to that we accept that as our working definition. That helps explain why philosophers often trade in conceptual or definitional work. One common philosophical tool is to try to conceive of or define situations that are not real but that instead reveal lessons for us. In a word, philosophers “experiment” in thoughts, rather than, as scientists do, in labs. These conceived or defined situations are thought experiments, the “What if?”s mentioned above.

Generally, thought experiments involve conceiving of or defining a situation where a few key details are changed from how they ordinarily are to test particular philosophical views. What if an evil genius did trick you into believing that the world around you were real when it really wasn’t? Does imaging that reveal anything interesting about the nature of knowledge? What if your body were slowly transformed into rock and no one around you noticed? Does imagining this reveal anything interesting about the nature of personal identity?

As we know from above, the first thought experiment is from an academic philosopher. The second is from a comic book writer. Each could be developed by either sort of person. Plato wrote semi-fictionalized dialogues, which encouraged readers to imagine themselves in particular situations. Most academic philosophers, before and since, write essays, treatises, or technical books—which are arguably less engaging than Plato’s work. While typical thought experiments, unlike Plato’s, are not presented in fiction, they can be. As philosopher Ross P. Cameron explains, “a typical fiction tends to be much longer than your typical thought experiment and hence can present you with a more detailed scenario” (31). Philosophers Jonathan Jenkins Ichikawa and Benjamin Jarvis even call philosophical thought experiments “mini-fictions” as opposed to the standard-sized ones. Likewise, philosophers Johan de Smedt and Helen de Cruz argue that, though both typical philosophical thought experiments and fiction rely on similar cognitive mechanisms, fiction “allows for a richer exploration of philosophical positions than is possible through ordinary philosophical thought experiments” (59). The exploration is richer not only because it’s more developed but also because, as cognitive research shows, readers of fiction are immersed in a way that readers of philosophy usually aren’t. Smedt and Cruz continue: “Regardless of whether they are outlandish or realistic, philosophical thought experiments lack features that speculative fiction typically has, including vivid, seemingly irrelevant details that help to transport the reader and encourage low-level, concrete thinking” (64). Readers take the scenarios more seriously.

These scholars contrast typical philosophical thought experiments with the longer scenarios in traditional science fiction and fantasy novels, but their points may apply even better to comics. Novels employ words to express ideas, while comics employ both words and images, and so reading a comic operates on an additional cognitive level. It can therefore be both more immersive and more challenging due to its multi-media form.

Of course, neither science fiction and fantasy novels, nor superhero comics, treat their scenarios explicitly as thought experiments. They don’t usually examine the assumptions involved and don’t draw broader lessons from them. And they certainly don’t consider whether the experiments were done under the appropriate conditions, say, by changing only a few features here and leaving the rest as is. There are no appropriate conditions, other than those that make their stories enjoyable. Superhero comics in particular aim first and foremost to entertain. Actual analysis is done better by academic philosophy, just as we’d expect.

Combining superhero comics and philosophy could be a powerful way to explore thought experiments because it merges the strengths of each.

Sandra Chevrier 05

[The paintings at the top and bottom of this post are by SANDRA CHEVRIER, and they’re currently at the top of my “What if we can use for the cover of What If?” list.]


Viewers tend to imagine the simplest solution to the puzzles created by placing two images next to each other. Since Scott McCloud published Understanding Comics in 1993, those puzzle-solving inferences have been called “closure.” To apply it to comics, McCloud focuses on the gutter, “that space between the panels,” as the site where “human imagination takes two separate images and transforms them into a single idea.”

McCloud borrowed the term from Gestalt psychology, which describes how, for example, a viewer mentally fills in the gaps between dots to perceive a dotted line as a “line” and not simply disconnected dots. In comics, viewers fill in gaps in a metaphorical sense—even though there’s usually a literal gap between the images too. But filling that conceptual space doesn’t involve imagining more images. We don’t mentally draw new panels. We just understand the implied content. The information is image-less.

In other visual arts, diptychs often create the Gestalt effect of closure by dividing a photograph in half or painting across two abutting canvases to create one visual field that is then physically framed in two sections and hung side by side. Medieval diptychs include literal panels joined by hinges—another metaphor for the gutter (which is itself a metaphor). Since “closure” has been the working term for three decades, I’m giving it a vacation this week and using “hinges” instead.

So in comics, hinged panels are any two-side-by-side images that connect in the viewer’s mind. Usually that connection is spatiotemporal. Unless forced to think otherwise, we tend to assume that the second image depicts the same setting after the shortest likely interval has passed. I was on vacation with my family in Europe last month, so I”ve created some hinged panels to illustrate. We were in the Czech Republic, Slovakia, and Austria–though in Italy these might be called “fumetti,” the Italian term for photo comics.

The first four provide a typical spatiotemporal hinge, sometimes with the viewer’s point of view remaining stationary and sometimes moving forward through the setting or shifting angles, as the subjects of the images move too:


The expected forward-moving hinge is so strong that it doesn’t matter what order the images occurred in reality. The next four hinged panels actually reverse the order that I snapped the photos, but the effect is the same. Time still seems to be moving forward–mainly because nothing within the images prevent that assumption:

The temporal hinge also helps to explain the spatial hinge, especially when the setting doesn’t repeat any overt elements. A viewer probably makes sense of the next pairing by inferring that the figures in the first image walked until they arrived at the cafe in the second–even though the backgrounds don’t have much in common.

But how much time passes in the hinge? Usually the shortest possible. So a viewer would assume that the above daytime images occur on the same day. But what about the next hinged panels? Unlike walkers wearing the same clothes as they walk and then later sit, the mannequins don’t give any action clues. Is this an hour later, a day, a month?

In addition to that forward-moving temporal hinge, the next hinge also connects the images by their internal shapes: the triangle of the first person’s body followed by the triangle of the next person’s arms. There’s a thematic hinge too. Both are photographs of people taking photographs of a Klimt painting. I wonder if that thematic effect overrides the temporal hinge so that a viewer doesn’t necessarily assume the two panels are in chronological order?

And what about this statue? The change in background quality and silhouette might imply a temporal hinge of several hours–or is that just because of the change of angle?

These next panels reverse angles too. And if you assume the point-of-view represents a character (the photographer), then you assume that they depict the central building at different moments in time. But without that assumption, would they instead appear to be simultaneous from two angles of an omniscient visual narrator?

And these panels are hinged by an effect very near to the original meaning of “Gestalt.” The two panels almost line-up, the second revealing that the cemetery in the first is raised. And to the degree that they do line-up, they probably also appear to be simultaneous. The gestalt hinge eliminates the forward-moving temporal hinge.

Things get even stranger here. Two versions of a Klimt: the original and a fake. Which is which? Which was photographed first? I suspect the hinge in this case doesn’t trigger any temporal inferences–or rather the two version are understood to exist simultaneously, and so their image here do too.

And what now? Two windows, but not quite the same window. Is this only a thematic hinge and so temporally the two can be read in either order? Or do you understand the figure in the second image as having stepped into the frame after the first image–and so a forward-moving temporal hinge despite the windows not matching?

The front of a painting and the back of a painting–or really a different work of art pretending to be the back of the first painting. Accepting the illusion though, the front and back of a painting obviously exist simultaneously, so do the images too? Or do you imagine someone flipping the painting over during a hinged moment?

And last and least: is there any spatiotemporal relationship between these two images? Or is the hinge entirely thematic: oddly placed mannequins? The actual temporal leap of the photos is about a year, and the spatial leap is continental, but the hinge doesn’t imply that.

You’re now free to return to your regularly scheduled terminology. But I do wish we could toss out the unnecessarily convoluted “closure” and replace it with a clearer term like “hinges.”


Tags: ,

Image result for I, Parrot

The graphic novel I, Parrot combines two unquestionable talents. Deb Olin Unferth is a major new literary voice, whose award-winning short prose has appeared in a range of top literary journals, and her book-length work includes two story collections, a novel, and a memoir, all published by prestigious independent presses. Elizabeth Haidle is creative director of Illustoria, a visual storytelling magazine for children, and she brings a smart, cartoon energy to Unferth’s writing. Together the two tell the story of their narrator Daphne’s struggles to win custody of her son, maintain her relationship with her boyfriend, and care for 42 exotic parrots. The three goals interweave since the parrots belong to her boss, and she is taking care of them to earn money to pay her lawyer, while also proving to the court that “inappropriate abusive men” are not “loitering the household,” a fact made increasingly clear as her boyfriend, Laker, takes on the positive role of step-father to little Noah. Despite dominos of mishaps, the fragmented family and their 42 adoptive pets pull through together.

Although Unferth’s family-oriented plot and Haidle’s style sometimes evoke children’s illustrated books, the occasional “fuck” in Laker’s dialogue clarifies the target audience. This is for grown-ups—and yet the intentionally simplistic rendering is more than surface details. While always expressive, Haidle’s faces are more geometric than human: circle cheeks, triangle noses, blocks of hair. The effect is counterweighted by the subtle gradations of her interior shading, which look brush-stroked in gray wash.

Image result for I, Parrot

Her page layouts are playful and evolving, varying from traditional gutters to open panels to maze-like circles, with a recurring motif of diagonal divisions.

Image result for I, Parrot

Image result for I, Parrot

Image result for I, Parrot

The overall effect creates the sense of a children’s world inhabited by adults—which also describes Daphne. The core of her life is not her job or her boyfriend but her son. Haidle’s art makes that fact palpable by rendering every element of Daphne’s story in a style most suited to a child.  Even when Noah isn’t on the page, he is still his mother’s cartoon heart.

Related image

As the authors describe in a Comics Alternative interview, the project originated with Unferth’s stick-figure sketch of the novel, which earned her a contract with her publisher who then introduced her to Haidle. The two worked together long-distance, via email, Skype, Dropbox, and one extended visit. While most comics collaborations begin with a written script, Haidle instead adapted Unferth’s visuals, while Unferth in turn revised to include not only Haidle’s input but her personal experiences—including her own son’s insights into the character of Noah.

While the relationship of writer and artist is always complex, the complexity is even greater here. Typically an artist receives verbal descriptions only, often with complete control of layout. But any given image choice may or may not have originated with Unferth, with Haidle translating rather than wholly inventing visual qualities. I suspect this somewhat reduced Haidle’s role as author. It also implies that Unferth is the primary author, and Haidle her illustrator, a common credits division, which the cover and title page resists by listing both creators side by side and without attribution. It is Unferth’s first graphic novel, a form she said was different and harder to work in than she expected. She calls I, Parrot “a plot-heavy book,” something she strived for since her wide experience in prose writing gave her expertise in plot but not image. She also gives Daphne’s voice a sharp but believable eloquence from the opening page, describing, for example, “relentless errands over a churning earth” and “the roar of the unhappy mind.”

The novel is less successful at exploring the intersections of word and image that define the uniqueness of the comics form. Often a panel’s figures and its narrated or spoken language overlap in ways that duplicate each other rather than provide independent information that combines in unexpected ways. Page one, for example, opens with the narration: “I finally found a job,” followed by a smaller script statement, “That’s me, Daphne,” and an arrow pointing at a woman smiling and waving as if at the reader. Daphne as drawn includes additional information (she has long black hair, etc.) but the image-text relationship is rudimentary. When Haidle draws Daphne slumped in a chair, Unferth’s talk balloon “Sigh” adds nothing the image did not already convey. While the redundant style further implies a children’s book aesthetic consistent to the novel’s theme, Unferth and Haidle rarely challenge those visual norms and so don’t use other effects available to literary graphic novelists.

Since Unferth drafted the novel in sketch form and it is her first graphic novel, it’s not surprising that she hasn’t plumbed the form’s full potential yet. While I, Parrot is a solid entrance into the field, I predict Unferth’s future graphic work will go further.

Image result for I, Parrot


[A version of this post and my other recent comics reviews appear in the comics section of PopMatters.]

Tags: , ,

While wordless comics date back to the woodcut novels of Frans Masereel and Lynd Ward of the 1920s and 1930s, and Max Ernst’s 1934 A Week of Kindness, “A Surrealistic Novel in Collage,” revealed the non-naturalistic potential of sequential art, one of the most successful explorations of wordless surrealism in graphic form is the more recent H Day (PictureBox 2010) by Renée French. Though French also publishes children’s picture books under a pseudonym, H Day is hardly for children. There is no overtly violent or sexual content, and while the imagery sometimes evokes children’s genres—a cute dog, a round-headed main character—the sequence is quietly disturbing in tone, with a narrative logic that teeters provokingly on the inexplicable.

Image result for H Day (PictureBox 2010) by Renée French

French divides the novel into seven sections or, as she subtitles all but the first, “stages.” The shortest is five pages, the longest twenty-four. After an initially and especially ambiguous introductory set of single images, each stage consists of full-page images paired on facing pages. The pages are atypically small, 6” x 7”, and so well suited to single images. (French worked with even smaller pages for her earlier Micrographica.)

The spine divides the novel in half both physically and narratively. left-hand pages feature unframed white spaces with a recurrent human figure composed almost entirely by outline. Its only interior marks are dots for nipples, a curved line for ambiguous genitalia, and occasional curved lines for knees. The figure has no ears, hair or facial features. It appears on roughly eighty pages. For the first thirty, the figure stands in empty white space, interior and exterior defined only by its outline. Shortly before the novel’s midpoint, French draws a lone bed, and on the subsequent page, the figure reappears and soon lies across it, where it remains until the sequence’s penultimate page in which the figure stands to leave, after which the bed is featured alone again in the concluding image.

Image result for H Day (PictureBox 2010) by Renée French

The figure, both while standing and lying, undergoes a range of surreal metamorphoses. An undefined circle appears within the figure’s head on its second page. Though composed of the same kinds of lines that indicate exterior body features, the circle suggests a physically interior element, as if the head, unlike the rest of the body, is viewed in cross-section—an effect created by the absence of facial features. The circle might also be understood as metaphorical, until it takes on overtly physical qualities, soon expanding beyond the outline circle of the head itself. The figure punctures the circle with its hand and extracts a ribbon-like object that possesses more interior shading lines than anything else on the page. The ribbon retracts soon, leaving a pucker mark from which a straw extrudes, wrapping itself around the head-replacing object. A range of even more extreme transformations follow: pillowy worm-like objects grow from the head and crawl about the bed; cage-like tendrils wind from the head to form a lattice of ropes with the headboard—all while the figure remains otherwise motionless. Because the setting is also stationary, the left-hand pages create a partial flip-book effect, with each metamorphosis occurring incrementally.

Image result for H Day (PictureBox 2010) by Renée French

Right-hand pages feature a radically different narrative, both in content and style. Where left-hand pages use stark open space, all right-hand pages create fully defined rectangular panels. Though frameless, the panels establish borders through meticulous pencil shading that also defines rich spatial environments of ocean, sky, clouds, curving hills, and, most prominently, a city-like landscape of uniformly shaded building-like monoliths creating street-like spaces between them. Though this ambiguous world is initially peopled by darkly shaded human figures, the protagonist of the sequence is a small, dark dog who possess the only facial features in the novel. It wanders alone, struggling against a set of antagonistic forces: swirling black smoke emitted by an industrial-sized chimney, swirling gray water emitted by a sewer-like drainage pipe, and swirling, ant-like dots that either swarm around dead bodies or emerge from them to merge with the smoke.

Image result for H Day (PictureBox 2010) by Renée French

Though surreal, the narrative effect is still largely naturalistic as the dog makes various attempts at escape. But French at times undermines even this quality when figures, including the dog, transform between pages into similarly shaped objects that appear to be formed of wound string—with no contextually implied explanations. “stage 5” also abandons the dog and its environment entirely to depict a sequence of transforming birdcage-like objects with no earlier presence in the right-hand sequence. “stage 6,” however, returns to the surreal worldscape as a new monolith sails toward the dog to open a door-like passage and extend a plank which the dog walks to vanish within the monolith’s unknown interior before it sails away.

The final right-hand page features a version of the bed from the left-hand sequence, only rendered in the more finely detailed style of the right-hand drawings and with cross-hatched walls and floor of a more fully naturalistic space. The effect is not only a surprising merger of the two otherwise unrelated sequences, but it retroactively suggests a metaphorical link between them throughout. Because the bed does not appear earlier on right-hand pages, it is understood to be the left-hand bed, which, given its concluding primacy, refigures the entire righthand sequence as an expression of the left-hand sequence. The dog and its worldscape are not “real,” but are figurative representatives of the transformations of the left-hand character. Though those transformations may themselves be understood as metaphorical, within the logic of the two-sequence pairing, they are the novel’s baseline reality and so are “real” in that sense.

Image result for H Day (PictureBox 2010) by Renée FrenchDespite all of the surreally ambiguous imagery, the two-fold narrative concludes clearly enough: the dog’s sailing away and the human figure’s leaving the bed are linked, positive outcomes. H Day has a happy ending. And while the imagery is open to interpretation—are the shapes that extrude from the human figure’s head visual embodiments of “thoughts” and the dog’s narrative the content of those thoughts?—French’s not-entirely-wordless novel does provide one unavoidable interpretation. The back covers states: “the artist illustrates her struggles with migraine headaches and argentine ant infestation.” While I appreciate the pleasures of a narrative hook, I wish the marketing team who wrote the summary did not include so reductive an explanation.

French has made similar statements in interviews; she told WOW x WOW: “My book, H Day (the version in France is called Céphalées, and it’s silent so that only affects the title) is an attempt to show what it’s like to have a migraine, from the outside and the inside. There’s an ongoing series of drawings on the left hand pages that take you through the pain part of it, showing a character who eventually merges with the bed in a pretty violent way. And then on the right hand pages there’s a story that I’d visualized for years, in order to distract me from the headaches. Even with that entire book, I still go back to that subject. Most of the portraits with things exploding out of the face or the skin warping around the head, are based on the migraines.”

But the back-cover summary, even aside from its erroneous emphasis on an ant infestation, does the novel a disservice. French’s decision not to include her explanation in an introduction or afterword establishes the novel as independent of the creative history that produced it. Yes, French suffered migraines and those migraines led to these images, but the novel is more than that personal chronicle. It both contains those autobiographical facts and exceeds them—an aesthetic effect undermined by the back cover.

And yet, while the title H Day is inherently ambiguous, the French title, Céphalées, means simply “headaches.” The difference of a single word—the novel’s only word—overwhelmingly defines its content. Where Céphalées is the impressionistic tale of a migraine attack, H Day is a surreal study in narrative ambiguity.

I recommend H Day.

Image result for H Day (PictureBox 2010) by Renée French

[A version of this post and my other recent comics reviews appear in the comics section of PopMatters.]

Tags: ,

I’m happy to report that Bloomsbury has green-lighted my and my co-author Leigh Ann Beaver’s book proposal: Creating Comics. It’s based on our hybrid creative-writing/studio-arts course, which we taught for the second time in spring. We’re busy drafting and illustrating now, plus we asked our students if we could include some their work in the book too. They said yes. So I’ll be happily posting work-in-process material on and off for probably the next year.

Working with Leigh Ann has been a massive learning experience for me, especially since I (like some of our more hesitant student) have uttered the dreaded sentence: “But I can’t draw.” So a lot of our first chapter is about building skill and confidence–mine included.  Here’s the best trick:

Working from photographs, preferably your own, gives a comic real-world specificity. Comics creators from a full range of genres and styles begin by staging a photo shoot. Robyn Warhol describes graphic memoirist Alison Bechdel’s “practice of taking snapshots of herself posing for each of the characters in every frame, then draw from the snapshots … to get every bodily gesture, every wrinkle in the clothing, every angle just right” (7). Bechdel does not reproduce every wrinkle in her actual drawings—her style in Fun Home is relatively sparse—but the poses add realism to what might otherwise appear cartoonish in its simplicity. Working in the Kirbyan dialect and subgenre of horror fantasy, artist Bernie Wrightson created the premiere Swamp Thing episode in House of Secrets #92 (July 1971) by posing friends in the roles of villain, damsel, and hero-monster. Bechdel dressed in a man’s suit and tie looks a lot more like her father than Wrightson’s friend looks like a mud-encrusted swamp creature, but the photographs still provided the necessary gestures and angles to give the comic a naturalistic edge. For his comics adaption of the silent film classic M, Jon J. Muth takes the photographic approach to its extreme:

All of the scenes in M were enacted by people in character. I cast friends, family, and strangers, gathered clothes and props, and decided where each scene would be shot… After directing and photographing a scene, I would make my drawings from the photographs… If I took a poor photograph—one that was over- or underexposed or blurry—then I did a drawing of a poor photo. I didn’t correct anything…. When I duplicated a photograph by drawing it, the drawing extracted a different range of emotions than the photo. This happened though I tried to be as faithful to the photograph as possible… This was a discovery, and not by design.” (192)

And this is the kind of discovery possible only through image making. No script can produce it.

The above illustration demonstrates a range of photo research examples. Each includes an original photograph and a drawing made from it. The goal isn’t fidelity—unless that’s your particular style. Sometimes the drawn image varies significantly, referencing the photo for general ideas. Other times the references are exact but edited—like Leigh Ann’s sparsely arranged bricks. Some of the images are traced on a light board; others are freehand. Chris made the tree and fence in Word Paint. Some of the images add details—Leigh Ann invented those beach balls but copied Godzilla from a website. She also photographed a colleague to copy in brushwork, showing the differences of media too. Our student Anna pulled a photo from her phone to use in her memoir about running, and Mims snapped pictures of her own hands while in class to use for a character. The four-panel strip at the bottom was taken from a class photo shoot with students posed with instruments and in animal masks. Your camera is an invaluable drawing tool. You’ll use it to refine images as well as create original content through your own photo shoots.

Below is my process of turning a photograph into a comics panel (which I definitely won’t be including in the book). I started with an image Lesley forwarded me from her phone.

The part of the photograph I liked most was also the easiest and most fun to draw, so I started there:

Those ground plants were my next favorite:

And once I have a base pattern, it’s easy to manipulate and insert multiple times:

Like everything else, the tree is just a crosshatch of intersecting  straight lines:

There’s no sidewalk in the photo, but I could picture it and liked how the lines interacted with the geometry of the fence:Another base pattern for the smaller tree:

And drop it in:Fill in spaces in the bark for more texture:And why not expand the fence another lopsided wrung?

Sadly, the perspective on the cenral tree was off, so even though I like how the tree in the photo loops in and out of the fence, I rearranged:

And I hoped some invented roots might help with perspective too:My image is hardly photorealism, but the differences are at least interesting:

Students in my spring term class “doodled” characters and then redrew them a few dozen times in different positions and actions untill their lines came almost effortlessly. I decided to follow my own lesson plan and created two characters while on vacation last week:

Our family vacation included Prague, Bratislava, and Vienna, so I Word-Paint doodled while in trains, airports, and early mornings in rental apartments.  Central European art museums are crammed with Klimt and Schiele, and I think their distortions were a happy influence. I named the skeleton “Wally,” after Klimt and Schiele’s favorite model and Schiele’s lover, and the demonish character “Egon,” Schiele’s first name.

I’ve also been toying with the idea of placing cartoons inside photographed environments. Our last breakfast was in Vienna’s Korb Cafe, which had the most amazing basement:

The room was closed for breakfast, so I was able to snap a dozen unpeopled pictures on my phone–more than enough for Wally and Egon. I didn’t have a story, so I just started drawing, seeing what gestures and positions emerged. I didn’t have any dialogue in mind either, but I like talk bubbles as a graphic element, so I matched styles to each character. Wally’s body and bubbles are made of squiggly lines, Egon’s are straight-edged.

If you applied for a writing job at Marvel in the 1960s, you were handed four pages of a Fantastic Four issue with all the words removed and told to fill in the captions and bubbles. Feel free to do the same with these:

Personally, I’m not sure words are needed. David Byrne says, “Singing is a trick to get people to listen to music for longer than they would ordinarily.” Maybe words can be a way to get readers to look at pictures too, but I think in comics they can distract from the images. I may experiment with words later, but first here’s the one-page version:


%d bloggers like this: