This study provides an introduction and rationale for the development of Comic Book Markup Language, or CBML, an XML[1] vocabulary for encoding multiform documents that are variously called comics, comic books, and “graphic novels” [2] as well as other documents that integrate comics content[3] or that share formal features with comics content. A markup language is a set of machine-readable textual codes, or “tags,” that are used to identify structure, semantics, and other features of documents and data.[4] The application of these codes to a document is typically a necessary stage, often the most crucial and informative stage, of editing, analyzing, indexing publishing, visualizing, and otherwise studying or manipulating texts in digital environments. The act of encoding a document is a form of discovery, or prospecting, in which the encoder maps a document's structure, identifies semantic elements of interest, and documents relationships internal and external to the document. Scholarly encoding is a form of both reading and writing. The reading, shaped by the constraints of a markup language, is inscribed upon and embedded within the digital text. As literary scholar and digital humanist Jerome McGann has noted, “When you mark up a text you are ipso facto reading and interpreting it. A … text marked up in TEI [a scholarly encoding language] has been subjected to a certain kind of interpretation”  [McGann 2001, 143]. Sperberg-McQueen, Huitfeldt, and Renear assert that markup is constitutive of meaning, markup is interpretive, markup is performative, markup acknowledges or licenses inferences about the text:

Markup is inserted into textual material not at random, but to convey some meaning.

An author may supply markup as part of the act of composing a text; in this case the markup expresses the author’s intentions, e.g. as to the structure or appearance of the text. The author creates a section heading, for example, by creating an appropriate element in the document; the content of that element is a section heading because the author says so, and the markup is simply the method by which the author says so. The markup, that is, has performative significance.

In other cases, markup is supplied as part of the transcription in electronic form of pre-existing material. In such cases, markup reflects the understanding of the text held by the transcriber; we say that the markup expresses a claim about the text. The transcriber identifies a section heading in the pre-existing text by transcribing it and tagging it as a section heading; the content of that element is a section heading if the transcriber’s interpretation is correct, but other interpreters might disagree; it is plausible to imagine discussions over whether a given way of marking up a text is correct or incorrect.[5]

In the one case, markup is constitutive of the meaning; in the other, it is interpretive. In each case, the reader may legitimately use the markup to make inferences about the structure and properties of the text. For this reason, we say that markup licenses certain inferences about the text.  [Sperberg-McQueen, Huitfeldt, and Renear 2000, 11]

Julia Flanders' discussion of scholarly text encoding privileges the role of the researcher/encoder and likens the encoded text to the scholarly article:

Perhaps we need to look to the pleasure of mutability. To recuperate XML, politically and aesthetically, we should be looking not to the paradigms of XML usage that arise from librarianship and from industry-level ideas of the separability of form and content, but rather to paradigms of performance of a different kind. By shifting our view we can understand XML as a way of expressing perspectival understandings of the text: not as a way of capturing what is timeless and essential, but as a way of inscribing our own changeable will on the text — in other words, as a form of reading. Seen this way, XML's presentational flexibility derives not from a separation of presentation and content, but rather from the shifting vantage points from which the text appears to us, the shifting relationships that constrain our understanding of it, the adaptability and strategic positioning of our own readerly motivations.

Ironically, this is a view which emerges most clearly at the margins of current digital text practice. It is not visible in the large digital library projects, whose workflow has come to resemble an industrial operation complete with offshore outsourcing, detailed division of labor, reliance on automation and robotics, and an emphasis, in the output, on uniformity and quantity (thankfully planned obsolescence has not yet become part of the strategy). But we can find it in the small projects designed by individual faculty, typically in conjunction with their teaching, to create digital versions of individual texts which serve as readings: often idiosyncratic, unscalable, representing private insight. They function more like an article than an archive, as a local, contingent expression of insight. [Flanders 2005, 60–61]

The development of a markup language that can support such scholarly reading and interpretation necessitates a careful study and analysis of the content, structure, and semantics of the class or classes of documents for which the language is designed — in this case, the comic book.

CBML is based on the Text Encoding Initiative P5: Guidelines for Electronic Text Encoding and Interchange . The TEI Guidelines are a mature conceptual model for digital representation of multitudinous and disparate document types: inscriptions and papyri; illuminated manuscripts; authorial holograph manuscripts; correspondence; printed books of prose, verse, and drama; critical and scholarly editions; born-digital documents; and more. The TEI Guidelines

make recommendations about suitable ways of representing those features of textual resources which need to be identified explicitly in order to facilitate processing by computer programs. In particular, they specify a set of markers (or tags) which may be inserted in the electronic representation of the text, in order to mark the text structure and other features of interest.  [TEI 2010c]

The TEI Guidelines are widely used in the digital humanities and academic library communities and are maintained by the TEI Consortium, an international body modules, including modules for general categories of documents, such as prose, drama, verse, and dictionaries.[6] The Guidelines also provide additional modules that address more specific textual features and metadata requirements, such as names and dates, manuscript description, linking, textual criticism, and so on. And in their most recent incarnation, the Guidelines provide elements and attributes for linking transcriptions to facsimile page images. This latter feature is especially useful for encoding comics and other graphics-intensive works. From these many available modules, one selects a subset that meets the needs of a particular document, project, collection, or analytical approach. The TEI Guidelines are extremely flexible, providing a vocabulary and mechanisms for encoding and describing a rich diversity document types. However, recognizing that not every document type and representational requirement may be anticipated, the Guidelines provide a well-documented system for customizing and extending the provided tag set with new and modified elements and attributes.[7] TEI, as delivered by the TEI Consortium, is remarkably well-suited to encoding many aspects of comic books; nevertheless, conceptual clarity and practical benefits may be gained from some modest modifications and additions to the stock TEI Guidelines. Hence CBML, a TEI customization with elements and attributes for encoding many of the structures and features found in comic book documents.

The language under discussion here is called Comic Book Markup Language in part to highlight the book-ness and bookishness of these documents, their material properties and bibliographic characteristics. Graphic narratives typically manifest as “books,” stapled or otherwise bound leaves, perhaps thirty-six pages, with an interesting and complex structure, incorporating the graphic narrative — the sequential art and text or “comics” content — alongside a rich assortment of paratexts: advertisements, fan mail, and so on. The emphasis on both “comics” and “books” in the title of the language signals an awareness of the full range of content in the material artifact and the integration of comics content with related paratextual content. The material properties of the book — the codex form, the leaves and pages, the physical properties of the paper — are inseparable from the structure, pacing, and design of the narrative. Certainly less page-bound organizational and compositional frameworks are possible, such as the newspaper strips with long-running narrative arcs, in which the daily “strip” of three or four panels is the basic structural unit.

The traditional grouping of panels into deliberately composed groups (often corresponding to the physical page or the “strip” of a newspaper daily) is being challenged by changing publishing and reading technologies. iPhones and other smartphones have become popular devices for reading newly published comics as well as “reprints” of older comics. However, a full-page grid of panels is not easily readable on the smaller screen of the typical smartphone, so the software interfaces on such devices focus on a single panel at a time. The deliberate juxtaposition of graphic and textual elements in the original composition is shattered by the interface requirements and limitations of the reading device. As new comics work is increasingly targeted at such digital platforms, the traditional grouping of panels into compositional units resembling pages may be abandoned for new compositional strategies. Larger format devices like the iPad and other tablets are better able to represent full-page compositions of panels while also allowing zooming in to focus on individual panels. The isolation of a panel from its surrounding context is not easily achieved in print media. While the reader's eyes and attention may focus on a single panel at a time, other panels on the page remain in the reader's field of vision. The migration of comics content to digital reading devices, and the structural and aesthetic implications of that migration, call attention to the impact of the material characteristics of the comic book document.

Comic books are often very formally self-conscious documents and express a fascination with their own bibliographic identities — creators (comic book writers and artists) become characters in the narrative, editorial notes refer to episodes from prior issues, comic books and paratextual elements, such as advertisements, are parodied within the comic book narrative, publication milestones (such as the first, fiftieth, or one hundredth issue are highlighted and celebrated. These literary, rhetorical, and commercial moves point to a self-awareness of the comic book as document and bibliographic object.

CBML is intended primarily for representing, modeling, and analyzing twentieth- and twenty-first-century comic books, daily comic strips, longer narratives or “graphic novels,” and Web comics and other comics content published on digital platforms, such as smartphones and tablets. CBML may also serve as a possible solution for encoding certain documents we might not normally characterize as comics or comic books, but which share many formal characteristics with comics. In his influential Understanding Comics, Scott McCloud's definition of comics encompasses Hogarth's narrative picture series, the Bayeux Tapestry, and pre-Columbian picture writing as found in the Codex Zouche-Nuttall [McCloud 1993, 10–17].

Comic books and graphic novels have been the subject of serious critical attention for some time, and increasingly so with the emergence of scholarly disciplines such as popular cultural studies and new areas of interest in traditional scholarly fields such as English and American literature. In 1992, Art Spiegelman won the Pulitzer Prize for his Maus, a comic book narrative of holocaust survival. In 2001, Michael Chabon won the Pulitzer Prize for his novel The Amazing Adventures of Kavalier & Clay, which relates the experience of two Jewish cousins working in the nascent comic book industry at the beginning of WWII. Comic books and the mythologies they have spawned continue to be a vital part of our popular culture and national consciousness. Witness the surprising and almost unprecedented popularity of the recent spate of superhero feature films based on the characters of Marvel and DC, the two largest publishers of comic books. In the 70s, 80s, and 90s, the Superman and Batman film franchises produced regular blockbusters. Since the more recent financial and critical success of Marvel’s X-Men and Spider-Man film franchises, the film industry has been flooded with films based on comic books, many based on mainstream super hero comics, others adapted from realistic, personal, and autobiographical graphic narratives, often published by smaller publishers. In 2012, Marvel’s The Avengers film from Walt Disney Studios broke all previous box office records with over $200 million in ticket sales for the biggest opening weekend of all time [Barnes 2012].[8] These few examples, which are frequently discussed in scholarly articles, monographs, and in school and university classrooms, demonstrate the continuing and perhaps increasing importance of comic books as an art form and cultural touchstone.

Like the study of other popular art forms — film, television, jazz, rock — comics scholarship is a thriving field of academic research. Scholarly books about comics are being published by university presses[9] and articles about comics appear frequently in peer-reviewed journals such as The Journal of Popular Culture. In 2008, English Language Notes, published an issue devoted to “Graphia: The Graphic Novel and Literary Criticism” [Kuskin 2008]. ImageTexT is a peer-reviewed online journal devoted to the study of comics. In 2007 ImageTexT published a special issue on “William Blake and Visual Culture,” recognizing the connection between modern comics and earlier art forms that integrate text and image [Whitson 2007].

In light of the ongoing cultural and scholarly relevance of comics, one goal of CBML is to support the study and analysis of comic books in the way that digital collections of, for instance, English poetry or American fiction support the study and analysis of more traditional literary forms. A large corpus of digitized comic books, along with encoded transcriptions and descriptive metadata, would allow scholars to search the text of comic books, search for keywords related to topics of interest, search for the appearance of particular characters, or search for works by particular writers and artists. Additionally, when exploited to its full potential, a large CBML collection would allow searching — and other forms of computer processing and computational analysis — based on structural, aesthetic, and informational and documentary features peculiar to the genre of comic books.

Large digital collections of comic books would support the types of searching that is now taken for granted in large digital collections of literary and other texts. This sort of functionality has proven incredibly useful and has transformed the ways in which scholars and students conduct research and “read.” This now commonplace functionality though can be particularly significant for the study of comics, given the complexity of the fictional universes that have developed in the works of many comics publishers. For instance, major characters like Superman, Batman, Spider-Man and Wolverine appear regularly in multiple titles. DC Comics' Batman is the featured character in Batman, Detective Comics, and other titles; he is a prominent member of the Justice League superhero teams featured in a number of different titles; and he shares adventures with other heroes in titles such as The Brave and the Bold and World’s Finest. Since Batman shares the same fictional universe with most of DC’s other characters, he will make frequent “guest appearances” across their entire line of publications. One will even find publishers occasionally cooperating to merge their fictional universes and publish works that feature, for instance, Superman from DC and Spider-Man from Marvel. Therefore, a scholar researching the development and representation of a particular character cannot confine herself to a small number of publications, but must consider almost the full output of a publisher or even multiple publishers.

We've been looking thus far at some of the more practical goals for CBML, providing a digital format to support digital research collections that in turn support different types of searching and digitally-enabled analysis. Another motivation behind the development of CBML is the desire to explore more generally the modeling and representation of the broader class of documents that tightly integrate pictorial images and text. Comic books are just one such type of complex graphic document; other examples include illuminated manuscripts; seventeenth-century alchemical manuscripts, with hand-drawn figures and graphic symbols; artists’ books; artists’ sketchbooks; illustrated poems like those of William Blake; letterpress productions like those of the Kelmscott Press; illustrated children’s books; newspaper and magazine advertisements; and even Web pages and other born-digital media. Comic books provide some of the more obvious examples of this integration and co-dependence of image and text, but lessons learned in the modeling of comic books may be applied to other forms.

Martha Rust, in her essay “ ‘It's a Magical World’: The Page in Comics and Medieval Manuscripts,” writes:

Both medieval book artists and contemporary cartoonists make use of the page as a device for giving their readers access to a domain of representation that is beyond the regimes of either pictures or words — yet somehow in the shadow of both.  [Rust 2008, 25]

The page is indeed one of the primary structural and compositional units of comics. In TEI the page is viewed as a “milestone,” an empty marker within the flow of the text. In comics books (and similar documents), the page is not an arbitrary milestone, but a compositional feature — a composed container for “panels” and text.

The miniatures pictured in Figure 2 and Figure 3, tightly integrate text and image. Moreover, we find discreet, bordered, sequential images, similar to comic book panels, and text representing speech, like comic book word balloons, originating from the figures’ mouths. In Figure 3, below the “panels” we find explanatory text, similar to the narrative captions found in comic books. Such examples illustrate formal and structural features of comics shared by many other document types. The analytical and descriptive strategies of TEI and CBML may prove useful in studying not just modern comics but an even more historically and culturally diverse assortment of document types.

Perhaps the most significant motivation behind the development of CBML is the wish to support for comic books and related documents the interpretive strategies and inscriptions of readings — discussed in the introduction above — that occur when a text is encoded. Unlike the traditional academic essay, in which a reading or analysis is sprinkled with relevant quotations from primary source materials, an encoded text may combine a digital representation of a full source text inscribed, in the form of markup, by a reading and performance of the text. Like the image and text of a comic book, the source text and the scholar's reading and analysis are inextricably intermingled in the encoded document. Encoded documents are metadocuments “which describe and enhance information but also serve as performative instruments.… As texts that describe a language, naming and articulating its structures, forms, and functions they [metalanguages and metatexts] seem to trump languages that are used merely for composition or expression”  [Drucker 2009, 11]. CBML facilitates more transformative reading strategies that have developed in response to the increasing numbers of high-quality digital and digitized documents. Literary scholar Franco Moretti has popularized the term “distant reading” to describe strategies for analyzing massive numbers of literary and historical documents:

[I]f you want to look beyond the canon … close reading will not do it. It’s not designed to do it, it’s designed to do the opposite. At bottom, it’s a theological exercise — very solemn treatment of very few texts taken very seriously — whereas what we really need is a little pact with the devil: we know how to read texts, now let’s learn how not to read them. Distant reading: where distance … is a condition of knowledge: it allows you to focus on units that are much smaller or much larger than the text: devices, themes, tropes — or genres and systems. And if, between the very small and the very large, the text itself disappears, well, it is one of those cases when one can justifiably say, Less is more. If we want to understand the system in its entirety, we must accept losing something.  [Moretti 2000, 57]

Digital technologies facilitate both close and distant reading strategies. Individual comic books and graphic narratives may be better understood through strategies of close reading, with careful attention paid to minute details of linguistic and visual language and document features, but comic books as a system and extended comic book narratives — which often unfold over decades in thousands or tens of thousands of individual documents by hundreds of creators (writers and artists) — benefit also from the distant reading (of text, image, and metadata) from comic book documents.

CBML aims to support such digitally-enabled and embedded performances and readings in comic books and across a range of graphics-intensive documents. This motivation is not unique to the study of comics or the development of CBML, but comic books and related documents do have particular, if not unique, features that must be supported in markup languages if we are to realize this goal.

The following sections examine features and content of comic book documents and propose encoding strategies for representing, documenting, and analyzing those features and content types. These encoding strategies incorporate both new elements introduced in the CBML customization and existing TEI elements.

The comic book is a particularly complex class of document. A typical comic book, from the early twentieth century to the present, contains diverse content types. Alongside and intermingled with comics content, we may find prose fiction, advertisements, editorial and promotional content, fan mail, and bibliographic metadata about creators and publication details.

The examples above illustrate how CBML can capture a transcription of the text found in a comic book. Other important features are also captured, such as the structure of the comic book; the sequence of pages and panels; the grouping of panels into compositional units that may or may not correspond to a physical page; classification of panel transition types and analysis of individual transitions from panel to panel. However, the examples above do not attempt to describe the pictures one finds in the comic book, nor should they. Comic books are a visual, graphic art form combining text and image. CBML/TEI/XML is a text format. While one could certainly use CBML to describe details about any or all of the pictures in a comic book publication, such an effort would undermine the hybrid form of the comic book. The visual, pictorial, and graphic design elements of the comic book simply cannot be fully or adequately described or translated as text. While many design features common to textual documents, such as text size and font characteristics, may be reasonably and usefully described using common TEI techniques, it would be futile and impractical to attempt to describe every detail of every picture in a comic book document. The encoded document, with markup containing and describing metadata, structure, transcription, and analysis, should co-exist with and be linked to digital facsimile page images of the comic book. Thankfully, in current digital environments and on the Web, such linking and display of text and image is relatively simple, if time-consuming, to accomplish. Section 11.1 Digital Facsimiles in the TEI Guidelines describes existing TEI elements and attributes for linking digital transcriptions to digital facsimile images [TEI 2010c].

On the other hand, depending on the goals of any particular project employing CBML, some scholars will wish to add to the encoded text some description and analysis of the pictorial aspect of the comic book. If one has such needs, CBML and TEI provide suitable mechanisms. Typed TEI <note> elements may be used to provide detailed descriptions of the pictorial dimensions of the document. For example, one might use <note type="panelGrpDesc"> for a “panel group description” or <note type="panelDesc"> for a “panel description.”

Another option is to use the @ana attribute, illustrated above, to apply analytical or interpretive schemes to the pictorial, graphic, and visual dimensions of the text.

Even when focusing solely on the text of comics, the transformation from handwritten or printed text to unformatted digital transcription may lose important information-bearing graphic and design features. For instance, in the examples in Figure 27 below, distinct fonts and balloons are used for particular characters in an attempt to represent graphically the tone, timbre, or other qualities of speech, resulting in a sort of visual or synaesthetic onomatopoeia. In such cases, TEI's <rendition> element and @rendition and @rend attributes may be used to describe such design features.

The basic structures of CBML were based on representative samples of common and relatively simple structures and panel layouts. For instance, panels normally are arranged in fairly regular grid patterns. 2 x 3 panel grids (three rows of two rectangular, nearly square panels) are very common; 2 x 2 and 3 x 3 layouts are also prevalent. But variation from strict and regular grid layouts is ubiquitous. Panels of every size and shape — circles, triangles, irregular polygons, and fluid organic shapes — are also common, as are borderless panels with ambiguous boundaries.

Simplistic approaches to encoding the spatial and sequential relationships among such panels are foiled by frequent variations and complexities in panel size, shape, and arrangement and particularly by ambiguous sequential positioning of panels. As determination of meaning in a textual document is dependent on a particular, usually obvious, sequence of words, so is determination of meaning in a comic book dependent on a particular sequence of images and panels. While a meaningful panel sequence in comics is usually apparent, the conventions of graphic expression in comics are not as fixed, well established, or as well known by general readers as the conventions for textual expression and narrative. And just as writers and poets break rules and conventions, so do comics writers and artists play with the rules and conventions of comics narrative. In Figure 30 below, a single obvious sequence of panels is not at all clear — multiple valid sequential readings are possible. Also note the playful use of diegetic text, with bibliographic information about the creators and title of the story embedded in the pictures. Near the upper left of the page is a blue-tinted panel with the names of the creators of the story: “LEE STERANKO SINNOT ROSEN in Another Epic!” The title of the story, “Tomorrow You Live Tonight I Die!,” appears on a calling card in the panel at the lower right-hand corner of the page. A <note> and omission of sequential numbering on the <cbml:panel> elements could be used in the encoding to address such sequential ambiguity.

Irregular and unusual panel arrangement, composition, and presentation may pose challenges for “reading” of the visual and textual language of comic books, just as difficult syntax, unfamiliar vocabulary, neologisms, or unconventional typography and punctuation may present challenges for reading primarily textual content. Most comics have panels with clear borders and generous gutters between borders. The borders and gutters between panels correspond to the punctuation and spacing used to enhance clarity and readability in written and printed texts. Panels without clear borders and gutters introduce another type of visual ambiguity in comics. Figure 31 illustrates a panel group with discreet pictorial moments that correspond to panels, but these “panels” lack conventional borders and gutters. At least six such pictorial moments or “panels” may be identified in Eisner's page from Figure 31. The top of the image depicts a man and woman running through Charles de Gaulle Airport; the next element shows, from a distance, the couple as they ride up an escalator. The lines of the escalator tube, extending more or less horizontally across the entire page, cause this image to serve simultaneously as a discreet pictorial moment and as a border or gutter separating the “panels” above and below the escalator scene. Two more “panels” are found directly below the escalator. On the left, the couple rush through the crowded airport; on the right, they arrive at their seats on the plane they have been rushing to catch. A very subtle “panel” emerges quietly below these two panels as the aircraft, seen from a distance, takes off through the clouds. Like the distant view of the escalators, this image does double duty as “panel” and as border/gutter. In the final “panel” of the page, the man and woman, reclined in their seats, relax and resume their conversation. In those images (the escalator and the ascending aeroplane) that function as both panel and border/gutter, we might say, borrowing terminology from textual criticism, that the substantive content has merged with the accidentals of the document. By encoding these graphic and textual moments within <cbml:panel> elements, the scholar encoder has imposed an interpretation and asserted various claims about the document. Additional @ana attributes might be used to indicate those “panels” that function as both panel and border/gutter. <note> elements might also be used to discuss other issues related to the general graphic ambiguity of the page.

In Figure 26 above, we saw an example of a speech balloon crossing the gutter from one panel into another. Another type of visual ambiguity is introduced when pictorial elements cross panel boundaries and co-exist is more than one panel. Since different panels often depict different physical spaces and different moments in time, the presence of a pictorial element in more than one panel can be particularly jarring and can be used to establish complex spatial and temporal relationships. Figure 32 below shows an example of this type of visual ambiguity in an early WWII-era comic book: The panels above are not uniform in shape or size. Pictorial elements, such as the gun in the fifth panel and the purple-suited figure and motion lines in the bottom two panels, cross the gutter separating the panels and co-exist in multiple panels. These graphic moves suggest interesting spatial and temporal juxtapositions and facilite visual transitions from panel to panel, breaking down the clear separation of narrative moments and instigating a flow approaching (though still very far removed from) the rapid frame-to-frame transitions found in film.

Figure 33 below, from Morrison and Jone's Marvel Boy [Morrison 2000], shows a more recent example of a page with both ambiguous panel boundaries and overlapping panels. The central action — of the blue-cloaked golden figure lifting Marvel Boy into the air — is witness to its own genesis, visually encroaching on the panel depicting some moments prior in time. As Marvel Boy is lifted up into the underlying panel he becomes a spectator — with the reader — of himself within the previous moment. Likewise the blue-cloaked figure plants a golden boot upon a future moment that proceeds from the central action. An encoded CBML document might identify and analyze such transitions. One could develop a taxonomy to classify the sorts of overlap phenomena found in Figure 32 and Figure 33 and use the TEI <interGrp> and <interp> elements and the @ana attribute, discussed above, to describe and interpret the graphic features and the resulting spatial and temporal relationships.

As is the case in many other art forms (fiction, television, film), serial publication of an ongoing narrative introduces a great deal of bibliographic and metadata complexity. While these issues are not unique to comics, I would argue that the degree of complexity is distinct in comics. Comic books typically contain serialized narratives. Often one finds “stand-alone” issues, which contain a complete story, and often stories extend over many issues, or even many issues of two or more serial titles.[17] In the case of mainstream super hero comics, regardless of whether any particular story is more or less self-contained within a single issue or extends over many issues, these narratives exist in the context of larger narratives developed over many decades by dozens or hundreds of individual writers and artists. Rather uniquely, characters like Superman, Batman, Wonder Woman, Spider-Man, and the Hulk exist in ongoing, decades-long, cross-generational narratives created by collaborative teams of writers, artists, and editors. Although we find similar issues of long-running narratives with changing collaborative teams in film and television, none approaches the longevity and frequency of production found in the comic book narratives.[18] Most long-running comic book features, from Fawcett's original Captain Marvel to Spider-Man to Archie, have complex bibliographic histories involving networks of changing titles, creators, even publishers. These complex bibliographic histories, with multiple creators, are relevant to textual theories about the nature of authorship and the social and corporate nature of document production. In comic books we have examples of collaboratively-produced narratives coupled with collaboratively-produced visual art. The many metadata structures available in TEI lend themselves to description of these complex bibliographic histories and relationships.

The evolution of characters over many decades in serialized tales poses documentary complications that may be addressed through encoding strategies. For instance, one may be interested in the DC super hero known as the Flash. The general concept of a super hero named the Flash, with the power of super speed, published by the comic book publisher DC, is some sort (or some sorts) of bibliographic entity — a character, a subject, a title. But there is not one Flash character (or title) in the DC imaginative universe. The first Flash, with the secret identity of Jay Garrick, was created by Gardner Fox and artist Harry Lampert in Flash Comics #1 (January 1940). This character was revived in the 1950s, with a new costume, new origin story, and new secret identity — police scientist Barry Allen. Allen “died” in the 1985 series Crisis on Infinite Earths and remained “dead” for over twenty (real — not fictional) years . During Allen's absence, his similarly super-powered sidekick Wally West took on the persona of the Flash, as did Bart Allen, Barry Allen's grandson. Many other characters have undergone similar evolutions and transformations.

TEI is well-equipped to capture these complex details of character and identity. The prosopographical features of TEI provide a framework for modeling relations among both real people (e.g., comics writers, artists, and other creators) and fictional characters.

Like most digital humanities projects and markup language development, CBML is an ongoing effort. Although many diverse examples have been shown here, the extent of textual and graphic variation found in the world of comics is astounding. Examples here have focused on superhero comics, the most enduringly popular genre in the history of the art form. However, many other important genres exist — romance, war, horror, science fiction, and fantasy comics — each with their own traditions and conventions. Increasingly, more realistic narratives, autobiographical works, and non-fiction works are presented as comics. Japanese manga and manga-influenced comics are very popular and possess a style and conventions distinct from western comics. Certainly many more difficult and challenging comic book documents will be found to test the general applicability of CBML and suggest additions and modifications to the markup language.

An important issue, not yet addressed in this discussion, is the issue of copyright. Modern comic books, which developed in the 1930s and since, are largely still under copyright protection. The most popular of these, those featuring the famous super heroes, are owned by massive media conglomerates. DC Comics (publishers of titles featuring Superman, Batman, Wonder Woman, the Flash, etc.) is owned by Time Warner. Marvel Comics (publishers of titles featuring Spider-Man, the Fantastic Four, Iron Man, the Hulk, etc.) was recently acquired by the Walt Disney Company. These media companies have reputations for jealously and fiercely protecting their copyrights. Copyright is doubtless a challenge for scholars studying comic books, just as copyright is a challenge for scholars studying film and many other twentieth- and twenty-first century works. But barring further restrictive copyright legislation, many of the older comic book publications will be entering the public domain in coming decades. Under current U.S. Copyright Law, Action Comics #1 (June 30, 1938), featuring the first appearance of Superman, would enter the public domain in 2033, 95 years after publication. And many comic books from the same period are already in the public domain because the copyright was not renewed, when such renewal was still necessary for an extension of copyright protection. In 2008, publisher Image Comics began their Next Issue Project, a series of new comics based on the narratives and characters from comic books published in the 1930s and 40s that had fallen into the public domain. The first Next Issue Project comic book was Fantastic Comics #24 [Fantastic Comics 2009]. It draws on stories and characters from Fox Feature Syndicate's Fantastic Comics, the final issue of which was #23, published in November 1941. The Next Issue Project has identified a number of other public domain titles, including Lev Gleason Publications' Silver Streak Comics (twenty-three issues), Quality Comics' Crack Comics (sixty-two issues), and Harvey Comics' Speed Comics (forty-four issues). These four titles alone provide over 150 comic books that could be used for further collaborative development of CBML and published on the Web and in other media. Certainly there is a great deal more comics content in the public domain or for which permissions may be secured, and nothing prevents a scholar from applying CBML markup to any text as part of a strategy for reading, interpretation, and analysis. The end goal of markup is not and should not always be publication of a digital surrogate. The encoding of a text may be a rigorous intellectual activity that has great value as process, not just as product.

CBML, as described here, focuses primarily on overall structure of the document, textual content, and metadata. Significant work remains to be done giving similar attention to the pictorial dimensions of the comic book. I have argued above that CBML should not attempt to go too far in description of individual images and pictorial details, relying instead on the presence of facsimile page images. Nonetheless, in order to analyze the visual grammar and conventions of comics, additional visual and pictorial features will need to be “identified explicitly in order to facilitate processing by computer programs” [TEI 2010c]. Future work on CBML will include modeling frameworks for analysis of pictorial and graphic features and developing taxonomies for identifying such features, beyond the basic structural components of panels, balloons, captions, and sound effects.

Comic books are an endlessly fascinating medium and one from which the digital humanities can learn a great deal. The structural complexities of the content rivals or surpasses the complexity of content famously treated by digital humanities scholarship: fragmentary classical texts; medieval manuscripts; Blake's illuminated work; and Rossetti's double works. The combination of text and pictures and the role of visual communication in comics prefigures the multimedia and new media of digital environments and the prominence of pictures, graphics, and icons found on the Web and other modern digital interfaces. The corporate and collective authorship of many long-running comic book narratives and the interactive relationship between creators and readers/viewers/consumers of created content provide models of authorship and readership that are relevant to the collaboratively created digital content that we make and study in the digital humanities.

Ongoing development of CBML is documented at http://www.cbml.org/. One may download the most recent schemas and TEI ODD files (which define and document the CBML customizations of TEI) and example CBML instance documents. CBML-L, an email list for discussion of CBML, is available at https://iulist.indiana.edu/sympa/info/cbml-l. In addition to ongoing development and refinement of the CBML markup scheme, future research will involve the creation of a large corpus of pubic domain comics encoded in CBML and the development of interfaces for representing, exploring, and manipulating CBML documents and associated facsimile page images.

I would like to thank DHQ Editor Julia Flanders and the reviewers for their very valuable feedback on this article. I would also like to thank my colleagues Paul Aarstad and Michelle Dalmau for their careful reading and insightful comments.