Mapping the reception of the Ody-C comic book with Twitter
This text details the method used to investigate the reception of Ody-C (2014-…) using Twitter discussions. The results suggest a quantifiable discrepancy between the creators’s preferred frames of reading and the readers’.
Ody-C is a contemporary comic book series, written by Matt Fraction and illustrated by Christian Ward. It re-tells the story of Ulysses, Ene and Agamemnon following their victory in Troy, but displaces these narratives in a science-fiction universe in which all men, save HE/Helen of Troy, have disappeared after having incurred Zeus’s wrath. This means that all the major protagonists of the classic tales are now women or « sebex », a new gender created by Promethene/Prometheus to allow the human race to procreate in spite of Zeus’s edict.
Though it is still in progress – 12 of the planned 36 issues have been published since October 2014 – the series has attracted positive critical attention since it debuted, with articles and interviews in places sur as Wired or NPR.
I first became aware of Ody-C in a conversation about one of the features of its first issue, an 8-page foldout which seemed conceived as a deliberate way to celebrate physical as opposed to digital distribution (Fraction writes: « [I] like finding ways to encourage the fetishization and collection of things » in answer to a letter in #2).
Reading it in a collected edition which ironically did not reproduce the foldout, I realized the series constituted a noteworthy variation of the trend towards dialogic and critical literary adaptations fostered by the development of the « graphic novel » (see Baetens and Frey, The graphic novel: an introduction, ch.8). This is in part because Ody-C references both Homer and science-fiction comics from the 60-70s texts, which serve as a thematic and graphic inspiration. Fraction and Ward have made it clear that they are interested in channeling Forest’s Barbarella as well as Moebius and Jodorowsky’s The Incal. Historians routinely use periodization such as « the long 19th-century » or the « short 20th-century » (1914-1989), and it seems only fair to suggest that there is such a thing as a « long-70s » comics period, spanning from 1964 (Barbarella) to 1981 (The Incal).
The text and the various peritexts – interviews, backmatter, covers – provided ample evidence of the role of these two sources of inspiration both in the conception and in the marketing of the book. I was curious, however, to see how readers dealt with these intertexts. Prompted in part by Ernesto Priego’s recent analysis of Donald Trump’s tweets, I decided I would use Twitter.
In a way, it was an easy choice, since Ody-C points to that social network, by reprinting tweets using the official hashtag – #ODYC – in the comics themselves. Predictably, some readers have then tweeted happily upon finding out that their tweets had been printed, generating some more positive feedback for the boook. It was also my hypothesis that there would be a congruence between social network users and comic books readers, an assumption which was later verified, as we shall see.
As I quickly discovered, Twitter does not readily provide access to historical data. Various academic testimonies suggest that such information is available but expensive. Services such as Twitter Archiver make it possible to store this data as it is produced, but they can only go back 7 days, which is hardly ideal. Given the scope of the project, and since I not anticipated I would eventually need the data, I looked into the possibility of gathering the tweets using Twitter’s public interface as opposed to its API.
This involved scrolling down a lot, to display a very large web page, copying and pasting its content into a file, then cleaning up the results. Because I was merely looking for recurring terms, the fact some of the information would not be readily usable (dates, for instance) did not bother me much.
The process proved tedious but efficient. « Automating » scrolling down helped, and even though I had to switch to a computer with enough memory to display 4000+ tweets in one browser tab, the data collection went well.
After an initial clean-up, I used Voyant tools to make sense of the data. I did not fine-tune the stop words to eliminate the noise entirely, reasoning that I could do so in Excel afterwards. Indeed, tracking down unnecessary lines and consolidating certain variations (« wonder woman » & « wonderwoman ») proved easy and fairly fast.
The full table of relevant results can be found at the bottom of this article.
5126 tweets were collected in the course of this study, from January 2015 to mid-April 2017, and more than 2800 accounts are mentioned in these messages. Though there is some noise in the dataset, this is still a substantial number, considering that the latest issue of Ody-C sold slightly more than 5100 copies in comic book stores.1
I was primarily looking for qualitative data about the reception, and that turned out to be disappointing. The adjectives used in the messages are flattering but not very specific: good, great, amazing, beautiful, favorite, awesome, epic, gorgeous, excited, fantastic, favourite, weird, trippy.2
The abundance of graphic descriptions and the higher frequency of Christian Ward’s name, mentioned twice as often as Matt Fraction’s, may simply be a reflection of the fact that Ward is present and active on Twitter, while Fraction is not.
I was more interested in the fact that The Odyssey was frequently mentioned (30th most common term or phrase excluding stopwords) but that the word « adaptation » was never used, unlike « retelling » (150th most common term). This suggested that my insistence on reading the book as an adaptation was in fact not widely shared. « Gender » was also used much less often than I would have expected, though fairly common.
In other words, the list did not yield the exact results I was expecting. It did however produce the following table, showing all the book and comic book titles mentioned more 30 times or more in the course of these 5000+ messages:
It is tempting to read this as a reflection of Image Comics’s promotional strategy. This is partly accurate, but it bears mentioning that The Walking Dead, Image’s best-seller by far, does not appear. Saga also significantly outsells Bitch Planet, yet it is mentioned less often.
In other words, it is my contention that this list defines the context in which Ody-C is read and discussed. Barbarella, which both Fraction and Ward often mentions, does not appear in these discussions ; The Incal is mentioned 5 times, and Moebius twice.
On the other hand there is a high degree of consistency in this list. In particular, Bitch Planet and Saga can both be roughly described as feminist (or gender-conscious) science fiction. Bitch Planet also plays on historical popular tropes, since it pastiches or reuses the codes of exploitation women prison films ; it also happens to be written by Matt Fraction’s wife. Significantly, the first two Marvel titles which appear on this list – Thor and Squirrel Girls – both also feature strong female protagonists.
Interpreting the results
What this suggests is that readers did not use the frames provided by the authors.
This is not entirely accurate, as The Odyssey features in this list, but Ody-C is a book which includes a two-pages essay about classic mythology in every issue, as well as numerous discussions on the subject in the letter column. The Odyssey only provides content for about 40% of the stories, and only a minority of readers mention the inspiration drawn from 1001 Nights (3 mentions) or Moby Dick (2 mentions), despite the fact that these inspirations are both visible and discussed at length in the comic book.
The reference to 70s comics is also present by association (« weird », « trippy » and Bitch Planet) but the central sources mentioned both by Fraction and Ward are absent from the discussion.
On the other hand, associations based on creators, genres, and gender representation appear to shape the conversation and the reception to a large extent. This points to the insular nature of comics consumption; a network of other contemporary comics to which a series can be compared appear to eclipse other points of references, be they emphasized by academics, critics or by the authors themselves.
|author||cjwardart ou Christian Ward||1693||33,03%|
|author||Matt Fraction ou Mattfraction||861||16,80%|
|genre||scifi or sciencefiction||115||2,24%|
|series||wicked + divine||71||1,39%|
|content||myth & mythology||62||1,21%|
|author||Kellysue De Connick||37||0,72%|
- The series enjoys better relative sales in collected editions, as evidenced by the choice to publish and expensive hardcover version of the first 12 issues [↩]
- « good » is also problematic, in that it is also used in negative sentences, unlike the rest of the list [↩]