Monster Map: Spatial representation of gendered “monsters”

monster hunt
Monster hunt; A brave 2D man facing a dragon (flickr-Jason Gil)

(This blog is a draft version of an essay soon to be posted on Esri Storymap under the name of “Monster Hunt: Spatial Representation of Gendered Monsters.”)

In our deepest nightmares, monsters haunt us. Monsters terrify us with their grotesque bodies, supernatural ability overpowering human strength and paranormal movements. Their existence as well as their heinous actions mock human attempt to understand, grasp, and pin down what they are. So, humanity’s hunt for monsters begin.

When just about to catch them… they escape!

They escape humans not only intellectually but also spatially. One cannot catch a monster easily. It gets away just when one is about to seize it and turns into a smoke or appears to have traveled thousands of miles a way in a second. Typical monsters that easily come to mind such as Count Dracula or the Creature from Frankenstein are particularly threatening because they move about the globe so effortlessly. They dart in and out of European countries let alone the “mysterious” East and Africa.

ghost house
ghost house with spirits bound to the location

They are everywhere: globally roving about but also at home.

However, there are other monsters who stay local and threaten relatively small number of people. Female vampires who do not leave the castle, witches who demonize neighbors, ghosts bounded to houses, a creepy man-killing cat rambling around a deserted fortress, etc. The fear that we have for this type of monsters are quite different from wide-roving creatures. Interestingly enough, it seems these local monsters are usually female: “Cats,” “weird sisters” of Stoker’s Dracula who can’t hunt for themselves since they are imprisoned to the castle, a victimized female ghosts seeking revenge. This project starts from a question whether gender plays a role in special representation of a monster and looks to find a connection between spatial patterns of a monster and the specific type of monstrosity monsters embody.

Monsters and sympathy

Investigating monster literature is an extra-ordinary job. We don’t just get “scary” tales in literature; we get a glimpse of what it is to be a monster. Take Frankenstein for instance. Unlike popular adaptations that prioritize the horrific appearance of animated corpse of the Creature, Mary Shelley’s Frankenstein is a text of interplay between various perspectives. Not only do we get Dr. Frankenstein’s narrative creating, escaping, tracking the Creature’s horrific murders, but significant chunk of the narrative devoted to the Creature and his side of the story. Being born (in most shocking way possible–electronic jolts!), abandoned immediately in cold and dirt, lost in the world (literally), seeking love and connection, brutally ostracized, feeling nothing but rage to revenge on his creator. As many critics have noticed, the eloquence of the Creature gives his narrative the level of sympathy that deserves questions regarding the intention of the author. And as many great literature does, it asks questions about the meaning of monsters and monstrosity. The narrative is set in a way that the readers can sympathize, or at least understand the reasons for his monstrous actions (murdering of a little kid, wife and best friend of his creator). Is he a monster if the society made him the monster?

Pertaining to this project, the travel the Creature initiates is actually driven by the society. He is banished from one home to another only to be kicked out of to another place. He cannot stay in one place because he cannot be a member of any society. Though he is made of human, made by human, speaks human language, he is an eternal exile bound to be around the world. Even his later journey is caused by the “war” between him and Dr. Frankenstein–a war between the creator and the creature–one the Creature would want to avoid but has to face now. Two characters chase each other all over Europe and even to the North Pole.

As seen in this case, the monstrosity of a monster is intricately tied to the location and the journey. The creature becomes (aware of the fact that he is) a monster as he changes location and commits sins. Within a Gothic castle of Frankenstein, he is a gothic monster of action who is hidden, appears, and kills. At the warm cottage environment of De Lacey family, his monstrosity pertains to the surprises the family members feel looking at his body. Even though he is a same “monster,” the type of monstrosity perceived by the characters is geo-spatially distinctive. 

Dr. Frankenstein leaving horrified at the sight of his creature
Dr. Frankenstein leaving horrified at the sight of his creature but notice also how the Creature himself is surprised at being “born” (wiki commons)

And as interestingly, this has a fascinating relation to the level of sympathy. For Count Dracula, it might be argued that the level of sympathy characters (or even readers) feel is lower because he comes from a far away land. He wears strange garments, speaks with an accent, and emanates weird energy. It is his foreignness that distracts the sympathy from others. Is the Creature more sympathetic character because he is “one” of “us?” 

So here are the questions this monster map explore:

  • Central Question: Does gender of the monster affect the distance traveled? How is the distance related to a level of monstrosity and level of sympathy?

Do female monsters show smaller trajectory and consequently lower level of monstrosity than male monsters? Do they have different habitation patterns than male monsters? How does spatial representation affect level of sympathy? Will the gender of the author complicate any pattern?

Will we find any unexpected patterns of monsters with this visualization? Let’s hunt some monsters then. But first… who are on the menu?

Monster Corpus

Now here’s a table of my monsters. Unsurprisingly, these are monsters in fiction. Yes, we do have monsters (supposedly) in real life. But here I use literary figures to explore how these literary representations because the purpose of this project is not really hunting monsters but to capture the essence of a specific culture and to enable a cultural analysis. How did people imagine a monsters to be like? Did they come from far away land or did they stem in our own soil? Did they move like the wind and disappear like the air or are they just stuck to one location? For future studies, analyzing oral histories would be interesting. But for now, I’m sticking to literature–the land of imagination filled with juicy layers of cultural perspective.

Included in the database are eighteen texts with thirty characters in total, fairly balanced between gender of the monster and author’s sex. We have seven male monsters by male authors, nine female monsters by male author, six male monsters by female author, and finally seven female monsters by female author. In addition to the gender of the monster axis, I added author’s sex mainly to test whether there’s any interesting pattern that comes out of it. Texts include Gothic literature, Ghost literature, and other horror literature that features in-between humans, animals, witches, etc. mostly from nineteenth-century literature.

text listNow, before we start hunting down these monsters, let’s first map out where they appear.

The actual monster map (dozens of them compiled for the convenience of our hunt)

As the esri Storymap is in-progress, I embed a video clip that walks through an ArcGIS map (desktop version). Start at 00:45. 

** As of September 1, 2021, the video is NOT working for the purpose of converting data into Esri Storymap. You can still watch it, but there’s no sound. 

Male monsters vs Female monsters

  • Features: Points are marked when a character is mentioned/indicated to be at such location, and lines demonstrate their travel trajectory. This line also allows the measurement of distance traveled. As will be shown later in the video, the krigging effect shows the level of monstrosity in a visualized manner. Simply, the redder the place is, the more monstrous a monster is.

First, we have a map that visualizes sightings and movements of male monsters by male authors. As you can see, they move globally and travel relatively large amount of distance. Part of the reason for the anxieties characters within these texts (and of course, for readers) is that they can never understand how monsters move so much distance at once. Combined with xenophobic fears in those eras, that they move about the world so effortlessly doubled their monstrosity.

The visualization is strikingly different for female monsters created by male authors. If male monsters by male authors showed large bounding of movements, female monsters by male authors show lesser degree of movement. The one long line that stands out here–which shows a movement from Paris to Damascus, Syria to Mauritas in Mauritania–is from Succubus. If we take this out as an outlier, the general pattern here is that female monsters almost never travel outside of her country. Their movement is well hidden from a large map and is almost invisible from this view (01:25). But if we zoom into the layer, we can see that female monsters actually move about but only to such a small degree that it’s not recognizable at the level of world-view map.

The problems of scale and what’s up with “measuring” monstrosity and sympathy?

As can be seen later in the video (03:00), the krigging effect shows the level of monstrosity. Like the travel journey of female monsters that couldn’t be shown in a large scale map, monstrosity of female monsters has to be zoomed up close to be seen. Because this map does not show the sequence of narrative it is not easy to see just by looking at the map what exactly happened or what made the monsters “red.” In the actual ArcGIS map, the users can click around each point to see the details of such information. The text title, gender of the monster and author, name of the character, the monster’s relation to protagonist, what happened at that location, etc. But even with this information, how did I measure monstrosity or sympathy? Isn’t that subjective response each reader has? How can you numerically tell the level of monstrosity? Is there a scale or a level for such thing?

Yes and No. To talk about the rubric, I first have to talk about how this project started in the first place. This monster map developed out of a group project that began back in 2016 lead by Dr. Ellen McCallum and Ed Schools. I had three undergraduate teammates under the name of “monster team.” Because we could not find a “canonical” monster level, we decided to come up with one for our map and made up a totally arbitrary one. Ex. If a monster murders an innocent, 5.0 on the monster level which is the highest. If a monster just stalks a character, 3.5 on the scale. Metamorphosis? maybe a 3.0. As we made up this rubric, we realized that we should have two categories at least. One for action and one for appearances, and we averaged the two to get a final monstrosity level (Again, questions arise: why 5:5 for action and appearance? Isn’t action more meaningful? Isn’t appearance what’s more significant in terms of defining a monster? Well… these are the questions we had to let go). Some of the description was hazy and we had to decide if “metamorphosis” is a monstrous action or appearance. But without a few exceptions like this, it was a quite clear categorization. Below is a screenshot of my monster rubric. On the left we have a scale from 0 to 5 each action or appearances attached to numerical number. Total 27 actions for monstrous action category from being just “creepy” to murders. Other actions include “dream attack 3.0,” “evil influence 3.0,” “intimidation 1.2,” “mind reading 3.0,” etc. Surprisingly, there aren’t whole lot of things a monster can do. Only about 27 types of actions. (Does this suggest Gothic/horror genre texts even in 18th and 19th century became formulaic?)

A screenshot of a monster scale that lists different actions/appearances of monster with an attached numerical value according to each action/appearance
A screenshot of a monster rubric that lists different actions/appearances of monster with an attached numerical value according to each action/appearance

Of course, doubts arise. How would you assign “intimidation” (1.2) as something less monstrous that “dream attack” (3.0)? Sometimes, “torture/imprison” (3.2) is worse than “mind control” (4.0). And this is where we step away from distance reading go into close reading, or at least, reading. What I decided I’d do is to read all of the text and check and adjusted the scale in order to make it more “universal” to all the text I had. I had to make some compromises to incorporate different levels of monstrosity. Sometimes, the description or the attitude of the narrator toward a monster who only “chases” (3.0) a character is so hostile that it is scarier than a matter-of-fact “murder of a stranger” (4.5).

And as I had doubts that maybe making up a monstrosity rubric wasn’t a good way to “distance read,” I came across a surprising phenomena. As a group, we all had similar scale set up. Every one of group member was working on different texts and thus, supposedly different rubrics. But when I went into read their text to change whatever was necessary on the rubric, I found that I had very little to challenge. Maybe the texts we chose were ones that leads to more general response. Maybe I was already prejudiced? I even developed my own rubric and compared the differences of the overall averages of monstrosity level but the differences were ignoble. Surprisingly so. Maybe we have this fantasy that each individual’s reading experience is so unique and subjective that we can’t come up with a universal response. Are we ignoring the fact that writers (sometimes…or usually?) intend to create a general response? Well, at the very least, a horror genre writer wouldn’t want the reader to find the text funny. A murder is supposed to be more dangerous and horrific than just killing a baby animal. How else do we have legal penalty system? But again, it is literature not laws. It could be argued that Mary Shelley has the Creature and Dr. Frankenstein chase each other for what seem like ages than just murdering confrontation because it would be more “monstrous” to have this constant influence over one another’s life. Some big theoretical questions remain in making the rubric…

Results: the gendered of monsters!

Mapping analysis result 1

Based on the distance analysis, the average distance of monsters have start difference according to gender. Female monsters traveled on average 423 km whereas male monsters traveled 4495 km. As expected from the visualization on the map, male monsters travel about ten times more than female monsters. On the other hand shows that female monsters show a slightly higher level of monstrosity. Considering that appearance and action assigned to 2.5 and 3.0  monstrosity have significant difference, this could be seen as a rather meaningful difference. On monstrosity-action scale, malicious attitude is counted as a 2.5 while 3.0 includes kidnapping, chasing, mind reading, accusing innocents, attacking innocents, dream attack, haunting, and imposing evil influence. For monstrosity-action scale, 3.0 monstrosity is described as one that’s not human but perceived as or disguised as human whereas one that’s human but little different or ugly is a 2.0 monster.

Mapping analysis result 2

It then seems then female monsters travel less but show higher level of monstrosity when male monsters travel much further but is perceived as less monstrous. Below graphs shows distance measurement and monstrosity level of monsters categorized by the gender of the monster and author (ex. MM: Male monster by Male author).

Mapping analysis result 3

What to make of these pattern?

One conclusion we can draw out of these result is that the types of horror male monsters and female monsters elicit are quite different. If male monsters threaten the globe infecting, killing, and haunting everyone, female monsters are local getting revenge to those around her.

One common feature of female monsters, which greatly heightens their level of monstrosity, is their sexuality. Although sexual aspect of male monsters such as vampires is seen as dangerous, this characteristic is not shared by many male monsters. For female monsters, they all emanate or evoke sexual energy and sometimes it is the only “evil” aspect of their monstrosity. Lois’ monstrosity mainly stems from the fact that she is beautiful and that she excites sexual desire in two religious men. And allegedly, the only wrong of succubus is her strangely excessive beauty that “shed[s] . . . joys around her” (312). Makala maintains that monsters–ghosts in particular–of nineteenth-century are an avenue for women writers to articulate their experience of ‘social terror’ against patriarchy (54). If eighteenth-century’s persecuted maidens traveled through labyrinths and dungeons of a debilitated castle and proved their physical strength and mental power (Moers 129), this heroinism is transferred into female monsters and ghosts. Strength and power of females that could not be expressed under normal circumstances of patriarchy was inherited by female monsters. Female writers place the placeless—monsters—sexual desire, frustration against patriarchy, injustices—in a much dramatic form of a “monster.” They do not have to travel much far to show their monstrosity. They stay within the bounds of the domestic house (or neighborhoods—extended “home”) brandishing their failure to be placed within existing society.

Spatial representation of monsters shows interesting differences between gender of the monster though it is complicated when the gender of the writer is included in the discussion. By taking a broader and higher perspective than the group mapping project, this project could discern meaningful patterns of monsters in terms of distance, locations, boundary of movements, and how these spatial elements are connected to characteristics of monstrosity and sympathy monsters demonstrate. The result of this project gives valuable support for humanistic approach of horror literatures validating the difference between gender of the monster and authors and their intention in creating such monsters.

Spatial analysis of this horror literature not only confirm traditional readings of texts but also brings insight to “outliers” that do not fit into the larger schema. Texts such as Frankenstein and Lifted Veil are interesting outliers of female works which have their monsters travel great distance with low-average monstrosity. Considering that Frankenstein is one of the most often referred text that brings controversy to the definition of “female gothic,” its unique spatial representation of the creature is another element that deserves further investigation.

We know their pattern… now what to do? How can I share it to the world?

Now that we’ve studied monster’s habitation pattern, it’ll be best if we can share this knowledge to the world. I’m developing a storymap on Esri website to publicize this map. As I used desktop version of ArcGIS, I first had to transfer my data to online ArcGIS account through which I can create layers of map that are translatable to storymap.

This turned out to be a difficult job. Since Online ArcGIS is like a free try-out version, it doesn’t support many of the complex analysis I had embedded in my map such as grouping layers ad krigging analysis. And because it doesn’t support grouping layers, I had to take individual feature in different export data. Say, I had “MM map” consisting of points, lines, krigging, minimal bounding of 8 different characters. It means I had 32 different features compiled in one map. In order to transfer that one map, I had to export each and every feature in different file so that online ArcGIS recognize that file. On top of that, because my meta data is so large (30 columns with more than 700 rows), online ArcGIS is keeps giving me bugs. As I mentioned before, my meta data includes text title, gender of the monster and author, name of the character, the monster’s relation to protagonist, what happened at that location, page number, duration, published date, any important things to note, etc. And when one hovers over or click each point, it should generate a information table that tells the audience what that point means. Without it, it’s just a point on a map. Unfortunately, Online ArcGIS is not allowing me to do that for some reason I still have to figure out.

spreadsheet metadata
A screenshot of metadata As you can see, each point has information about the character, page number, relation to the protagonist, level of monstrosity, monstrous action and appearance, etc. which unfortunately were not transferable to online ArcGIS at all…

Thus my failure. What is DH without years of failure, errors, and staring blankly at the dataset and coding logics! Besides, working with storymap was also fun. I put a lot of thought picking out style/format of my map and decided to pick one that is best to embed long essay with the map at the background. This style I chose also allows the audience to be interactive with the map zooming in and out to see where monsters are seen.

DH and Computational methods: it’s not the objective truth

Uncertainties, assumptions, compromises and continued challenges

  • Uncertainties/Assumptions: With mentioned compromises above, almost in every text, there are pointed locations that are based on assumptions. Moreover, the range of accuracy vary from point to point. For example, in Lois the Witch, I used two historical maps of Salem to indicated sites of prison, execution place, courts, etc., but locations of character’s houses were selected randomly out of several points of residents. The difference of scale should be noticed since there are large differences in detailedness of geographic information. Whereas texts like The Succubus gives out even street names, The Vampyre or “Tomb of Sarah” sets its characters in large settings such as “Rome,” or “Greece.” Fictional spaces were also mapped within a larger context of the text. Even when ‘Dwoldling,’ a set place for “Phantom Coach” turned out to be fictional, it was mapped according to the information that it was 12 miles from ‘Wyke,’ a moorland far north of England. Besides, the distance traveled is mainly conjectured out of a straight line connected between each point of location.
Map of Salem I used as a reference when I was mapping fictional locations in Lois the Witch. (http://salem.lib.virginia.edu/maps.html)
Map of Salem I used as a reference when I was mapping fictional locations in Lois the Witch. (Salem Map (University of Virginia)

I also had the conflict between my two selves–a humanities scholar who wants to challenge things and the author of this map who wants to visualize my map in the best way possible. I have my male monsters in green and black and female monsters in pink and red. I have to say I still doubt whether this was a smart choice. But I had to pick my battle. Since I am publishing this to a general public, I thought I’d use colors that best visualize the purpose of my map–compare and contrast between genders of monster.

Where to go now

I plan to publish this storymap online and polish it enough to enter it into storymap contest. I’ve looked through past year’s winner and looks like there’s no category for “fictional” maps. Usually it’s travel, history, migration, etc. with one honorary fictional map that a dad and 5 year-old girl made. I’ve also received feedback from my colleagues that many of them would like to use this as a pedagogical tool when they’re teaching. Not only for classes that read gothic novels but also for theory class as well. What does it mean that we don’t usually track down locations and look at maps when we read novel? Does it mean that readers are distanced from imagining those situations less realistically? Does that enhance a sense of horror? Or does it hinder it? This map doesn’t just “hunt” monsters. It’s a question mapped out: of genre, characterization, visualization, and above all, reading experience.

Special thanks to all the contributors and advisors who made this happen.

Thank you Monster group teammates, Dr. Ellen McCallum, and Ed Schools!

The Aged in History Past–Classroom assignment using digital tools or technology

What did people in Enlightenment era think of aged people? In an era when progress, human perfectibility, development, and above all, reason prevailed, how do contemporaries understood aged people who decline into physical or mental “decay”?

  • Summary of Assignment:

This assignment is designed to help you learn how to conduct a history data research in order to answer some of the questions we’ve been asking this semester about the representation of the aged and cultural understanding of old age, especially thinking in terms of the cognition and the concept of “selfhood” in Enlightenment era. First, you’ll learn firsthand how to search for materials in early modern era and how to effectively find sources you need out of hundreds of thousands of documents. Then you’ll try out a text mining tool to (maybe) pick out an interesting pattern in those documents.

  • Goal:
  1. Learn how to search for primary source
    1. To search for primary texts (ECCO, EBBO, Mind as Metaphor)
    2. To come up with fitting keywords
  2. Try out digital tools that incorporates an analysis (Voyant)
  3. Think about what you discovered in the connection to the idea of cognition and selfhood and ultimately what it says about cultural understanding of old age.
  • Assignment:

First, read Ch 3. The New Science: Aging and Agency, Age and Identity in Eighteenth-Century England  by Helen Yallop and summarize key points she makes about the agency of the aged in Eighteenth-Century Britain and select 10 vocabulary that epitomizes eighteenth-century contemporaries’ understanding of the aged (ex. definition, metaphor, etc: “old and poor,” debilitated, “idiocy,” “storehouse” of wisdom, “dependency,” “gout,” etc. ) Also mark three early modern texts mentioned in the book you find interesting or think most significant while reading.

Second, utilize the primary search digital tools such as ECCO or EBBO. Search for three texts and type in the keywords you found. What are the result? Are you seeing any pattern? Also, if you’d like, type in some keywords in Mind as Metaphor database. What are some synonyms or overarching concepts that appear?

Third, pick out about 1000 words from primary text that includes some of the keywords and run it through Voyant. What do you see there? What do these pattern challenge/support/complicate what we’ve been discussing so far? Does it match what Yallop is saying?

Finally, post one or two critical question using padlet.

 

EBBO (Early English Books Online)

ECCO (Eighteenth Century Collections Online)

Mind as a Metaphor

Voyant

padlet

 

 

  • Things I learned + Things could be incorporated

Padlet & Voice Thread

Different (fun!) adaptations (ex. poetry, drawings, comics, films, writings, screenplay, script, etc.)–explanation for chosen medium.

Wikipedia analysis: using voyant

Using movie script/letters/literary works and analyzing it to voyant

Twine game

Chronology time

Why important question!

My frustrating experience searching for aging-related primary source…

Since DH seminar this week is for moving images and I’m not one of those people who’ll be diving into that… I want to record my frustrating experience I am having for a few weeks now and also a few happy moments searching for aging-related primary text in eighteenth-century.

Age studies is a relatively new field having its first significant rise in the 70s and the intersection of age studies and literature has only began to take good shape. In addition to it still searching for possible texts for analysis, the fact that there has been a terminology change in how people expressed what we consider as now “aging” is giving me a difficult time finding just the right source. But of course, this is the challenge for anyone doing a historical research or anyone searching for primary text in general.

(Paul Sandy, ‘Wine Seller’ 1759: an apparently low-class old woman selling wine looks healthy and happy)

Since I was blocked by ECCO (Eighteenth Century Collection Online)–which said there is no interesting reference to aging in eighteenth century literature (I love you ECCO but I don’t trust you…), I started searching for university libraries that lists some databases for long eighteenth century primary source sites.

Memorial university libraries: early modern Britain – 15th to 18th C

Unique English Imprints, pre-1800: US National Library of Medicine

London Lives 1600 to 1800:Crime, Poverty and SocialPolicy in the Metropolis ☆

Early Modern Resources

British History Online

EuroDocs

Early English Books Online

USC Libraries

William Hogarth: Links to Online Image Archives of Paintings and Prints by Hogarth ☆

MSU

17th and 18th Century Burney Collection Newspapers  ☆

Gale News Vault ☆

Eighteenth-Century Journals Parts I through IV  ☆

Periodicals Archive Online ☆

Eighteenth Century Journals, Part I-V ☆

British Periodicals, Collections I and II ☆

British Newspapers 1600-1950  ☆

Empire Online

London Low Life  ☆

British History Sources 1500-1900 ☆

Early English Prose Fiction  ☆

Eighteenth-Century Fiction  ☆

English Drama

English Poetry Database

English Poetry (2nd edition)

Eighteenth Century Drama: Censorship, Society and the Stage ☆

Jane Austen’s Fiction Manuscripts ☆

Proceedings of the Old Bailey London 1674-1913 ☆

The Gazette ☆

English Historical Documents ☆

I hope there could be a search engine or a software program that enables me to type in a key word and pulls up all the record from various sites.

For spring break, I plan to visit these sites and to write down the summary of their usefulness and categorize them also. For now I have a little star next to the link to mark the sites that was useful.

Project Management

  • Revised project plan below

So… it’s time to get serious with my mapping project. Since it’s been more than a year I worked on monster map (and thus forget even what my data represented and how many texts are included in my data) and since I don’t know anything about esri story map (except a few great samples I found), let’s plan how I’ll go about publishing my map. I used workplan template provided by my DH course and it seems I only have three agenda:

  1. Review of data set
  2. Learn how to use story map (esri)
  3. Upload & Tweak around the map

The only dependency is that I’d have to be in the library to work on ArcGIS. I’d assume for now that uploading and revising how my map looks would take the longest (is it aesthetically pleasing? What text should go beside my map and in what order? What would be the most effective way to present the “story” map? And what am I ultimately trying to show?). Though the program and the result of story map looks simple and easy enough, from my last experience, I know it is not. Even though I only have three steps now, I’ll probably have to make a sub-mini-plans and schedule them in order finish this before the semester.

Revised Project Plan

 

“The Mechanic Muse”: Text Analysis and Distant Reading: Ted Underwood

It has been almost seven years since Moretti’s radical claim disturbed the humanities. The fundamentals of reading a text, “close-reading”–which I just taught my undergraduates last week and will continue to do so, is now being questioned. Given that it is humanities’ job to challenge whatever is the “fundamental,” it may not be so surprising that we are now questioning our own method of study. But this questioning is more than a challenge to the method; it is an examination of our approach to art and ultimately what we are trying to get out of this field.

I can’t tell for sure how much text a scholar must read but the comps list (comprehensive exam list: a list of texts you propose to read in order to write your doctoral dissertation) has around 150 works of literature (both primary and secondary). Say you are an avid reader and read around 150 during your BA and MA, and 150 more taking graduate seminars and for other responsibilities (including pleasure-reading). As a second-year PhD student about to compose my comps list, I’m around 450 (given that I already read the text for my comps exam). As NYT puts it, Moretti asks, “so what?” Even if you are an avid reader and read 1,000 texts, what about the other texts?

My immediate response would be it is not about the numbers of texts you read but about what each text represents and about how to exercise your own critical eyes on that text. Right? But again, even if I do not totally agree with Moretti, there is something insightful about the so-called “distant reading.” I do believe some concerns about quantitative approach to literature such as “how can numbers represent the “infinate variation of human perception” (Drucker, 2017)?” But they do provide an additional perspective. They do not have to be the alternative but another axis that humanities might consider.

I have recently been to a distant reading workshop here at MSU with Ted Underwood. His project involves machine learning which is basically a program which learns from examples rather than definitions and finds patters that might be meaningful to a looking eye. For example, his recent publication “The Life Cycles of Genres” investigates the development of genres using word frequency calculating program and finds patterns. When did the “Gothic” give way to, say (if), science fiction? In order to answer such question, one might start with defining what Gothic is. Despite decades of scholar’s effort, however, it is not easy to grasp what Gothic really is. Instead of finding a “definition” of this elusive genre, quantitative method finds a list of text, the corpus, and finds patterns that occur throughout. As Underwood admits, this might find things that affirm our previous findings if it does not come up with “novel” findings. But isn’t it what we “traditional” scholars do anyways? And at least, we will know, quantitatively, it is proved.

click to view Ted Underwood’s new publication “The Life Cycles of Genre” (Image from the URL below)

The most intriguing part about Underwood’s talk was how he accepts these models as they contain biases and utilizes that fact for his analysis. Like how would one corpus set developed by librarians as “science fiction” would differentiate the result from another data set by goodreads? This talk made me think maybe it isn’t just about the distant reading approach itself but the mere lack of study that is already out there that makes it seem troubling. The same would’ve gone for close-reading. If there is only a hundreds of scholars analyzing 19th century novel, their consensus, if there is one, about that time period would not mean much. If we have various data sets questioning or at least bringing in multiple assumptions/bias/social norms, it would be the start of looking at art in a meaningful way.

 

Related Links

New York Times: Distant Reading

Topic Modelling

DH and Visualization

Text and Images in DH

“DH is text heavy, visualization light, and simulation poor.” (Champion i25)

One takeaway from Champion’s writing is the definition and thus differentiation between model and simulation. One concern I constantly had for my mapping project is how my map represents “text” into visual. How would I translate the subtlety of written words–which enables different interpretations and mental visualizations (if possible)–and force it into a materialized visual forms? Building his argument on the symbiotic relation between image and text and furthering it with the need for “visualization literacy” (i27), Champion’s explanation for simulation made me realize that visualization will be approached in a similar fashion. It does not have to be a “model,” “a physical or digital representation of a product or process,” but a simulation, “the reconfigurative use of a model to reveal new and potential aspects of a model” (i27).  I thought of my mapping presenting or allowing only one selective aspect of a text even furthered by the fact the selection depends on the choices I make about which to visualize. But this differentiation between model and a simulation allows me to see that my maps will be a simulation that has the potential of another interpretation and that bears the thought of an embedded interpretation.


“[A] diagram or collection of data showing the spatial distribution of something or the relative positions of its components.” (Meirelles 115)

The definition of map alone supports Champion’s understanding of visualization of text material. Maps show the “relative” positions. Be it location, scale, movement, temporal relation, etc, mostly it is a selective visual encoding of data element that cannot be simply described as objective nor subjective. On page 127, Meirelles lists categorizes variables of the images–point, line, area, size, valuable, texture, color, symbols, etc. This also reminded me of Posner’s reading of digital humanities and how humanities scholars (should) challenge the grids of the fundamental. What would I achieve, and at what expense, if I chose to depict male monsters in pink and female monsters in black? Is my mapping project an experimental project that questions the “taken-for-granted” or public-accessed and tools for making the text be “seen” in a different medium?

! Another thing I found helpful was different hierarchical tree types of Cartesian systems. Perhaps use this for esri story map?

 

Reading:

 

Digitally Mapping Literature–what good can it do?

What is the advantage of a literary work that is digitally mapped?

This year for my DH project, I’m thinking about completing my mapping project I started exactly a year ago. I will be keeping my old data and add some new features as it can intersect with my research interest–i.e. age studies. Most of the work will involve transforming and exporting my data into a publishable data set. I will be using excel, ArcGIS, and mostly storymap tool. Storymap is an opensource tool, an interactive map sharing program. As I have engaged in ArcGIS tool before, I know it is an incredibly rich program–meaning that it involves learning complicated and oftentimes time-consuming techniques.

In addition to the technical honing and creating an aesthetically/functionally-pleasing storymap, I want to take time to think about what I can do with this opportunity. Last time I admit I was too concerned with the limitations of such project that I did not think of the gains of digital mappings. “Distant reading” as Moretti calls it or creating a mega-data (“mega” compared to close-reading one literary text as traditional literary critics do) containing more than 30 texts to compare and contrast, find patterns, pick out an outlier itself produces a different but no less valuable result. And I also want to take time to think about some of the doubts I had about my own project.

  • Are there any other ways to “measure” monstrosity or sympathy level? Is it valuable to measure such subjective literary effect? Are there other ways than numeric scale?
  • What techniques/tools/alternatives are there to project the uncertainties about the point of locality or distance range in the map?

Going back to creating a visual map, I also will be thinking about turning literary text into a visual text.

  • What will it mean for my viewer or for my own project that I pick the color pink for female monsters and black for more monstrous characters? Should I choose between visual-grasping i.e. “can be grasped at first glance” and challenging ingrained connotations of such connections?

Environmental scan: What other projects similar to mine are out there?

On official esri site, there are less than 10 storymaps under the category of literature or novel–the small amount of which is surprising. (Maybe I should find a better tag?) About half of them pinpoints the point of a character’s appearance on a map–official or manipulated–and have a text next to the map or over them as one scrolls down or clicks the point.

As I have tables and graphs that will compare/contrast monster’s distance and monstrosity by gender, I will not only use the map (bulleted) but also the scroll-down feature (side accordion) to explain the result of the comparative data.

 

(Marvel Superhero Origins, https://storymaps.esri.com/stories/2017/marvel-origins/)

 

I found three mapping project that interested me in esri. The first one above is a mapping of marvel hero and heroine’s origin. It is a rather simple story map that pinpoints the location of hero’s birth and action with a side accordion that explains each character’s name,birthplace, first appearance, ability, and origin story. The map is not so detailed to street name but the map has a visually corresponding aesthetic matching to the general theme. Another project, not listed on esri but on DHcommons and that uses ArcGIS storymap tool, is “Mapping at the Mountains of Madness” by Matt Mckinley. This mapping project is similar to mine in that it is mapping the locale of real as well as fictional places on earth. It maps locations detailed in Lovecraft’s novella “At the Mountains of Madness.” The map shows two different colors of points–one fictional and one real. The author uses “buffer” effect to speculate and demonstrate that the location is speculated. This project also, however, is not so detailed in its rationale as well as argumentation why one needs such map. 

 

(Mapping at the Mountains of Madnesshttps://www.arcgis.com/apps/MapSeries/index.html?appid=365b9c489db64fcd8486ae3533ed53c5)

 

One model storymap I found is “The Two Koreas.” Though this one does not concern any literature, it contains many features I think are adaptable to my project. First, it is enable in another language, Korean, as well as English which attracts more audience and makes the project shareable to larger spectrum. What was most attractive about this storymap is how it utilizes text with the visual. It explains very well what this project is about and why (though implicitly) such visual materials are integral for understanding the theme. It involves picture, moving maps, overlapping of maps, as well as a lineal historical line (this will be helpful for my project!), and graphs. It seems incredible amount of effort, work, and creativity were invested in this project.

 

(The Two Koreas: https://storymaps.esri.com/stories/2017/two-koreas/index.html)

 

Meaning of Digital Humanities and DH Scholarship

Miriam Posner, “What’s Next: The Radical, Unrealized Potential of Digital Humanities.”

I have to admit, I have been a long doubter of digital humanities before I was involved in DH projects–and maybe long after that. I come from another culture and the general “feels” about digital humanities in East Asia, or the critical institution that I was apart of, is very negative. It is not even “not positive” stage. In a country hosting Samsung and LG, benefiting with world’s one of the most developed technological practices, people in subway all looking down at their cellphones, it is maybe ironic. And this may be the reason more so for the humanities professionals to be against anything that combines digital with the humanities.

And before I say every doubt you have is legitimate as well as beginning to be answered, here are some of the questions I had: What do humanities have to do with anything digital? Aren’t we supposed to be persons who endorse reading?  Will it need the professionals in humanities who were trained to read, analyze, think critically, etc. to work in realms fueled by corporate-driven funding and computer coding? As I learned to realize, most of the doubts I had was due to lack of information or understanding of what DH is. And now it seems even embarrassing to write down the questions I had.

By chance and will, I’m in an institution where the university as well as faculty members are open and even eager (wow) to the promotion of DH. In my first semester as a doctorate student, I did a mapping project using ArcGIS program, visualizing how “monsters” and ghosts of 18th and 19th Century differ in terms of gender of the creature and authors. I pinpointed the locale of creature’s existence, sightings, the amount they traveled, the course they took, the monstrosity of their appearance as well as actions, etc. I used excel to make a data collection, did tons of calculations that I already forgot how to, used technology to visualize the intensity of atrocity of monsters. It was a frustrating experience largely because I had to learn how to master a program that I have not even heard existed but mainly because I had “doubts.”

As a person who were used to “close-reading” analysis, and who believed that was the “legitimate” way to “do” literature, counting word frequency, making a monstrosity level spectrum (i.e. murder being the highest, haunting being the lowest, for example) as well as empathy level spectrum, and above all, presenting literature as a spectacle one could understand at first glance was disturbing for me. (And perhaps the fact that I would show this to non-humanities-professionals and they say “it’s cool!” was discouraging.) Still, my data proved (and “showed”) that female monsters have less travel-distance (i.e. home-bound) and were more empathy-evoking. I concluded the project linking female monsters/ghosts demonstrating their fears and angers as well as limitations about domestic spheres they were confined to. Though my findings were wonderful and received great reviews, I still was hesitant. How can I judge the Creature from Frankenstein’s murder of a little boy is less monstrous than Count Dracula’s possession which will later lead to another killing? How can I assess a female ghost’s vindictive haunting to her failed lover is less monstrous than a male monster’s random killing?

Miriam Posner says, this is exactly what humanities in DH should be doing. Digital humanities is not just about using software, transferring data to an online collection, archiving, producing visual data that layperson can look and convinces themselves of granting you the funding. Our job is to ask questions, to tease out the assumptions, especially ideologically-driven ones, and to ask for change.

In the chapter above mentioned, she focuses her attention to gender and race. As a person who does Age Studies, I wonder the same argument could be made for age-category. Just like the distinction or the opting of black/white, male/female, we take the distinction of age granted. Does “senior years” start with 65? Why? Mostly because it is the time when pension begins not because of some inherent quality associated with the number. Oftentimes when I’m doing DH projects, I have the need to categorize humans, cultures, events, not only identifying them but also making inclusive-exhaustive definitions. Sometimes I am frustrated with this process because it is necessary to get the “result.” It is not so satisfying (not for me though) to have many unclassified/double/triple/or more overlaying elements in your data. And sometimes you have to make sacrifices. When I was using z-axis map, it incorrectly assumed words that carry negative connotation to be used to describe negative event/emotion. But without making some sacrifices, it seems meta-data is not possible. But is it? Posner suggests that we will get results and data, only that it might not conform to already existing social norms. If we add–or figure out a way to represent/categorize gender, we will not have how many women vs men have jobs but something radically different. It could be women vs men vs many other categorizes but also could be of different ramification. It might even further the idea of the futility of gender categorization–or the opposite.

I am not familiar with technology. I use my phone until it breaks, rarely install anything that doesn’t come with the laptop, doesn’t have a twitter account, just started trusting Starbucks phone app. And the talk of digital presence scares me still. But I don’t believe DH is about one’s preference. It’s about the responsibility as well as the ability of people in humanities to ask meaningful questions about a cultural phenomena and contributing to it with valuable doubts.

Mapping Monsters: Spatial Representations of “Monsters”

Result of data analysis for monster mapping

Example of Male monster minimal bounding

Mapping Monsters

Spatial Representations of “Monsters”: Gender Difference and Its Meaning

: This project was done as a part of a graduate course at Michigan State University. To open the map, ArcGIS is required. Web-based map can be viewed but this map does not allow any group-layer features. Screenshots of the map shows some snapshots of the map.

  • Description: This project concerns how gender affects spatial representation of so-called “monster” characters and how these representations are related to effects they produce such as levels of monstrosity and sympathy. Monsters that easily come to mind such as Dracula or the Creature from Frankenstein scare us readers not only with their evil doings and strange appearances. Part of their monstrosity lies also in their ability to escape human grasp. They move from one country to the other with the least effort and contaminate entire humanity unavailing our effort to pin them down to one place. However, there are other monsters who stay local and threaten relatively small number of people: female vampires who do not leave the castle, witches who demonize county people, ghosts bounded to houses, etc. The fear that we have for this type of monsters are quite different from wide-roving creatures. This project starts from a question whether this distinction stems from gender of the monster and looks to find a connection between distance traveled/areas of coverage and type of “monstrosity” monsters embody.

This map has largely three layers that allow above comparison analysis, each one including available points, lines, minimal bounding, and kriging: 1. Female monsters vs. male monsters 2. Female author vs. male author 3. MM vs. MF vs. FM vs. FF (gender of the monster/gender of the author (F: female, M: male), for example: MM means male monster created by male author).

  • Central Question: Does gender of the monster affect the distance traveled? How is the distance related to a level of monstrosity and level of sympathy?

Do female monsters show smaller trajectory and consequently lower level of monstrosity than male monsters? Do they have different habitation patterns than male monsters? How does spatial representation affect level of sympathy? Will the gender of the author complicate any pattern?

  • Explanation and Compromises
  • Corpus: Included in the database are eighteen texts with thirty characters in total, fairly balanced between gender of the monster and author’s sex. Texts include Gothic literature, Ghost literature, and other horror literature that features in-between humans, animals, witches, etc. mostly from nineteenth-century literature.
  • Features: Points are marked when a character is mentioned/indicated to be at such location, and lines demonstrate their travel trajectory. This line also allows the measurement of distance traveled. Minimal bounding, which is a smallest polygon that includes all the points of a character, is used to show and calculate square measure of each character’s coverage of area. Kriging analysis shows the different numeric number each character scored in that specific location and allows one to see at one glance at which point the highest numeric value exists. I used different symbols and lines to represent different category each element belongs to.
  • Uncertainties/Assumptions: Almost in every text, there are pointed locations that are based on assumptions and a range of accuracy vary from point to point. For example, in Lois the Witch, I used two historical maps of Salem to indicated sites of prison, execution place, courts, etc., but locations of character’s houses were selected randomly out of several points of residents. The difference of scale should be noticed since there are large differences in detailedness of geographic information. Whereas texts like The Succubus gives out even street names, The Vampyre or “Tomb of Sarah” sets its characters in large settings such as “Rome,” or “Greece.” Fictional spaces were also mapped within a larger context of the text. Even when ‘Dwoldling,’ a set place for “Phantom Coach” turned out to be fictional, it was mapped according to the information that it was 12 miles from ‘Wyke,’ a moorland far north of England.
  • Numeric Scale: To quantify a level of monstrosity and sympathy, I developed a monstrosity scale and sympathy scale.

Monstrosity scale measures how monstrous a character is displayed by each text. I collected every monstrous actions and appearances that is mentioned in a text and used that list to make a scale and ordered/ revised the numeric system considering the context of each. By making a corpus of monstrous action/appearance, I could overcome the difficulty of translating a subtlety of representation of literature to a numeric value. For instance, it could be said that ‘mind controlling’ is less ‘monstrous’ than ‘murder attempt’ at first glance and thus should score a high level of monstrosity. However, when it is taken within the context of the text, say, it leads to a cruel victimization of an innocent person or is described in such a detail that it leads to horrify its reader, these contexts cannot be overlooked by ‘objective’ measures. This is not to say this scale is objective. When considering the ‘context,’ cartographer’s own interpretation of the situation and emotions each display of monstrosity is eliciting are embodied. However, it can be said that it is more consistent and coherent as a whole.

Sympathy scale measures how a character is viewed by other characters and by the narrator. Having a numeric value for sympathy proved to be very challenging. Sometimes, characters differ in their view of the monster characters and the narrator describes the monster in a subtly sympathetic way even when s/he is received as not sympathy-worthy. There are also texts that evoke sympathy in readers not through other character or through narration but with plot itself. It is less uncomfortable for readers to feel sympathy toward a ‘monster’ who haunts her murderer or who hurts one who killed her baby. I included these different levels of sympathy each text arouses in readers to determine how a text depicts monstrosity.

  • Limitations

Even when there’s a difference, features that did not have enough numeric value—male monsters’ monstrosity, sympathy scale for both male and female monsters—could not be demonstrated using kriging tool.

  • Answers from the map:
    1. Distance-Monstrosity

Male monsters travel ten times more than female monsters on average even though female monsters have higher level of monstrosity. This shows that the nature of their monstrosity as well as horror monsters elicit differ by gender.

 

2. MM, MF, FM, FF

Although there is a stark difference between male monster and female monster, adding author’s sex in the comparison enables much further analysis. Although female monsters created by female author show longer distance traveled, minimal bounding proves that they move about the same place, thus has higher numeric value for distance but has limited roving.

 

Tag: #monster #gender #horror #ghost #witch #literature #sex

http://www.arcgis.com/home/item.html?id=97826243075245579beab18a21e82a37

 

Screenshots:

Male monster minimal bounding

female monster points and lines

female monster minimal bounding

MF points and lines

FM points and lines

FF points and lines

female monster krigging

female monster krigging zoomed