Blog #2 Post Prompt – Revised

Blog #2: “On Distant Reading”
Due: 9/24

In this blog post you are going to use the full compiled transcription of the Linn diary.

What is the value of distant reading? How can it help us to examine a text and illuminate aspects of that text that might not be so clear to us through traditional reading methods? They invite us to play with texts in a way we have not been able to before. Despite the temptation to dismiss this play as trivial, many critics see visualization as an interface in an iterative process of play; as part of the process of problem-generating. So what do you think? DOES distant reading (and in this case textual visualization) help us to examine a text in a meaningful way? When you look at a textual visualization – a Cirrus (or word cloud), a network of names, a graph that shows word frequencies – what questions do you start to ask about that text? What connections do you see? Remember that while the computer generates patterns, the computer cannot ask the questions. You must determine what those patterns mean.

After the battle at Roanoke Island, Linn makes this observation in his diary: “War is horrible. I first saw the pomp & circumstance – the battle field – the dead and wounded now the prison ship.” (Tuesday, February 18)

Jakacki hypothesis: I would propose that this short statement, coming roughly halfway through our text, demonstrates a profound shift in James’s perception and a loss of innocence.

Prompt: Can a distant reading of the text confirm or refute my hypothesis?

Example: I’m going to start by giving you an example: I ran the Linn Diary text file through Voyant: (http://voyant-tools.org/) as per the instructions I gave you last week. The word in the Cirrus that drew my eye was  “company” (I don’t know why), and so I clicked on it and this is what the dashboard revealed to me: Screen Shot 2014-09-23 at 6.28.52 PM

The word “company” appears 68 times in the text. Looking at the corpus reader, I see that there are places in the text where the word appears more frequently. Looking at the Word Trends panel, I see that there are two spikes in its usage: in segment 6 and 9. When I click on the Word Trends graph at points 6 and 9, I see that those spikes occur at February 6 and March 27. Perhaps equally important (or confusing), there is a sharp drop in the use of the word “company” just before March 27.

So can I apply the word “company” and come away with a definitive yes, or no to the research question? I don’t think so. I’m disappointed. I thought I would have an “ah ha” moment and be able to show that yes, the word patterns show me that my hypothesis is correct, that a word chosen at random will support my perspective on Linn’s development as a person.

But I’m determined. What happens if I add a second word to see if I can adjust that pattern?

I choose the word “wounded” (by typing it into the “search” bar below the Word Trends graph.) Now I see this:

Screen Shot 2014-09-23 at 6.40.13 PM

Focusing on the Word Trends panel, I now see that there are some correlations between Linn’s use of the words company and wounded. The word wounded appears only 31 times in the text, and seems to follow along the same trajectory as the word company in the second half of the text. Does the combination of these words tell me definitively that Linn loses his innocence after the battle of Roanoke? I’m not sure. What it does is encourage me to ask a new question: why does Linn write these two words more frequently in the second half of the narrative? What is the correlation between company (his comrades, his most immediate fellow soldiers) and wounded (the violence that Linn saw happening to around him during the two battles)?

So for your post, I would like you to do the same thing. Follow the same process I did above, but choose your own word from the Cirrus. Once you’ve examined that word in the various panels, look for a second word that you think might correspond to (or contrast with) the first.

What do the words that you’ve chosen reveal about Linn’s experience in battle? Do they reveal anything at all? Remember that you may find that your words help you to support the hypothesis I have presented (in which case you should explain why you think they do) or that they help you to refute the hypothesis I have presented (in which case you should explain why you think my hypothesis is faulty.) You may also discover that my question ultimately serves only to lead you to one that you think is more important (in which case you should explain what that new research question is.) You may conclude that this form of textual analysis is helpful when asking these questions, or that it doesn’t help you to consider the questions that *you* want to ask about Linn and his diary.

Because this is different and we’re all experimenting with how to make sense of this, I encourage you to look at one another’s posts to see how they are approaching the question. If you’re interested, take a look at what Professor Faull’s students are thinking about in their blog posts for her section of HUMN 100. They may have considered some things about the relative benefit of distant reading that you might want to adapt for your own thinking.

Reminder: give your post the category “Blog #2.” Add two (2) screen images – similar to the ones I’ve shown above. Include five tags that you think reflect the work that you submit. Please review the Blog Assignment description and rubric and email me with any questions you may have.

Experimenting with text at 30,000 feet

In class on Wednesday we discussed how distant reading approaches can offer us as we analyze texts. I showed you a quick demonstration of Voyant Tools and how different visualizations encourage you to tease out different aspects of a large text or “corpus” (a collection of texts).

Now it’s your turn to experiment in preparation for submission of Blog Post #2 “On Distant Reading” (due Tuesday 9/23 by 11pm).

Instructions:

  • Go to the Voyant Tools home page: http://voyant-tools.org/
  • Upload the file of the Linn diary draft compilation (the file I emailed to you on Tuesday 9/16) using the “Upload” button, then “Reveal”
  • You should now see a dashboard with several panels. In the top left panel is a “Cirrus” or word cloud. You’ll notice that the most prominent words are conjunctions and pronouns (the, and, of, we, they, etc.) To get at the meatier words, you’re going to want to create a “Stop Words List.” Click on the gear button (see screenshot) and in Options panel pulldown menu choose “English (Taporware)” and tick the box “Apply Stop Words Globally.” Click “OK.”

Voyant1

 

 

 

 

 

 

Choose different words that you find compelling (times of day, military terms, etc.) Make a list of five words that you want to look at further. What happens when you look at the words in the Corpus panel? The Word Trends panel? What else appears to you? What can you learn about word frequencies and words that appear in relation to others that you might not “see” so readily when you’re reading in more traditional methods?

Now you’re going to experiment with different types of visualization.

  • Go to the Voyant Tools Index: http://docs.voyant-tools.org/tools/
  • Upload your text to Links, RezoViz, and ScatterPlot (note that you’ll have to do this for each visualization tool you choose).
  • Go through the Stop Words List application again.
  • What does each tool show you in terms of the text? What word associations do you discover? How can you manipulate those associations?
  • Keep in mind that there is more to these visualizations than just pretty pictures. They help the researcher to ask questions about a text that might not otherwise come to mind.

Try the same experiment with the diary entry you transcribed. Granted, this is a much smaller body of text, but can you see any words or names that are important in your text AND in the overall diary text?

On Monday we’ll be experimenting with more complex text visualization tools.

 

 

 

Week Four Assignments, Readings, Exercises

Monday 9/22

  • Reading: Complete reading transcribed Linn diary
  • Discussion: keywords from distant reading as metadata
  • Lab: More complex analytical/visualization tools (cross-corpus analysis)

Tuesday 9/23

Wednesday 9/24

  • Discussion: Distant reading & blog assignment wrap-up

Blog post #2 due (11pm)

Friday 9/26

  • Reading: Watch “Secrets of the Dead: The Lost Diary of Dr. Livingstone” (available to stream on our course Moodle site)
  • Timemapper exercise introduced
  • Research topics chosen

Week Three Assignments, Readings, Exercises

Monday 9/15

  • Share out transcriptions – group help session
  • Transcriptions due via email (11pm)
  • Reflections due via website (11pm)

Tuesday 9/18

Full diary (rough) transcription file distributed to class (via email)

Wednesday 9/17

  • Reading: Whitley article
  • Lecture: Distant reading introduction (including Whitley)
  • Lab: Distant reading analytical/visualization tools
    • begin to form questions about diary text for distant reading blog post

Friday 9/19

  • Reading: Diary
  • Meeting with Nancy Frazier, research librarian

Mapping Operation War Diary – Bringing the Battlefields of Yesterday to Life | Operation War Diary

More and more frequently interesting new Digital Humanities projects are discussed in the media. This morning, Professor Faull sent me a link to this article. Here is a project called “Operation War Diary” that is using GIS mapping (with which we’ll be working later in the course) to assess how much the landscapes described in the war diaries have changed in the 100 years since they were written.” Look at the bottom of the article: see how the team researching this project ask readers to help them determine which might be the best approach to doing this type of work.

Mapping Operation War Diary – Bringing the Battlefields of Yesterday to Life | Operation War Diary.