Prelude to a Proposal: As both Cecelia and Gwen have indicated, our conversations from the last week have been marked by a lot of confusion, tension, and doubt about our project and goals. In an attempt to bring some clarity back, we are each posting a potential project for Prism this week. As I sat down to write about one such project, I began by reading through project proposals of some of my fellow team members (or Praxisers?). Given the frustration of the past week, I was surprised to discover that there is a notable amount of agreement over our concerns about Prism and possible directions that our project could take. As I walk you through my proposal, I will touch on what I see as some emerging themes.
Apprehensions about a Prism Proposal: I am extremely interested in how people make meaning from a text. However, I worry that deploying Prism in its present form (or a form that only allows for the highlighting of certain categories) might not allow for data collection that captures the process of meaning making. Like Gwen, I share a concern about the interpretation of text without a consideration of the text as a whole. In order to get at the polysemic nature of a text, we need to know more than how individuals interpret specific words or phrases. We need to know what message the reader pulls out of the text as a whole. In order to apply a more structural approach and to understand how the internal relationships within a text generate meaning, we would need to be able to mark more than categories. We would also need to indicate how various parts of the text relate to each other. Despite these concerns, I offer a proposal which could be conducted within the current Prism premise by including a few additions to the processes of analysis and visualization. Though the particular text is not important, I have included a small excerpt from a wedding ceremony as a test case.
A Question for Prism: In cultural studies, we often talk about the way cultural references come bundled together. We can think of this in two ways. In the first, the elective affinity between two theoretically distinct worldviews causes the two to become intertwined (for example, scholars often discuss the way in which Protestantism and capitalism have a shared ethic of work, allowing for an easy expansion of capitalism in the protestant world). However, bundles also exist within the minds of individuals. People can hold associations in their heads between beliefs and concepts which are not necessarily connected in practice. For example, when someone reads the word “capitalism,” it may also signal concepts such as “the West,” “democracy,” “freedom,” or “exploitation.” In another example, the word “love,” may bring concepts such as “marriage,” “commitment,” “fate,” or “betrayal” to mind. One use for Prism would be to assist scholars in identifying these bundles and figuring out how they frame the interpretive process.
Two-Stepping: In order to do this, I think there would need to be a two-step process. Like other Praxisers, I am concerned about imposing my own categories upon the interpretive process. So, I suggest that specific Prism projects would benefit from a more open stage. However, as our transparency activity from last week demonstrated, completely free form interpretation failed to produce anything sensible in the aggregate. To solve this problem, I would introduce some constraints, asking users to highlight phrases and label them with a single word of their choosing. Either a researcher or a semantic linguistic program would then work to develop a limited number of categories relevant to the text. In stage two, the same text would be posted for markup again with the user-generated categories. This process mimics a well-established interpretive processed used in the social sciences (called Grounded Theory Coding: see Charmaz and Strauss and Glazer). However, a tool like Prism allows for an innovative modification: the initial categories to be developed by users (those who actually make meaning), rather than the researcher.
Analysis and Visuals: As indicated by several other Praxisers, I think it would be crucial to keep each user’s interpretation autonomous. Although I think some sort of aggregate interpretation is important, it would also be necessary to be able to sort interpretations according to users. In order to see how cultural bundles might influence meaning, I would want to be able to answer two key questions. What is the actual content of the cultural bundles that influence the interpretation of this text? Can a user’s interpretation of a particular phrase indicate how that user will interpret subsequent phrases? Answering these questions would require that I can sort the aggregate interpretation to show only certain users. For example, I might want to see a visualization of all the users’ interpretations who marked a particular phrase with a particular category. Did all those users who indicate that “one true love” is about fate mark the text in similar ways? Do they tend to make the same interpretation of the phrase “lawfully wedded?” Are there patterns that let indicate the specific content of cultural bundles related to love and marriage?
I, (Bride’s Name), take you, (Groom’s Name), to be my lawfully wedded husband, secure in the knowledge that you will be my constant friend, my faithful partner in life, and my one true love.