2 0 0 6
Demonstration of the MINERVA System for Study of Literary Texts
(originally presented at the first Colloquium on Digital Humanities at the University of Chicago in November, 2006)
- Sample screen images
THE MINERVA SYSTEM FOR STUDY OF LITERARY TEXTS
Dr. Cora Angier Sowa
The Minerva System is a set of tools, partly automated and partly non-automated, for planning and carrying out a project in literary study. Techniques of systems analysis, borrowed from the scientific and business world, are adapted to analyzing the insights of traditional belles-lettres literary criticism. A project is defined as an enterprise that has a goal and an organized way of achieving it.
In recent years, many databases of literary materials have been assembled, some containing just texts, others having pictures, topographic maps, film clips, etc. Programs have been developed for using these databases, but mostly to search for information in them. Less attention is paid to just what will be done with all the data that has been collected. This contrasts with work done by earlier humanists using computers, especially in the 1960's, who started with a real problem that they wanted to solve, then tried to find a way that the computer could help them. Also too little discussed today is the relationship between technology and humanities. What do they have to do with each other?
My contribution to the field is called the Minerva System, named after Minerva (Athena), who was goddess both of wisdom and of skilled crafts. It is from the same desire to know, to appreciate, and to shape that the humanities and technology both arise. To quote Henry Petroski, To Engineer is Human, "The work of the engineer is not unlike that of a writer. How the original design for a new bridge comes to be may involve as great a leap of the imagination as the first draft of a novel."1
The Minerva System:
The Minerva System is a set of tools for planning and carrying out a project in literary study. Methods of Systems Analysis, emphasizing diagramming techniques and modularity, are used. These are techniques for taking a large, vague question and breaking it down into smaller pieces that can be worked on separately. Top-down methods are emphasized, but bottom-up methods are also used, for example in writing down a group of ideas and letting them coalesce into a pattern. Problems should, however, always be grounded in literary (and historical and social) concerns, such as beauty, sublimity, and social importance (or scariness and boringness!) which alone confer significance (or lack of it) on any endeavor. Any scientist will tell you that these are as important for the scientist as for the humanist. The trick lies in analyzing the language of criticism, finding out how vague concepts can be expressed in precise language, and identifying what is quantifiable.
The Minerva System consists of three elements: (1) The framework of steps to follow in carrying out a project, e.g. descriptive paragraph, worksheet of key words, hierarchical diagram of modules, flow charts, program to be used, synthesis of computable and uncomputable parts, and evaluation of results. These steps are partially automated; tools are provided to aid the user in the form of the Minerva Project Planner. (2) The Minerva Program Suite, a set of programs that can be used as a stand-alone system, that were designed using the above methods, created out of interchangeable, reusable subroutines. The Minerva programs currently include programs to make concordances, search for words and cooccurring words, perform simple statistical tests on the distribution of grammatical categories, "compose" original compositions, and look for cooccurring clusters of words. A set of "OwlData" programs aid the user in preparing his or her own data to use with the programs. (3) The CD The Loom of Minerva. This CD is a self-study version of classes that I have taught at various colleges, especially at St. John's University in Queens, NY. The CD includes both text chapters and the suite of Minerva/OwlData programs. The chapters include a discussion of the Minerva System, as well as historical chapters describing work done using mechanized aids in the study of literature, and a manual of Visual Basic using modular methods, for those who want to play with the programs, for which source code is provided. The first chapter of the text, A Guide to the Labyrinth, can be read on my Web site at www.minervaclassics.com/loom1.htm.
The discussion of each program in The Loom of Minerva is introduced with one or more pieces of literary criticism (the more romantic and apparently uncomputable the better!). Examples are a case study of Coleridge's "The Rime of the Ancient Mariner" (with criticism by Walter Pater and others) and an analysis of Baudelaire's Les Fleurs du Mal (with criticism by Sainte-Beuve). The study of Coleridge leads to a demonstration of the program LOOKUP, which tabulates the frequency of chosen categories of words (e.g. "sea," "supernatural animals," "revenge," "forgiveness") in the poem. The study of Baudelaire leads to a demonstration of CATMAP, which is used to create a visual map of the progress from "beautiful" words at the beginning of the poem "A celle qui est trop gaie" to "ugly" ones at the end. A program called COMPOSE "composes" sentences "in the style of" an author like Shakespeare or Chomsky, demonstrating how such a program can test rules of grammar and syntax.
Advantages of the Minerva System -- requires no special resources:
The Minerva System does not use data that is in a proprietary format. It uses plain ASCII text, such as that which can be downloaded from the Internet. The OwlData programs can be used to reformat the text into the format used by the Minerva programs. The Minerva programs do not use sophisticated mathematics or linguistics. One of my aims is to show scholars -- and especially to show students -- that the fundamentals of computer work can be understood without having such specialized knowledge.
The Minerva System is designed to be extensible. New modules can be added to it without disturbing the whole. The latest program that I have added is the CLUMPS program to find clusters of cooccurring word-stems. The original version of CLUMPS, written in PL/I, was used in a study of the Homeric Hymns by myself and John Sowa, published as "Thought Clusters in Early Greek Oral Poetry," in Computers and the Humanities, May 1974.2 It is now available on my Web site, at www.minervaclassics.com/clumps.htm. Joseph Raben, of Queens College, NY, is using the CLUMPS program to further his project to study the poetry of Shelley. For any programmer who wants to add new programs to the Minerva System or modify existing programs, I only ask that I be credited with being the original author of the programs and system.
- Henry Petroski, To Engineer is Human: The Role of Failure in Successful Design, New York: St. Martin's Press, 1985, p. 78.
- Cora Angier Sowa and John F. Sowa, "Thought Clusters in Early Greek Oral Poetry," CHum 8:3 (May, 1974), pp. 131-146. Results of this research were incorporated into a larger study of the Homeric Hymns in my book Traditional Themes and the Homeric Hymns, Mundelein, IL: Bolchazy-Carducci, 1984.
2. Sample screen images:
TO SEE THE LATEST SCREEN IMAGES, SEE THE ILLUSTRATIONS IN CHAPTER 1 OF THE LOOM OF MINERVA, "A GUIDE TO THE LABYRINTH, WHAT YOU WILL SEE WHEN YOU RUN MINERVA."
All selections on this site, unless otherwise identified, are copyright by Cora Angier Sowa.
Send e-mail to Cora Angier Sowa.
On this Web site, you are visitor number: