01/04/94
CONTACT: Stanford University News Service (650) 723-2558
STANFORD -- Sometimes scholars need to know what an author chooses to mean when a word is used. The Academic Text Service offered by Stanford's Libraries and Information Resources now includes a tool called Searcher, a program that allows users to search a database of texts for occurrences of certain words, a way of determining both the history and the meaning.
The program is useful for investigating the nuances of how words were employed - how often and in what context - helping humanists to study imagery, symbolism or connotation.
Does the use of the words for colors denote symbolism in The Lord of the Flies? When an author writes "The King likes pears," does it make a difference that the word "like" can have more than one meaning?
Previously, such searches - some done by hand - took a long time, and many were impractical, even impossible. The Searcher program allows researchers to find what they want very quickly and with more flexibility than any previous tool.
"Major texts have had concordances [indexes of words] for many years. Even in those cases, though, you don't have the kind of versatility of search procedure that you have in Searcher," said Philip Ivanhoe, assistant professor of philosophy.
"The software in the Searcher will allow you to define searches in a way that precludes your needing to go through an initial search and then picking out a subset of those search results, and so on, with more definition. It allows you to be very intelligent with the first pass."
The program was written at Stanford and has been licensed elsewhere. It works on Macintosh computers and is accessible at the Stanford library or for professors and students from their offices or dorms.
One graduate student is researching the history of the word "virgin." Another is studying the history of the phrase "to be about to," while a third is studying "hot news" forms of "have," as in "The Titanic has sunk!"
A search begins by selecting a work from the program's "Bookshelf," the database containing more than 70 texts. Searcher then opens a window that allows the user to instruct the program. The basic command is FIND, which prompts the user for one word. Searcher then locates all the occurrences of that word and opens a new window to display them.
Jumping to that window allows the user to scroll through the results. Each word appears in 2,000 characters of its surrounding context and with a location identifier such as page number or act or scene numbers.
FORWARD and BACKWARD buttons will scroll through additional sets of 2,000 characters to help clarify the context.
Fancier searches are easy. The FIND OR command will hunt for up to five words at a time, and more are accommodated by using a synonym list. A prefix search easily finds all forms of a word: the prefix "lik" will find "like," "likes" and "liketh." Searcher also hunts for suffixes.
Features such as "Within Context" can be used with most plays to find out if a certain character says the word. The feature "Near" looks for near-word phrases - such as "if . . . then" - and allows the user to specify how far from the first word to look for the second word. It is also possible to restrict the word list to words that occur only a certain number of times.
The most popular feature to manage results is the concordance.
The "Key Word in Context-Style" lists the location followed by the word in 60 characters of surrounding context. The "Book Style" concordance lists only the location. For more context, an archive file can be created. There is also a graphing feature that displays the distribution of matches throughout the text.
The texts available for search include historical documents, plays, novels, poetry, philosophical treatises and the Old English corpora, the basic tools for scholars in medieval studies. Most of the texts are in English. More texts can be added easily, either by locating them in electronic form or by scanning them in.
"The range of the corpora is increasing steadily," says Elizabeth Traugott, professor of linguistics. "The most exciting new one is the Helsinki Diachronic Corpus of the History of English, with one and a half million words from the history of English from 700 to 1710, organized by period and sample genres from each period."
Several of her students used it last quarter, she said.
Traugott also sees great potential value in being able to test research ideas with the Searcher.
"It's easy to see if your project is going to be viable," she said. "You can find out in a couple of hours with the Searcher. If you're clear on what you're looking for and just don't find [it], then you can decide, well, I'd like to write a paper on why it isn't there. That's OK. Or you can say, well, I don't want to bother with that since it isn't there; I want to find something else."
Ivanhoe has found the Searcher particularly useful to students unfamiliar with the works.
"It really opens up the text to the reader to a level of familiarity that used to require years and years of reading it over and over again to know where something is," he said. "Now people can very quickly get a handle on the text."
Despite such advantages, he cautions that the classical approach to close textual study has not been replaced by the Searcher.
"It augments it, it doesn't replace it," Ivanhoe said. "There is a difference between being able to find all the occurrences of 'benevolence,' say, in the text, and having them in your head, along with other things, because then you can make associations that you can't in the other way as easily."
-jns-
940104Arc4510.html
This is an archived release.
This release is not available in any other form.
Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images,
some of which may be available to you online.
Direct your request by EMail to newslibrary@stanford.edu.