Corpus-based Latin Word Comparison

This exercise attempts to show students that words do not have fixed meaning, but rather are contingent on their usage and context. It teaches students how to identify "standard" majority meanings as well as deviations from these norms and their relative unusualness. Finally, it teaches students how to compare individual words using corpus-based searches across large collections of texts.

This exercise would be most appropriate for small to medium classes of advanced language students (e.g., those who have a firm grasp on grammar and vocabulary but need to develop a better understanding of nuance). It could be completed in a short time (a 15-minute demonstration in class to show how to use the search tools and identify changes in words, along with one or two days outside of class). Students will need computer access to the Perseus site, Thesaurus Linguae Latinae, standard lexica, and indices verborum. This could be adapted for any Greek or Latin class at an intermediate/advanced level.

1) Students will select one word from a text we are reading.
Example: "deliciae" in Catullus poem 2.1

2) Students will look up the various meanings the word can have using the Thesaurus Linguae Latinae, the Oxford Latin Dictionary, and commentaries on their authors.

3) Students will search the rest of Catullus's work for instances of that word in all its forms using an index verborum or the Perseus search tools: http://www.perseus.tufts.edu/hopper/search
Example: word appears at poems 3.4, 6.1, 32.2, 45.24, 68a.26, 69.4, 74.2

4) Students will compare the usage in each case by comparing context (e.g., genre, surrounding narrative, modifiers, etc.). Students will keep track of how often the word appears in a given context.

Example: "deliciae" is used of women in Catullus twice, of a friend once, of an object three times, of a dead individual three times, etc.

5) Students will perform the same search for one or two related authors.

6) Students will compare what seems to be the standard meaning in their author, in related authors, and in the reference books and identify any discrepancies. They will also examine their specific instance of the word and explain why it might be the same as/different from the standard meaning.

Students should avoid words that appear too frequently (e.g., conjunctions or the verb "to be"). Teachers will also have to define what counts as a "related author."

Assessment of this exercise can be performed by examining the students' results in step 6 and determining whether their analysis is confirmed by the evidence they found in their corpus searches.

http://www.perseus.tufts.edu/hopper/search