How does one write a research paper using a corpus analysis involving Martin Luther King's "I Have a Dream" speech?
A corpus is a collection of "thousands or millions of words" that are considered to be authentic samples of the native speaker's language usage (Muller, F., & Waibel, B., "Corpus Linguistics--An Introduction," University of Freiburg). The samples are stored in electronic databases for analysis, and analysis is usually done through computer software (Muller & Waibel). Hence, corpus linguistics can be defined as "the study of language" by analyzing these "naturally occurring language samples" (Muller & Waibel). One purpose of corpus linguistics is to monitor the evolution of "specific features in the history of English" (Muller & Waibel). Muller and Waibel gives us the development of the words "gonna or wanna" in the English language as an example.
The very first step to writing a research paper in which you conduct a corpus analysis is to decide exactly what you want to discover through your analysis, in other words, to figure out your research question. If one were to use Martin Luther King's "I Have a Dream" speech, one might use corpus linguistics to study the progression of the African American variety of English, called Ebonics by linguists, and compare the progress of Ebonics to the language King uses in his own speech. In doing so, you might find similarities and differences between the typical Ebonics variety and King's own language usage. You can then compare his language usage to the language usage of other African Americans of his time period, especially of the educated class, to see if his own language usage is an emerging pattern.
Once you have determined exactly what you want to figure out through your corpus analysis, you next have to conduct corpus linguistics research in the correct database. The Corpus of Contemporary American English looks like one of the best existing databases for varieties of American English. Another possibility may be the Freiburg English Dialect Corpus, and you may need to use your school library to further research corpora databases. Multiple corpora databases exist based on research need. Therefore, depending on exactly what research question you decide you want to answer through your corpus analysis, you may need to look at other popular corpora databases such ARCHER and Helsinki, both of which offer good historical corpora, dividing corpus texts by historical time periods.
Once you have figured out your research question, gathered your data, and analyzed your data, you should be ready to write your thesis, outline a draft of your argument, and write your research paper.