In Ronald White's 2005 New York Times editorial he describes how the inaugural address for presidents' second terms tend to have more "I" language and less "we" language. In the book The Eloquent President: A Portrait of Lincoln Through His Words Ronald White writes:
Each of these five presidents began his inaugural address with a reference to the “confidence” of their fellow citizens. One cannot miss, however, the self-referential quality that was advanced by many different personal pronouns in each of their opening words. There was a steady staccato of “I” in all of these inaugural addresses.
Except for Lincoln’s. The opening words in his Second Inaugural are all the more remarkable when heard against the backdrop of the five previous second inaugural addresses. He nowhere speaks of the “confidence” or “approbation” of the electorate. He uses personal pronouns only twice, I and myself, in the first paragraph, and never again. Because we approach the Second Inaugural through the larger lens of his previous speeches, we should not be so surprised. He never spoke of himself, did not use one personal pronoun, in the Gettysburg address.
Others have similarly commented on the use of personal pronouns in speeches, such as Stanley Fish's comments on Obama's use of "I" vs. "we", along with Mark Liberman's challenge of those assertions.
For this program we will be:
Gathering data
Writing and using tools to analyze all the presidential inaugural addresses
Providing a description of our results and discoveries
You will turn in a directory called analysis which will contain these three components. You must zip this up before submitting, so the zipped version will be called analysis.zip. Each of the components is discussed in turn below. If you do not turn in your materials in the structure described below you will get a grade of 0.
Gathering data
All the inaugural addresses can be found at the UC Santa Barbara American Presidency Project where word counts are included as well. You may want to store these as separate files, or you may want to merge them into one big file. You may want to add information on date and political affiliation (Democrat, Republican, ...). You may want to refer to Bartleby's background information for inaugurals, or the list of "firsts" at Inaugural ceremonies.
The data files must be in a subdirectory called datafiles, whether there is one file or multiple files. While you may not share your programs, I don't mind if you share your datafiles with each other.
Writing a tool to use in analyzing inaugural addresses
You must write a program called analyze.cpp (which should be directly inside the analysis directory) to provide the items shown below for each of the 56 inaugural addresses:
(5 points) Word count
(5 points) Sentence count
(5 points) Paragraph count
(5 points) Average sentence length (in words)
(5 points) Average word length
(10 points) Number of first-person singular pronouns (I, I’ve, I’ll, me, my, mine, myself )
(10 points) Number of first-person plural pronouns (we, we’ve, we’ll, us, our, ours, ourselves)
(5 points) Ratio of first-person singular pronouns to plural pronouns
Running analyze.cpp should result in the display of a single table (5 points) where each of the 57 rows represents a single inaugural speech, and all the above items are given for each row. Failure to display at least part of the above table items for each speech will result in a grade of 0, in spite of how much work you may have done.
Think carefully about the best way to store your data for use by your program. (Note that in C/C++ a dynamically allocated string can be very long.) You may also write other programs or versions of analyze.cpp that give additional results. Do not include the compiled executable versions of your programs in the directory you turn in, but only include the source code (.cpp) versions.
Description of results and discoveries
Include a Microsoft Word file called userid.doc, where userid is your uic netid, or if working with a partner your two netids separated by an underscore. In other words since my netid is reed I would turn in reed.doc. If I were partnering with someone with netid psanch then we would turn in reed_psanch.doc. This file should have:
Your identifying information at the top (Name(s), Lab, Date, 1 line description of program)
Requirements summary of what you did/didn't get done. Clearly indicate here any extra credit additional programs or analysis that you did (see below for extra credit description.)
The table of results of running your program (from step 2)
Your code from analyze.cpp, nicely formatted as text so we can annotate it with comments and return to you.
Extra Credit (up to 10 points)
In addition to what is required, for additional credit you may want to consider other questions such as: Are there particular linguistic characteristics of the best speeches vs. the worst? Is there some pattern based on political affiliation, decade, century, number of quotes, whether or not it is a president's second term? How do the linguistic characteristics connect to the historical context?
Be sure to include a description of any additional tools you created and the results these generated. To get any extra credit your results must be meaningful and interesting. Simply doing extra work will not give you any extra credit, rather you need extra results obtained through the tools you wrote, possibly combined with research.
To summarize, you will zip up and turn in the following:
analysis ----
|
|---- analyze.cpp
|
|---- (any other .cpp program tools developed)
|
|---- netid(s).doc
|
|---- datafiles ----
|
|---- your datafile(s)
|