Primary Sources and Secondary Sources
What is a Primary Source?
A primary source is a document that was created at the time of the event or subject you've chosen to study or by people who were observers of or participants in that event or topic.
If, for example, your topic is the experience of workers in the Chicago packinghouses during the first decades of the twentieth century, your primary sources might be:
- Chicago newspapers, c. 1900-1920, in a variety of languages.
- A short film, such as an actualité, made during the period that shows the yards.
- Settlement house records and manuscripts.
- Novels about the packing yards, such as Upton Sinclair's The Jungle (1906).
- U.S. census records concerning neighborhood residents for 1900 and 1910.
- A mechanical conveyor system, used to move carcasses from one room to another at the time and place you are researching.
- Autobiographies of meat packing executives, workers, etc., published even many years later.
- Maps that show the location of the packing house plants, made during the period you are studying.
- Music, such as work songs or blues ballads, made or adapted during the time you are researching.
- oral histories of packing house employees' experiences, though a historian's comments on those oral histories would be a secondary source.
The medium of the primary source can be anything, including written texts, objects, buildings, films, paintings, cartoons, etc. What makes the source a "primary" source is when it was made, not what it is.
Primary sources would not, however, include books written by historians about this topic, because books written by historians are called "secondary" sources. The same goes for historian's introductions to and editorial comments on collections of primary documents; these materials, too, are secondary sources because they're twice removed from the actual event or process you're going to be writing about. So while a historian's introduction to Upton Sinclair's novel The Jungle (1906) is a secondary source, the novel itself, written in 1906, is a primary source.
What are Secondary Sources?
Once you have a topic in mind, you need to find out what other scholars have written about your topic. If they've used the same sources you were thinking of using and reached the same conclusions, there's no point in repeating their work, so you should look for another topic. Most of the time, though, you'll find that other scholars have used different sources and/or asked different questions, and that reading their work will help you place your own paper in perspective.
You want to move past just looking for books in the library. Now that you're doing your own history research and writing, you should step up to the specialized bibliographies historians use for their own work. Don't stop looking for secondary sources until you begin to turn up the same titles over and over again. Put those titles you see most frequently and those that are most recently published at the very top of your list of things to read, since they are likely to be the most significant and/or complete interpretations.
After you've located and analyzed some primary sources and read the existing secondary literature on your topic, you're ready to begin researching and writing your paper.
Remember: when lost, confused etc., ask a reference librarian! They are there to help.
[adapted in part from Peggy Pascoe's site at the University of Oregon]
Questions to Consider When Reading Primary Historical Documents
- When and by whom was this particular document written? What is the format of the document? Has the document been edited? Was the document published? If so, when and where and how? How do the layout, typographical details, and accompanying illustrations inform you about the purpose of the document, the author's historical and cultural position, and that of the intended audience?
- Who is the author, and why did he or she create the document? Why does the author choose to narrate the text in the manner chosen? Remember that the author of the text (i.e., the person who creates it) and the narrator of the text (i.e. the person who tells it) are not necessarily one and the same.
- Using clues from the document itself, its form, and its content, who is the intended audience for the text? Is the audience regional? National? A particular subset of "the American people"? How do you think the text was received by this audience? How might the text be received by those for whom it was NOT intended?
- How does the text reflect or mask such factors as the class, race, gender, ethnicity, or regional background of its creator/narrator? (Remember that "race" is a factor when dealing with cultural forms of people identified as "white," that "men" possess "gender," and that the North and Midwest are regions of local as well as national significance.)
- How does the author describe, grapple with, or ignore contemporaneous historical events? Why? Which cultural myths or ideologies does the author endorse or attack? Are there any oversights or "blind spots" that strike you as particularly salient? What cultural value systems does the writer/narrator embrace?
- From a literary perspective, does the writer employ any generic conventions? Use such devices as metaphor, simile, or other rhetorical devices?
- With what aspects of the text (content, form, style) can you most readily identify? Which seem most foreign to you? Why? Does the document remind you of contemporaneous or present-day cultural forms that you have encountered? How and why?
Asking a Good Historical Question; Or, How to Develop a Manageable Topic
When writing a historical research paper, your goal is to choose a topic and write a paper that
- Asks a good historical question
- Tells how its interpretation connects to previous work by other historians, and
- Offers a well-organized and persuasive thesis of its own.
Let's take this one step at a time.
- Asking a good historical question:
A good historical question is broad enough to interest you and, hopefully, your classmates. Pick a topic that students in the class and average people walking down the street could find interesting or useful. If you think interracial relationships are an interesting topic and you find the 1940s to be an equally fascinating time period, come up with a question that incorporates both these interests.
For example: "How did white and African-American defense plant workers create and think about interracial relationships during World War Two?" This question investigates broad issues—interracial romance, sexual identity—but within a specific context—World War Two and the defense industry.
WARNING: Avoid selecting a topic that is too broad: "How has war affected sex in America?" is too broad. It would take several books to answer this question.
A good question is narrow enough so that you can find a persuasive answer to it in time to meet the due date for this class paper.
After selecting a broad topic of interest, narrow it down so that it will not take hundreds of pages to communicate what happened and why it was important. The best way write a narrow question is to put some limitations on the question's range. Choosing a particular geographic place (a specific location), subject group (who? what groups?), and periodization (from when to when?) are the most common ways to limit a historical question. The example above already contains a limited subject group (whites and African-Americans) and a short time period (WWII, 1941-1945); simply adding a place, such as "in the Bay Area" or "in Puget Sound" further narrows the topic: "How did white and African-American defense plant workers in the San Francisco Bay area create and think about interracial relationships during World War Two?" is a much more manageable question than one that addresses all defense workers.
WARNING: Avoid a question that only looks at one specific event or process. For example, "What happened on Thursday, Dec.12, 1943 at the Boeing bomber plant in Albany, California?" is too narrow. Perhaps there may have been several important events that day, including a fight over an interracial relationship. However, this question does not position you to explore the larger processes that were taking place in the plant over time, nor why they are important for understanding sex, race and gender in American history.
A good historical question demands an answer that is not just yes or no. Why and how questions are often good choices, and so are questions that ask you to compare and contrast a topic in different locations or time periods; so are questions that ask you to explain the relationship between one event or historical process and another.
Examples (why and how, compare/contrast, explanatory):
- "Why and how did Latina women in Texas challenge their traditional sexual identities in the 1960s?" or "Why and how did captivity narratives define interracial romance in colonial America?"
- "Gay liberation over time and space: The Stonewall Uprising and Harvey Milk assassination protests compared;" or "Sex and gender after the war is over: The contrast between the post-World War One and World War Two eras."
- "Go West, Young Woman: the rise of the popular newspaper, western boosterism, and the origins of women in professional journalism;" or "Sit-coms, kitchens, and Mom: TV and the redefinition of femininity and domesticity, 1950-1975."
A good historical question must be phrased in such a way that the question doesn't predetermine the answer.
Let's say you've decided to study the Tillamook Ku Klux Klan. You're fascinating by the development of the Klan, and repelled by its ideas, so the first question you think about asking is "Why was the Klan so racist?' This is not a good historical question, because it assumed what you ought to prove in your paper that the Klan was racist. A better question so ask would be "What was the Klan's attitude and behavior toward African Americans and immigrants, and why?"
- Connecting your interpretation to previous work by other historians:
Once you have a topic in mind, you need to find out what other scholars have written about your topic. If they've used the same sources you were thinking of using and reached the same conclusions, there's no point in repeating their work, so you should look for another topic. Most of the time, though, you'll find that other scholars have used different sources and/or asked different questions, and that reading their work will help you place your own paper in perspective. When you are writing your paper, you will cite these historians—both their arguments about the material, and also (sometimes) their research findings. Example: "As Tera Hunter has argued concerning Atlanta's laundresses, black women workers preferred work outside the homes of their white employers"(and then you would cite Hunter in a footnote, including page numbers).
- Offering a well-organized and persuasive thesis.
Think of your thesis as answering a question. Have your thesis answer a "how" or "why" question, rather than a "what" question. A "what" question will usually land you in the world of endless description, and while some description is often necessary, what you really should focus on is your thinking, your analysis, your insights.
Consider the following questions when reviewing your thesis paragraphs:
- Does the thesis answer a research question?
- What sort of question is the thesis answering?
The thesis paragraph usually has three parts: (1) the subject of your paper, (2) your argument about the topic, and (3) the evidence you'll be using to argue your thesis.
- Is the thesis overly descriptive? Does it simply describe something in the past? OR,
- Does the thesis present an argument about the material? (This is your goal.)
- Is the thesis clearly and succinctly stated?
- Does the thesis paragraph suggest how the author plans to make his or her argument?
Examples of Thesis Statements: From Bad to Better
"Dorothy Richardson's The Long Day is a provocative portrayal of working class women's lives in the early part of the twentieth century." This is a weak thesis for a paper, since it is overly vague and general, and is basically descriptive in nature. The thesis does not suggest why or how Richardson's book is "provocative."
"The narrator of Dorothy Richardson's 1905 work, The Long Day, exemplifies many ideas and perspectives of the early twentieth century's new feminism." This is a bit better, since the author is actually suggesting that there might be an argument about early twentieth-century feminism. But note how the language is still vague. What ideas and perspectives? To what effect does Richardson's work deal with these ideas?
"While The Long Day's narrator exemplifies many tenets of the new feminism, such as a commitment to women's economic independence, her feminist sympathies are undermined by her traditional attitudes towards female sexual expression." OK. Now we are getting somewhere! This is a solid thesis. Note that the language is specific (commitment to women's economic independence, as example). Also, the author has detected a contradiction in the text, a tension that the paper can fruitfully analyze. It could be strengthened further by suggesting HOW Richardson's sympathies are undermined by her traditional attitudes.
How to Document Your Sources
In history courses, you should use the traditional endnote or footnote system with superscript numbers when citing sources. Do not use parenthetical author-page numbers as a general rule. Exceptions include: short discussion assignments; five page analytical papers where you have been assigned the specific texts that you are analyzing.
The preferred guide for citations in history is The Chicago Manual of Style. The University of Wisconsin's writing center page offers a helpful introduction to the traditional method of citing sources laid out in The Chicago Manual. Also visit U of T's advice file on documenting sources for a concise overview on the traditional method.