As a historian, I have always been fascinated with the underlying assumptions and questions that characterize our craft. Perhaps this fascination derived from my earlier forays into mathematics and physics at Wesleyan University, subsequent home of the journal History and Theory, in the 1950s. A Harvard University course with Morton White on the philosophy of history renewed that fascination. (I wrote a paper on Trotsky's philosophy of history, using set theory.) Later, I agreed to take my turn and teach the philosophy of history course required of all history majors at Washington University in St. Louis. I had such fun that I have persisted ever since, despite the necessary distractions of being a Russian historian, a department chair, and a dean.

The so-called "methods course" raises nagging questions about doing history that often go unasked in conventional courses. What do we mean by a cause, an event, evidence, a primary or secondary source, a narrative, an argument, a manuscript, or a document? What is the difference between explaining why and explaining how an event happened? What is the function of a footnote? What is plagiarism? How much can a historian speculate without hard evidence? And so on.

Many students are impatient with such questions. They want to get on with reading or researching and writing history themselves. But those who persist get a deeper understanding of the many subtleties of a subtle craft. They remain wary of simplistic explanations and narratives in books, in the classroom, and in films. They learn to think about the distinctions between fact and fiction, and the rewards of discovery, as well as construction, in narrating and analyzing the past. They learn better than some professional historians how to take notes and acknowledge their sources correctly. They become subversive, skeptical, and careful thinkers about history, and budding historians.

At Davidson College, there is no required course in historical method. Instead, all history majors must take one of several sophomore-level topical courses each of which considers questions of method, historiography, research, and writing. The sophomore year is a good time to raise method questions. Students have begun to study history in college, but more advanced courses and research opportunities lie ahead. My own course on the theory and practice of history is one of those courses

My course is a discussion seminar grounded in required readings. Students write a number of small reaction papers based on those readings, which we discuss in class. They ultimately research and write a topical paper that emphasizes questions of method and historiography. They present an outline, a draft, and oral reports on their work before turning in the final paper.

Norman McLean's Young Men and Fire (1992) provides an excellent beginning. McLean dramatically tells the story of how some young smoke jumpers died in a forest fire in Mann Gulch, near the Missouri River in Montana, on August 5, 1949. McLean is a writer, not a historian. But McLean interviews survivors, explores Forest Service archives, studies fire science, plots wind patterns, walks the ground, and tries to imagine what it is like to be young, and scared, and about to suffocate or burn to death in a firestorm while racing uphill. He also tells the story of trying to find the story. In so doing, he raises a number of key questions of historical method. And he shows explicitly how the messy story of one's research differs from the much neater final story that one ultimately constructs.

We then shift from the microscopic to the macroscopic level, and read a general work about doing history, Mark Gilderhus's very fine History and Historians (fifth edition, 2002), and, more recently, my own essay and method collection, The Historian's Toolbox (to be published soon by M. E. Sharpe). In discussion, we begin to identify broad questions of evidence, narrative, argument, sources, and causation that characterize all historical research, writing, and thinking.

To study causation, we read a collection of primary and secondary sources by Kenneth Stampp, The Causes of the Civil War (1959). Here we examine causes, necessary and sufficient, for an event to occur, and the distinction between explanatory patterns and predictive laws. Students also learn about historiography, schools of thought, conventional wisdom, and historical revision.

To examine the role of the individual in history, we then read Steven Ozment's The Burgermeister's Daughter (1996). Ozment's heroine, Anna Buschler, manages at least two love affairs with two different men in her youth while living in her father's house. Set in Reformation Germany, Ozment's story makes an ordinary woman seem quite extraordinary as she engages in a lifelong struggle to gain her rightful inheritance in a court system run by men and a society ravaged by religious warfare.

John Keegan, the foremost military historian of our time, helps students understand historical debate and revision with his brief book on The Battle for History. Re-Fighting World War II (1995). Students learn about official histories, war memoirs, official secrecy, and military revisionism. We then explore issues of identity, race, class, and gender by reading Gerda Lerner's Why History Matters (1997).

Having tortured my students one year with Keith Jenkins's The Postmodern History Reader (1997), I discovered student impatience with pontification, obfuscation, and the esoteric. Instead we now examine a single "postmodern" history, Natalie Zemon Davis's The Return of Martin Guerre (1983). Students learn how a gifted historian can retell an old story of double identity in late 16th-century France with a new interpretation based more on context than text, on imagination and speculation more than on new evidence. Robert Finlay's American Historical Review essay and Davis's response (AHR 93:3, June 1988, 553-71; 572-603) help students understand the peer review and criticism that marks all historical debate.

Finally, we circle back to more general thoughts about history with Richard Evans's In Defense of History (1997). Students come to appreciate that history is both discovery and construction, that fact and fiction are not identical, and that postmodernism is a fad that has run its course (or discourse) without leaving much trace on the world of historical writing and research. They also are able to connect Evans's discussion of objectivity, causation, and social history with their own research papers, now in progress.

I am not one of those historians who must be dragged kicking and screaming into a conversation about the underlying assumptions, methods, and questions of the craft. Doing history is fascinating and complex, a craft that combines elements of both art and science. The sooner students understand this, the better able they are to discover the past, construct historical narratives and arguments, and live with ambiguity in a world where truth is always probable and contingent, never certain and predictable. Teaching historical method courses is not an obligation, but an opportunity.

—Robert Williams is the Vail Professor of History at Davidson College.