The final book in a trilogy I have been working on since the 1970s (gasp!) is being published this spring, under the title Separated by their Sex: Women in Public and Private in the Colonial Atlantic World (Cornell University Press).  Ironically, when I was younger I liked to joke to myself about senior historians who announced projects involving trilogies that were never completed.  Having pulled it off, I would like to offer three rules for success.

First, start a trilogy without announcing it, or even without realizing that is what you are doing.  In the early 1970s, when I was an assistant professor searching for a project to follow my dissertation on the loyalist exiles in England (published as The British-Americans by Little, Brown in 1972), I originally planned to write my second book on the coming of the Revolution.  But then I began to read the first published articles on American women’s history, and I was hooked.  Those initial articles examined women in antebellum America; they made assumptions about eighteenth-century women I thought were incorrect.  So I set out to write an article to correct them.  Eight years later I had a book:  Liberty’s Daughters: The Revolutionary Experience of American Women, 1750-1800 (Little, Brown, 1980).  Based on extensive research into the correspondence and letters of women in the era of the Revolution, it sought to answer the question:  what was the impact on women of the Revolutionary War and its accompanying political upheavals?

I discovered in response that women, who in the 1750s tended to see themselves as non-political, living confined to what they termed a “private sphere,” avidly engaged in political activities and discussions during the 1760s and 1770s, even while often apologizing for doing so.  First, they participated in prerevolutionary boycotts of British goods (especially taxed tea); and some also joined groups that symbolically spun yarn in public places, to encourage Americans to produce and wear cloth of their own manufacture rather than textiles imported from Britain.  Second, as residents of the war-torn independent states, they followed, reported, and developed their own opinions about military and political news and events.  Although I did not follow their politicization into the 1790s, I was not surprised when Rosemarie Zagarri’s Revolutionary Backlash: Women and Politics in the Early American Republic (2007) demonstrated that women and men assumed that women would be involved with partisan activities until well after 1800.

Second,plan a volume that then morphs into two—or three—instead of one.  After having written two books on the revolutionary era, I was ready for something completely different.  Even as a graduate student I had worked only on the eighteenth century, so I switched gears and moved backwards in time into the seventeenth century, in fact to the beginnings of English settlement in North America.  This time my initial question grew out of scholarship on gender:  in the early seventeenth century, when English governmental theories analogized between the family and the state, I asked, what did it mean for women, men, and family life that the king was believed to be “like a father to his people” and the chief source of theoretical authority in society was the Fifth Commandment, which was “Honor thy Father and Mother” (my italics)?  Did the prevalence of the family analogy mean that women could play an acknowledged role in both politics and family?  And as a follow up:  what then (if anything) happened when the family analogy was replaced by contract theories of government promoted by John Locke and other Whigs?  In brief, what were the consequences for women when the origins of the state, instead of being seen as organically stemming from families in which women as well as men wielded authority over children and servants, hypothetically lay in decisions made by a group of men in a state of nature?

After I had spent several years reading court records from the early English colonies (the only abundant source of evidence from a period with no newspapers, few printing presses, and low literacy rates in general), I realized that what I had envisioned as a book with two halves—one focusing on the early colonies, one on the post-1688 era following the Glorious Revolution and the rise of Lockean thought—would have to be two books. After a few more years of research, I recognized that even completing the first book would take much longer than I had thought would be required for the entire project.  In the end, Founding Mothers & Fathers: Gendered Power and the Origins of American Society (Alfred A. Knopf), covering the years until approximately 1670, did not appear until 1996.

The answer to my initial question emerged during the course of the work.  Not all women, but some—high-status women who could be seen as metaphorical mothers—were numbered among the wielders of authority in colonial society. Eventually, my findings led me to conclude that in seventeenth-century Anglo-America rank was a more crucial determinant of social and political standing than was gender. As a modern American feminist who began with the opposite assumption, it took some years of reading seventeenth-century documents for the evidence to register on me.  Once it did, though, that conclusion became key to my understanding of early colonial society, including, for example, revealing one of the major reasons why the high-status Anne Hutchinson had such a significant impact on early Massachusetts.

I did not allow myself to think about the difficulty of embarking on the next book—the second half of the original project—until completing Founding Mothers & Fathers, because studying the decades in the late seventeenth and early eighteenth centuries poses particular problems of research in the American context.  Colonial historians refer to those years as a “black hole” because of the paucity of sources.  For a variety of reasons, almost all surviving court records from the period lack the substance and detail of earlier documents, so the research strategy I employed in Founding Mothers & Fathers would not work.  Nor would the research strategy of Liberty’s Daughters: most colonial women before the 1730s did not enjoy sufficient leisure to write letters or diaries, even if they had access to scarce and expensive paper and knew how to write, which most did not.  The first newspapers that contained more than basic commercial information were not published until the early 1720s.  On top of all those problems, any work focusing on gender and public life would have to confront an event that I feared would simply take over any such study: the Salem witch trials.

Those considerations, plus a reluctance to begin another large-scale research project that I anticipated would consume a number of years, led me to decide to write about Salem witchcraft separately.  Though planning at the outset to focus on gender issues in the trial records, as had so many others, I ultimately concluded that the context of the late-seventeenth-century Indian wars in New England provided the critical key to understanding that famous episode.  Gender played a role in my book on the subject, In the Devil’s Snare: The Salem Witchcraft Crisis of 1692 (Alfred A. Knopf, 2002), but not as prominently as I had originally imagined it would.  Thus it constitutes a side excursion, not a major part, of the larger project.

With the witchcraft book finished, I finally had to confront the problem I had postponed for so long:  how could I write a book about gender definitions and political power in a period with so few sources available to address such topics?  That question leads to my final piece of advice to future trilogy-writers.

Third,write the second volume last.  If one knows how the story starts and how it ends, filling in the blanks in the middle is relatively easy, even if sources are scarce.  In this case, for me the crucial blanks in the middle revolved around the ubiquitous language from Liberty’s Daughters—references to women’s “private sphere,” statements that politics were “outside” such a “sphere”—that was wholly absent from the documents I had consulted for Founding Mothers & Fathers and In the Devil’s Snare.  When had such concepts been introduced into Anglo-America?  When had gender replaced status as a defining characteristic of wielders of public power?  How had these ideas become dominant in the minds of both women and men?

To answer those questions and to solve my ever-present source problem, I turned to subjects and materials that had seemed merely of remote, comparative interest during my previous work:  English events and publications.  In the many decades since I conceptualized and began research for the first two books, colonial historians, including myself, had become much more aware of the influence of transatlantic cultural currents on the colonists.  Precisely because North America lacked many printing presses, literate colonists eagerly consumed English magazines, pamphlets, and books throughout the seventeenth and eighteenth centuries.  Commercial ties and some continued immigration kept the colonists in regular (if somewhat delayed) touch with the home country, and, I reasoned, English publications could serve as a proxy for nonexistent American ones.

Accordingly, when I headed to England in September 2005 for a year’s appointment as Pitt Professor of American History and Institutions at the University of Cambridge, I planned in advance to research the English side of the story, starting around 1688.  But after some months it became evident that my tale truly began earlier, and that the most fruitful seventeenth-century sources on gender and politics in England stemmed from the English Civil War of the 1640s, when large groups of women and men regularly petitioned Parliament, and newsbooks—in effect, weekly newspapers—commented on their activities.  The climax came in a fascinating genre of pamphlets on the theme of a “parliament of ladies,” as men fantasized about governments controlled by women at a time when most political verities were being called into question.

That material gave me a starting point, but I still hoped to find what I came to call the ur-text of public and private:  the first time in England or America that an author defined women’s “private sphere” as the opposite of the public realm.  And, one day in the rare book room of the British Library, I found it:  a 1702 pamphlet written by the multifaceted printer John Dunton, Petticoat Government: in a Letter to the Court Ladies.  The accession of Queen Anne, the first English queen whose husband was not a king, led Dunton to distinguish between the queen’s public role as monarch and her private role as wife in the household.  Strikingly, his descriptive language presaged that of the nineteenth-century cult of domesticity; a good wife, he wrote, displayed “Devotion, Modesty, Chastity, Discretion, and Charity.”

Dunton’s formulation took decades to gain full acceptance.  Even Joseph Addison and Richard Steele, in their influential Tatler and Spectator, did not employ the language of public and private, while nevertheless simultaneously revealing their belief that political activity of any sort was inappropriate for women.  And here the long delay in my attacking this part of my project worked greatly to my advantage.  The advent of online keyword-searchable databases of published materials such as Eighteenth-Century Collections Online (ECCO), Early American Imprints (EAI), and America’s Historical Newspapers (AHN) allowed me to identify with relative precision the decade in which the dichotomy public/private acquired gendered meanings.  By the late 1730s the trope was established on both sides of the Atlantic, so much so that one anonymous English author (whose essay was reprinted at least twice in the colonies) declared that Queen Elizabeth and other politically astute females of the past must have been hermaphrodites!

And so my unintended trilogy on the theme of gender and political power in early America is complete.  Research for it led me in each iteration in so many unexpected directions that I do not know what to anticipate as I embark on a new project, that long-postponed look at the years immediately prior to the American Revolution.  But I do know that the book, informed by the past decades of work on the trilogy and its sidelight volume, will be very different from that I would have researched and written in the 1970s.