The Captive's Position

9780812239584: Hardback
Release Date: 16th November 2006

Dimensions: 152 x 229

Number of Pages: 240

University of Pennsylvania Press, Inc.

The Captive's Position

Female Narrative, Male Identity, and Royal Authority in Colonial New England

Reconsidering captivity narratives published between 1682 and 1707, The Captive's Position explores the ways in which two generations of New England Puritan ministers reacted to internal and imperial challenges to colonial authority by seizing upon representations of captive women to negotiate and to shape a distinctive male identity.

Hardback / £45.00

Why do narratives of Indian captivity emerge in New England between 1682 and 1707 and why are these texts, so centrally concerned with women's experience, supported and even written by a powerful group of Puritan ministers? In The Captive's Position, Teresa Toulouse argues for a new interpretation of the captivity narrative—one that takes into account the profound shifts in political and social authority and legitimacy that occurred in New England at the end of the seventeenth century.

While North American narratives of Indian captivity had been written before this period by French priests and other European adventurers, those stories had focused largely on Catholic conversions and martyrdoms or male strategies for survival among the Indians. In contrast, the New England texts represented a colonial Protestant woman who was separated brutally from her family but who demonstrated qualities of religious acceptance, humility, and obedience until she was eventually returned to her own community. Toulouse explores how the female captive's position came to resonate so powerfully for traditional male elites in the second and third generation of the Massachusetts colony. Threatened by ongoing wars with Indians and French as well as by a range of royal English interventions in New England political and cultural life, figures such as Increase Mather, Cotton Mather, and John Williams perceived themselves to be equally challenged by religious and social conflicts within New England. By responding to and employing popular representations of female captivity, they were enabled to express their ambivalence toward the world of their fathers and toward imperial expansion and thereby to negotiate their own complicated sense of personal and cultural identity.

Examining the captivity narratives of Mary Rowlandson, Hannah Dustan, Hannah Swarton, and John Williams (who comes to stand in for the female captive), Toulouse asserts the need to read these gendered texts as cultural products that variably engage, shape, and confound colonial attitudes toward both Europe and the local scene in Massachusetts. In doing so, The Captive's Position offers a new story of the rise and breakdown of orthodox Puritan captivities and a meditation on the relationship between dreams of authority and historical change.

Chapter 1. Female Captivity, Royal Authority, and Male Identity in Colonial New England, 1682-1707
Chapter 2. The Sovereignty and Goodness of God in 1682: Mary Rowlandson's Narrative and the "Fathers'" Defense
Chapter 3. Deference and Difference: Female Captivity and Male Ambivalence
Chapter 4. The Uses of Female Humiliation: Judea Capta, Hannah Dustan, and Hannah Swarton in the 1690s
Chapter 5. Hannah Dustan's Bodies: Domestic Violence and Third-Generation Male Identity in Cotton Mather's Decennium Luctuosum
Chapter 6. Returning to Zion: Cultural Competition and John Williams's The Redeemed Captive
Chapter 7. The Seduction of the "Father(s)"

Coda: Dux Faemina Facta/Dux Faemina Facti


Chapter 1. Female Captivity, Royal Authority, and Male Identity in Colonial New England: 1682-1707

This book begins with an historical question: why do narratives of Indian captivity appear in New England between 1682 and 1707? During this period a new kind of narrative emerged about colonial women who had been ripped from their families by "savage" men and forced to undergo extraordinary physical and spiritual trials in the wilderness. While North American narratives of Indian captivity had been written before this period by French priests and Spanish and other European adventurers, those stories had focused largely on Catholic conversions and martyrdoms or male strategies of survival among the Indians and self-promotion in the mother country. In contrast, the New English texts represented a colonial Protestant woman who was separated brutally from her family, but manifested culturally valorized qualities of religious acceptance, humility and obedience until she was "redeemed" eventually to her local colonial community. Most strikingly, these narratives were eagerly supported, disseminated, prefaced, and even written, by American-born New English ministerial elites.

Dominant second and third generation ministers like Increase Mather, Cotton Mather, and John Williams appropriated women's narratives during the same period when the traditional political and religious authority of their English-born fathers and grandfathers became newly threatened from abroad as well as from within. Many historical studies of these narratives have traced their transformation from religious to political texts, from theological to sentimental and sensational uses, and from high cultural to popular cultural dominance. Still others have examined such broad transformations in the light of questions about female authorship and agency. Most neglect, however, to explore captivity narratives in their relation to shifts in political and religious authority occurring in this particular period. Thus, although this study draws on earlier work, it places the captivities within that specific historical frame in order to open up the question of the uses which the captivity structure and the representation of the woman captive served for powerful ministerial elites.

What was at stake—personally as well as socially, politically as well as religiously—in prominent New English ministers' appropriation of the position of the female captive at this particular moment? This project argues that a popular literary form, developed from stories about orthodox New English women's captivity among Indians, helped dominant male colonials to address and to negotiate profound transformations in their own sense of personal and cultural identity during a crucial transitional period at the end of the seventeenth century. Analyzing how and why religious narratives of women's captivity came so powerfully to represent a distinctive identity position for powerful second and third generation colonial men is the burden of this project.

I. The Political Contexts of Captivity
Several political contexts surround the writing and publishing of captivity narratives, contexts that intersect with, but are not entirely limited to the literal Indian/colonial conflicts in whose terms they are often understood. Well-known to political historians, such contexts have not been examined in more than a passing way by most literary or cultural historians, yet as this study will show, they cast illuminating light on the growth, development, and uses of women's captivity narratives.

Between the publication of the first text considered here (Mary White Rowlandson's narrative of 1682) to the last (John Williams narrative of 1707), seven changes in government occurred in Massachusetts alone. Internal changes intersected with international conflicts in Europe that deeply resonated at the local colonial level. Four larger political events are especially influential not only for their obvious effects on Massachusetts governmental structure and practice, but also for their effects on the decision of certain second and third generation New English ministers to support and to use women's captivity writings. The first event is the threat to and then the loss of the original Massachusetts charter in 1685. Responses to threats to the charter begin as early as the Restoration of 1660 and come to a head during the entire period under discussion. The second event involves the effects of the so-called Glorious Revolution in England of 1688, in which William, the Protestant Dutch Prince of Orange, successfully seizes the English throne from his father-in-law, the Catholic James Stuart (James II). The Glorious Revolution is used by some colonials to justify the 1689 overthrow of the first royal governor of the Dominion of New England, Edmond Andros and his colonial political supporters, but reactions to it also inform the new Massachusetts charter negotiated with King William by Increase Mather (1691-92), color the regime of William Phips, Increase Mather's choice for the first royal governor under the new charter, and affect the regimes that follow Phips. Continuing conflicts surrounding the charter and the Glorious Revolution in turn intersect with two imperial wars—the War of the League of Augsburg in 1689 and the War of the Spanish Succession in 1702—which spill over into Massachusetts in "English" terms—as "King William's" and "Queen Anne"s" Wars. These wars, the third and fourth events that orient this study, exacerbate continuing issues over legitimate European succession, rights of possession, and expansion that are experienced in New England as boundary and trade wars that equally involve shifting Indian allies. Joint Indian and New French alliances lead to numerous assaults on historically contested New English borders to the North and especially to the East that result in the seizing of a large number of New English captives. A recent study suggests that between 1675 and 1713 (the Peace of Utrecht) up to seven hundred such captives were taken.

Colonial interactions with England and other European rivals indicate how this political context necessarily implicates a religious context. Like the rhetoric informing the charter negotiations and the colonial response to the Glorious Revolution, "King William's War" and "Queen Anne's War" are framed by ongoing rhetorical attempts to link religious affiliation to emergent national definitions and identifications. As the "French" and their allies become negatively typed as overwhelmingly Catholic, so the "English" become defined as broadly Protestant. Traditional colonial New English religious elites like Increase and Cotton Mather, who, as theocrats, directly relate their religious legitimacy and authority to the colony's political legitimacy and authority, react to these changes in a variety of ways. As historians have shown, after initial and open opposition to royalist intervention in New England in the mid 1680s, such leaders come to rally around William as "Englishmen" after the Glorious Revolution, and deploy a new Whiggish rhetoric of political rights and religious toleration. They do so, however, to protect certain traditional New English charter and Church privileges which deny rights and toleration to those who dissent from them politically or religiously. While claiming their identification with a newly defined Protestant political/religious culture in post-Restoration England, they thus at the same time, out of a competing identification with their own first-generation fathers and grandfathers, persist in beliefs and behaviors that draw these new affiliations into doubt.

Three broad local complications arise out of the attempt of ministers like Increase Mather and his supporters to construct themselves at once as tolerant "Protestant" Englishman of the new kind and as the loyal "sons" of New English nonseparating congregationalists. An early and ardent member of a self-proclaimed traditionalist faction before the abrogation of the old charter in 1685, Increase Mather becomes the negotiator of a new charter in the late 1680s, first with James II and then with William III. Almost immediately upon Mather's return from London in 1692, he is emphatically denounced by those who, in his absence, have construed themselves as the true defenders of traditional New English political and religious orthodoxy against outside English intervention. If traditionalists of both generations violently attack Increase Mather because he has, for them, bent too far to English political demands, other second and third generation moderates, attracted to newer English forms of church practice and polity, begin to suggest that cultural changes should accompany political changes in New England. Encouraged by the current rhetorical emphasis on religious toleration in New England, an important competitor of Mather's own generation, Solomon Stoddard, begins even more openly to publish work begun before Mather's departure which directly undermines the traditional covenantal basis of New English church structure from within rather than without.

Each of these groups threatens the balancing acts that Increase Mather, his son Cotton, and their allies come to see themselves as performing and reperforming over the Massachusetts charter. If the first, nonyielding position lays the colony open to royal charges of insubordination or treason, the second and third positions, whether through an overattraction to an unfamiliar and condescending Williamite England or through a more local dismissal of traditional New English ways of understanding legitimacy and authority, seem to betray the identification with the first-generation "fathers" that, for Increase and Cotton Mather and their supporters, grants them a more broadly cultural as well as a political and religious influence.

Thus, while the spill-over of imperial European conflicts to the colonies during this period apparently eventuates in a new rhetoric of a shared "English Protestant" position, especially but not only against the French, it also aggravates and helps to precipitate conflicts within Massachusetts involving those who claim adherence to a particular version of the authority of the first-generation and those supporting varying degrees of cultural as well as political change. Whereas each of the captivity narratives considered in this project can therefore be read as a reaction to external threats, whether Indian, European or both, each ought also to be read in terms of internal conflicts and competitions within New England, specifically, within Massachusetts itself. Although captivity narratives can be argued to play a part in an ongoing construction of "Englishness," as a number of scholars have suggested, this construction surely must be analyzed in terms of its conflicted intersection with competing colonial understandings of "New" Englishness as well. Insofar as issues of "Indianization" obviously inflect captivity narratives, they, too, should be considered in relation to, not as separate from, period constructions of what is English and New English. In addition to expressing a range of complicated attitudes towards real Indians, representations of native peoples during the entire colonial period also express overt or covert interpretations of both European and internal colonial relations in need of further exploration.

Extending and nuancing suggestions made by Roy Harvey Pearce in 1947, literary scholars have argued for a broadly Puritan base to early captivity narratives, a base that becomes increasingly less religious. While this scholarship has opened up important dimensions of these popular texts' uses and transformations of more general Puritan doctrinal or literary assumptions—European-derived notions of providence and typology, for example—it has not, as a rule, recognized how particular features of narratives representing women in Indian captivity significantly link these texts to fears, beliefs and positions espoused by male elites whose dominance is threatened during a specific historical period. Unexplored is the intersection of the publication and republication of four major captivities with different moments of political conflict in Massachusetts, an intersection that points to the personal as well as the religious influence that Increase and Cotton Mather clearly exert on the appearance of these texts.

Mary White Rowlandson underwent captivity in 1676, during the second year of "King Philip's" War of 1675-76, but her narrative of her experience is not published until 1682, in a climate when renewed charter threats are emerging. Possibly prefaced, certainly supported by Increase Mather, the Rowlandson captivity concludes with the final jeremiad of her husband, Joseph Rowlandson, a supporter of Increase and a mediator in the bitter separation of Boston's Third Church from the First Church over the Halfway Covenant. (The Third Church accepted the covenant, while the First Church did not.) Cotton Mather uses the sensational story of Hannah Dustan of Haverhill, taken captive in 1697 during the last year of "King William's War," in three different venues at three different moments of political transition in Massachusetts. The first is appended to a fast sermon delivered in 1697, three years after the ignominious fall and death of William Phips, the Mather-supported royal governor, and just before the expected announcement of a new royal governor. The second appears in Decennium Luctuosum of 1699, Cotton Mather's history of the Indian wars of the past decade, prefaced and concluded by comments to Lord Bellomont, the latest royal governor. The third use of Dustan's narrative appears in Mather's Magnalia Christi Americana which, written sporadically from around 1693 onward, is finally published in 1702, thus coinciding with the appointment of Joseph Dudley, son of the Massachusetts colony's second colonial governor, to the royal governorship. The narrative of Hannah Swarton of Casco, Maine, whose captivity occurs in 1690, during instabilities occasioned by Andros's removal, appears in 1697, appended to the sermon in which Cotton Mather first tells Hannah Dustan's story and defends Governor Phips. Finally, the narrative of John Williams, in which the story of the male minister significantly replaces that of the female captive, appears in 1707, soon after Cotton Mather publishes two virulent treatises attacking the royal governorship of the New English-born Joseph Dudley. John Williams, minister of the frontier town of Deerfield, was the husband of Eunice Mather Williams—Cotton Mather's cousin and Increase Mather's niece.

Connections like these prove useful for isolating the four texts considered here and for asserting the need to read Indian captivities as cultural products that involve colonial attitudes towards both Europe and the local scene in Massachusetts. This inquiry draws on the contexts sketched above to argue that each of these narratives performs a role in local conflicts over colonial legitimacy and authority aroused by new imperial interventions in New England. At the same time, however, simply linking these four captivities to political or religious positions espoused by Increase or Cotton Mather neither fully explains ministerial support for them nor accounts for the complexity of the ways in which early captivities intersect with internal and external controversies. This linkage cannot, in itself, entirely account for the reasons why conflicts over legitimacy and authority become related to the range of meanings these powerful men ascribed to the position of the female captive and the narrative structure in which she was represented.

II. The Question of the Female Captive
Why and how did colonial elites like Increase and Cotton Mather turn particularly to stories of female captivity to represent reactions to internal and external threats to colonial male power and legitimacy? Two possible reasons immediately present themselves. The first is that the experiences of adult colonial women provided the predominant pool out of which captivities, whether written by or about them, would come. The second is the fact, just noted, that Mary Rowlandson and John Williams had personal connections to the Mather family. But other reasons linking understandings of the domestic realm to broader religious and political conflicts over authority become evident once one considers general features of the role played by gender in Puritan social thought and theology.

Edmond Morgan long ago remarked upon the analogical potential of a colonial Puritan social structure grounded on gender hierarchies assumed to be divinely mandated. This mandate not only ordered that woman be subservient to man within the marriage covenant, it also dictated that the marriage covenant itself be used as the basis for explaining and justifying all other social covenants. Although studies of actual gendered practices might belie such hierarchical assumptions, especially towards the end of the seventeenth century, they nonetheless play an important role in ministerial deployments of captivity narratives in which representations of women are so central. Scholars have shown that colonial ministers were aware of the possible contradictions involved in allowing women, whose culturally prescribed duty it was to remain private, subordinate and silent to write or even to be represented as characters in public, heroic narratives. But the narratives' perceived usefulness in justifying and persuading their readers of certain political and religious beliefs, especially after the enormous popularity of the Rowlandson narrative, clearly outweighed hesitations about their publication.

Given the New English Puritan reliance on typological exegesis, ministers could point to the representative quality of the woman captive's experience; she did not stand for women's experience alone, but, viewed in Scriptural terms, for the experience of the entire colony. Similarly, as Ivy Schweitzer, Amanda Porterfield and others have more recently argued, ministers could further draw on the rhetoric of Puritan theology to argue that all elect believers in fact inhabited the woman's position in the spiritual realm. While the secular social realm might be hierarchical by its nature, the spiritual realm allowed for an equality based on men's inhabiting the passive, obedient and humbled position before God that they ideally assigned to women before them in the secular realm. Noteworthy in the case of captivity narratives is how this feminized, spiritual position becomes so strongly read and promoted as a political position as well. In times of political stress, especially during or just after wars, male conflicts often are played out through attempts to stabilize the meanings of women's position. Such was literally the case in the period at hand, which saw not only the writing of the female captivity narratives considered here, but also the Salem witch trials and the increased executions for infanticide of female fornicators.

Historians have read the witchcraft "outbreak" and the ensuing trials at Salem Village in 1692 as responses to the political loss of the original Massachusetts charter and to wide social controversy and unrest about the colony's future. Carol Karlsen has both nuanced and challenged this claim by arguing that the trials arose out of a related confusion and anger about the rising social power and position of some women under new economic conditions stemming from renewed English contacts. Karlsen also links the trials to another related social change occurring in the 1690's: the increasing number of executions of women for an infanticide linked to fornication. Whereas both men and women had been held equally culpable of such sins in the preceding decades, the 1690s exhibit a markedly punitive focus on women. Drawing on the work of anthropologist, Mary Douglas, a variety of scholars have argued that female fornicators are executed only in part because of their personal or even their theological guilt; more important are their transgressions of religious and social boundaries that are felt to threaten the community as a whole. In another turn of this argument, however, the transgressing woman becomes a more representative figure, whose breaking of communal covenants more specifically mirrors the shared guilt, not the vulnerability, of the entire community. As a representative figure, "she" becomes less an inside threat to the community than its scapegoat, her death necessary to cleanse all "the Land," as John Williams put it, of its shared "pollutions."

Another meaning implicit in the notion of the woman as scapegoat complicates reading the fornicator's death simply as a means of restoring community purity. Frank Shuffelton has called attention to the breakdown in the eighteenth century of traditional community measures for supporting as well as exposing those community members, especially the poor and single women, who were in social or religious straits. Examining the late seventeenth century beginnings of such breakdowns, Laura Henigman has recently explored the dearth of community involvement in a number of pregnancies occurring out of wedlock in the 1690s and the concomitant colonial reliance on recent and rigid infanticide laws to punish women offenders after a child's death. These analyses again suggest that instead of restoring community boundaries, the scapegoating of witches and fornicators at this time indicates the community's complicity in such women's boundary-crossing desires and its awareness that the community supposedly cleansed and restored by their deaths has in fact already broken down and been transformed irremediably, both by contacts with different cultural groups, whether Indian or European, and by the desire for land and for goods aroused by these contacts and conflicts. Such readings of the witch and the fornicator point us towards different uses of the woman captive's position than those suggested so far.

Like the witch and the fornicator, the woman captive crosses boundaries, in her case cultural as well as literal. Unlike them, however, this figure is represented as submissive, obedient and loyal to the tradition of the New English "fathers" and their God. Her forced crossing of boundaries is constructed not only as an affliction for her sins, but also as an opportunity to demonstrate her appropriate repentance and belief that God alone can physically and spiritually redeem her. Ministers involved in the publication or editing of the captivities considered here seem to have recognized and stressed clear textual as well as thematic distinctions between their representations of the captive woman and female offenders like the witch and the fornicator. Supporting, editing and even writing narratives for captive women, they wrote blistering execution sermons for female fornicators.

John Williams's first published text, for example, is the sermon he preaches at the execution of one Sarah Smith in 1699. Several years before "improving" and editing the narratives of Hannah Dustan and Hannah Swarton, Cotton Mather preaches a similar execution sermon for Elizabeth Emerson, Hannah Dustan's own sister, and his initial publication of the Dustan/Swarton stories is followed by a collection of execution sermons, Pillars of Salt of 1699. Joseph Rowlandson preaches no published execution sermon for a woman, but Lancaster town records do reveal a particularly nasty confrontation with a local woman, Mary Gates, who disputes his ministerial authority and is duly chastised for it, while his final jeremiad, appended to his wife's narrative, draws pointedly on the figure of the adulteress. On the surface at least, the border-crossing position of the captive woman presents a marked contrast to these transgressors of legitimate community boundaries, these treasonous repudiators of traditional ministerial authority and, given the political crises surrounding these events, of state authority as well.

Or does it? Such attempts to distinguish the captive's position from those of the witch and the fornicator only provoke other questions about what is at stake in ministerial support for this figure. While the captive woman, in contrast to these other figures, is represented as a submissive victim who is taken unwillingly, might she also have been interpreted as actively using her passivity to realize desires for other and different kinds of cultural and political connections and identifications? Could the community's approbation for this figure be based on desires that not only uphold, but also transgress its own boundaries? Although she is represented as claiming a lack of agency, could it be precisely the culturally-valorized passive position of the woman that allows her to cross a variety of cultural, political, and religious norms?

The tempting alternative of captive women's resistance rather than their orthodoxy has been variously considered in a range of studies. In its reversal of expectations—i.e., simply replacing orthodoxy with some form of resistance to it—this approach to women's captivity narratives cannot sufficiently address the complexity of representations of the woman captive's position during this time-period. Too often assuming that the female captive alone occupies a position resistant to orthodoxy as a result of her experience, her gender position, or some combination of the two, however, this argument overlooks the possibility of men's identification with the representation. Those claiming that female captivity invariably belies orthodoxy ignore the fact that there are equally strong and important aims involved both in these texts' identification with orthodoxy and in the kinds of identificatory positions they assume towards the nonorthodox, whether this concept is used to construct other colonials, other English Protestants, French Catholics, or a variety of Indian others. In order to understand the variety and significance of the identity positions religious women's captivities represent for threatened colonial men, other theoretical approaches to these narratives need to be engaged.

III. Ambivalence and/as Colonial Exceptionalism
While a number of feminist critics have analyzed individual women's agency in captivity narratives, particularly their resistance to orthodox men's discursive attempts to control their experience, another group has built on these insights in order to explore questions about broader cultural work performed by women's texts. Tara Fitzpatrick, for example, has moved the question of women's captivities' resistance to ministerial control to the context of older discussions of American exceptionalism and national identity. Revising frontier theory, Fitzpatrick links the domestic to the national by arguing that colonial women's captivity represents an initial moment in the forging of an American identity made exceptional by captive women's spiritual and literal experience with the wilderness and its peoples. Scholars like Nancy Armstrong and Leonard Tennenhouse have read both the resistance and exceptionalism represented in captivity narratives quite differently. They argue that printed representations of a captive colonial woman's voice, in expressing her loyal resistance to assimilation to her captors, mark a moment in the creation of a distinctly English bourgeois identity. A further line of the scholarship on exceptionalism, represented most currently by the work of Michele Burnham, draws on the work of postcolonial theorists such as Homi Bhabha to argue that the captive's liminal position encourages a sentimental response by readers that dissolves boundaries between captives and captors as well as those between nations. In this reading, it is the ambivalence of women's captivity narratives which resurfaces as a defining feature that invariably undermines exceptional identity, whether American colonial or English.

Sharing an interest in American exceptionalism with Fitzpatrick, this study, too seeks to explore a specifically colonial exceptionalism, but only by redefining it. Rather than reading the female captive's experience in the wilderness as engendering a resistance that rewrites Puritan communalism as "American" individualism, however, this argument considers late seventeenth century attitudes towards the English-born "fathers" and towards Europe resonant in the representations of captivity and wilderness used by American-born ministers. In this reading, far from being simply or only figured as a site of an individual female captive's physical testing and conversion, the "wilderness" often renews or re/constructs associations with the moral and psychological "Babylon" of Catholic Europe and post-Restoration England on the one hand, and with the questionable spiritual status of colonials considered disloyal to the traditions of the founders, on the other.

Like the work of Armstrong and Tennenhouse, this inquiry, too, poses the broader question of the roles played by printed representations of women's captivity in figuring a collective, public identity. But a concern with the ways in which (usually) American-born colonials on the margins differentially use women's captivities to imagine their relations to a shifting English center leads to questions about the identifications of certain colonial "New" Englishmen rather than about a broadly "English" identity. Finally, like Burnham's work, this study draws on the psychological concept of ambivalence to dispute the claim that some ultimate single identification, some stable identity, is finally expressed or achieved within these texts. But the reading of the ramifications of the ambivalence expressed in these texts differs from Burnham's conclusions. An historically framed reading of these narratives draws into question the assumption that ambivalence necessarily or always undermines any concept whatsoever of colonial exceptionalism.

The term "ambivalence" has most generally been used to refer to "the actions and conflicts resulting from a defensive conflict in which incompatible motives are involved." Recent scholars of ambivalence point out that such a broad interpretation of ambivalence weakens the precision of the concept as it had first been understood. J. Laplanche and J.B. Pontalis maintain that if "ambivalence" is to continue to possess any distinctive analytical rigor, the concept should be construed more narrowly as referring to "specific conflicts in which the positive and negative components of the emotional attitude are simultaneously in evidence and inseparable, where they constitute a nondialectical opposition which the subject, saying "yes" and "no" at the same time, is incapable of transcending."

A number of elements of this elaborated description seem especially relevant to scholarly discussions of women's captivity in its relation to colonial exceptionalism. The concept answers the need for a theory that can account for the presence of both resistance and orthodoxy in these narratives. Ambivalence, strictly considered, does not refer to conflicts in which one position necessarily undermines the other, but to an emotional attitude in which two positions exist simultaneously. Therefore, not every conflicted response is ambivalent; to locate conflict is not always to locate ambivalence. This definition of ambivalence both refines and draws into question interpretations of captivity which argue that women's narratives invariably display the true/r preeminence of one attitude over another. It does not limit uses of the woman's position to an a priori resistance to or complicity with an "orthodoxy" that may itself be split.

This restricted notion of ambivalence further limits the ways in which the concept can be applied. Ambivalence as a term of analysis should be read and understood in relation to the historical specificities informing the nondialectical "emotional attitude" which it structures. The concept should not be read as thus similarly applicable across all historical moments, but as responding in variable ways to different historical events and changes.

For the purposes of this project, the concept of ambivalence is suggestive and useful for three reasons. Its focus on the ambivalent emotional attitude of an individual subject offers first, the difficult but crucial opportunity of extending this emotional attitude to specific groups as well as to certain individuals—in this case, to some second and third generation men among the New English ministerial elite. While scholars of typology have often acknowledged Puritanism's general capacity to link private to public events, the assumption in what follows, shared with scholars discussed below, is that the concept of ambivalence can be used to reveal shared, collective, and competing, fantasies of men who not only were, but who also consciously constructed themselves as members of particular generational groups. Second, the concept permits focus on the specific historical and cultural fields within which an ambivalent emotional attitude is produced and with which it is implicated and thus provides support for the claim that ambivalence should not be read as similarly and unproblematically applicable across multiple periods and genres utilizing representations of women's captivity. Finally, in questioning ahistorical assumptions about the meanings of this "emotional attitude," a historically informed, limited concept of ambivalence also contributes to the rewriting of the idea of New English colonial exceptionalism offered here. My analysis of the emotional attitude produced within late seventeenth-century captivity narratives suggests that the colonial male identities implicated within them might not have been undermined, deconstructed or even, under all circumstances, psychically distressed by, so much as they were sustained by forms of ambivalence particular New Englishmen found expressed in these representations. To what extent could certain colonial men's profoundest sense of being or becoming exceptional—as groups as well as individuals—have been addressed in the distinctive forms and representations of women's captivity narratives? Considering this question turns us from a broad psychological description of ambivalence to the work of an earlier group of scholars of American Protestantism.

Throughout the 1960s and '70s historians like Philip Greven and John Demos analyzed in cross-generational terms continuing and changing features of American Puritanism—examining, for example, church practices, family structure, domestic and public education, and inheritance customs. What emerged in all these studies was a general consensus about how many first and some older second-generation fathers engendered feelings of helplessness, dependency, suppressed rage and guilt in colonial sons who came of age in the last half of the seventeenth century. These interpretations of a general second and third generation ambivalence towards fathers inform, in different ways, literary interpretations of Puritan sermons and poetry written largely by ministerial elites.

Both Emory Elliott and David Leverenz, for example, have read the images and structure of Puritan sermons as expressing filial ambivalence. While their understanding of why the imagery of these sermons changed is different, each scholar nonetheless puts his formal analysis of ambivalence into specific historical and generational frames. Elliott argues that second generation ambivalence is largely "resolved" by the third generation at the end of the seventeenth century while Leverenz maintains that the "energizing ambivalence" of the first generation "declines" by the century's end.

More recently, feminist critic Ivy Schweitzer has returned to the question of Puritan ambivalence to offer a new reading of Puritan lyric. Drawing specific attention to Puritan ambivalence as gendered, Schweitzer explores how Puritan male poets used culturally-constructed images of women to resolve their own conflicts over patriarchal authority. In Schweitzer's reading, the problem of their ambivalence is "solved" in the structure of the Puritan conversion narrative as men move from inhabiting positions represented as feminized humiliation to achieved positions of "sonship" with the father.

Schweitzer's reading of Puritan lyric raises the important question of the role played by representations of women in negotiating an ambivalent Puritan male identity. While Leverenz's far more psychoanalytically-framed study addresses how different styles of Puritan parenting are expressed in Puritans' sermonic structure and imagery, both his analysis and that of Elliott overlook certain ramifications of Puritan men's appropriations of culturally-constructed representations of women. Unlike Schweitzer's largely unhistoricized work on the lyric, however, Leverenz's and Elliot's work on sermons importantly underscores the necessity of interpreting literary representations of ambivalence in their historical dimensions. Indebted to all three scholars, this analysis explores certain second and third generation men's uses of women's captivity narratives in relation to particular historical moments of their appearance.

The interest here lies less in ultimately claiming that such men's filial ambivalence is allayed, resolved, or declines within these texts, however. This study instead directs attention to reading the multiple identificatory positions towards a variety of objects which representations of female captivity displayed and allowed to particular Puritan men, and to interpreting these positions in relation to a range of changing outer events. Considering how and why ministers of two New English generations seized upon popular narratives by or about captive women to locate, explore, and sustain their own mixed responses to cultural and political shifts in authority, this inquiry seeks to make a case for the larger historical usefulness of a concept like ambivalence for addressing the differences within and between distinctive colonial situations as well as the exceptional identities shaped in a continuously shifting relationship to them.

David Leverenz asks readers to remember that Puritan men and women, whether English or American-born, whether first, second, or third generation, experienced a variety of emotional attitudes—they were not simply ambivalent. An argument about certain New Englishmen's uses of representations of women's captivity does not deny this useful caveat. What it does consider are the questions of why and how different forms and representations come to display not only different emotions, but even the same emotion differently at different historical moments. Specific political, religious and social contexts informing the historical appearance and uses of these gendered narratives offer extraordinary insight into their inner workings as texts. At the same time, the inner workings of the captivities themselves, their own interpretive fluidity as representations, also clearly offered their writers and supporters the opportunity to try on a range of complicated identificatory responses to the changing political, religious and social contexts within and through which these texts are produced. It is in this sense that narratives by and about female captives, rather than serving merely as responses, however complex, to delimited historical phenomena, should be read as themselves related though different forms of that history. The multiple identifications they perform and shape, while exhibiting loyalty and faithfulness to what is variously construed as a traditional vision of authority and legitimacy, simultaneously reveal the doubt, dread, hatred, and desire that sustain and threaten that vision.

At the end of the seventeenth century, the appearance of religious narratives of female captivity indicates a felt need on the part of certain colonial New Englishmen for a new, popular, cultural form through which a nontranscendent, nondialectical "yes and no" towards their "fathers" and towards new relations with a changed England can be expressed and maintained. Briefly yet productively shaping and negotiating such men's ambivalent identities, women's captivity narratives offer one means for us newly to raise questions about the relationship between interiority and history.

V. Summary
The contexts explored above can now be constellated around the following set of claims. From the Restoration of 1660 to the Peace of Utrecht of 1713 imperial conflicts over the larger meanings of political legitimacy and authority in Europe both exacerbate and evoke a crisis in cultural identification for some second and third generation New English ministers which eventuates in a defensive reworking of their English-born grand/fathers' original "errand." This defense should not be read simply as a defense of the first generation, however, as a range of scholars has variously maintained. It should also be read as revealing an equal defense against their own desires to change. Three assertions follow from this claim which this study examines through contextually-embedded analyses of a specific group of captivity narratives published between 1682 and 1707. First, while their identification with the orthodox woman captive can be used to express second and third generation men's loyalty to a shared vision of their "fathers'" authority in the face of perceived inner and outer threats to it, both the structure of women's captivity and its representation of the woman's position suggest varying views about the nature of that authority and about their relation to it that threaten as well as support their views of the "fathers." Second, while identifying with the female captive's position expresses these men's difference from the positions of other internal and external groups—whether Indian, imperial European or colonial—the interpretive slipperiness of the captive woman's position equally suggests the possibility of their identifying at the same time with other positions, whether to defend the "fathers" or to separate from them. Third, the specific emotional attitude towards the "fathers'" authority and towards renewed identification with European—especially English authority—which is produced and performed by the captivity narratives considered here, follows a nondialectical trajectory and comes to a more or less determinate end. If women's captivity narratives of the late seventeenth century thus give shape to a range of identificatory positions informing a sense of personal and collective identity for certain late seventeenth century New English male elites, their capacity to do this work—as a discrete form—gradually breaks down.

The second chapter of this project argues that Mary Rowlandson's Narrative of 1682 should no longer be read simply as a commentary, whether orthodox or resistant, on "King Philip's" (Metacom's) War. Its publication and support by Increase Mather and ministers aligned with him indicate that they also wished to use it to respond to internal (colonial) and external (royal English) challenges to their cultural control of the construction of traditional religious and political authority in Massachusetts. Read in contexts such as Edward Randolph's challenges to the Massachusetts charter and Increase Mather's debates with William Hubbard, the narrative's representations of the passive position of the woman captive, of her relations to her Indian captors, and of her seizure and return could all be mobilized to define and defend a range of older and emergent theological, social and political readings of the "ways" of the "fathers."

At the same time, as the third chapter argues, when it is read in its relation to particular second-generation men's ambivalence towards the first generation, Rowlandson's narrative also supplies its male supporters with reasons for and fantasies about separating from and replacing "fathers" perceived as limiting and constraining their "sons'" authority. Strikingly, the ambivalent emotional attitude towards fatherly authority that these men find registered in a woman's text allows them at once to experience and to avoid choosing among old and newer cultural identifications.

The ability to express and sustain an ambivalence expressed for elite second generation men in an orthodox woman's narrative of her Indian captivity might suggest a developmental moment culturally in New English male identity-making in which the lack of a fixed choice of identification could be at once experienced, tolerated, and, indeed, even desired. But this emotional attitude, however distinctly marked by some third generation ministers' renewed interest in women's captivity, becomes transformed in their texts.

The fourth and fifth chapters of this study examine Cotton Mather's use of the story of Hannah Dustan, who famously bludgeoned and scalped ten of her captors during the opening year of "King William's" War. In his fast sermon of 1697, "Humiliations Follow'd with Deliverance," Cotton Mather attempts to use the figure of Judea capta—whom Dustan is to represent—as a way of reaffirming a covert New English position in the face of the humiliation and death of the first royal governor under his own father's new charter, the colonial-born Sir William Phips, and the threat of a new English-born governor to come. At the same time, however, his sermon seems equally directed towards blaming other New English men for the colony's current position as humiliated "daughter of Zion." If these international and local aims reveal Mather's renewed use of the passive female captive for defending his own version of the grand/"fathers'" "ways," Mather's text also reveals his extraordinarily convoluted attempts to make Hannah Dustan's active position as Indian-slayer conform to these ends.

As a result of difficulties with the real as well as the represented Dustan, Mather turns, in the sermon's published form, to features of Mary Rowlandson's 1682 narrative to write and then to append the narrative of a more orthodox captive, Hannah Swarton to Hannah Dustan's story. Swarton's text, dealing with a captivity that occurred in 1690 and possibly written almost entirely by Cotton Mather himself, provides the opportunity for him both to complete his sermon's original political and religious intentions and to contain the story of Hannah Dustan. Through its representation of a new—specifically French Catholic—arena of captivity, Swarton's narrative ostensibly offers Mather a different rhetorical mode of defending the "fathers," unifying community and defending himself from violent emotions towards both of them aroused by the Dustan narrative.

The fifth chapter addresses the question of why Cotton Mather splits off Hannah Dustan's brutal story from both his sermon and from Hannah Swarton's narrative in his Decennium Luctuosum ("The Sorrowful Decade"—his 1699 history of the Indian wars of the preceding decade. While this oddly secular history appears to be directed to the new royal governor, Lord Bellomont, in order to demonstrate New English loyalty to England's cause against imperial France, it also seems targeted at a wide variety of New English colonials. Read in the double contexts of the history's unremitting focus on the destruction of New English families and of Mather's own execution sermon for Dustan's sister, Elizabeth, who was accused of fornication and infanticide, Hannah Dustan's violence against an Indian "family" becomes indicative of Mather's ambivalent desire not just to compete with other New Englishmen who struggle for the "fathers'" power, but also to move beyond that form of power altogether. While Hannah Dustan's story can thus be used both to defend and to revolt against the by now traditional construction of a New English fatherly authority, it also hints at the need to replace and reinstate authority and legitimacy in the person of a new "father."

The sixth chapter interprets John Williams's The Redeemed Captive Returning to Zion and the sermon written concomitantly with it, God in the Camp, as complicated responses to the governorship of Joseph Dudley, the late-born second-generation New Englander who eagerly replaces Lord Bellomont in 1702. Williams's narrative of the male replacement of the female captive appears in the context of growing anger among self-described New English traditionalists over the cultural as well as the political direction of Dudley's governorship. In spite of Dudley's attempts to use Williams' text for his own ends and Williams's own professed gratitude to Dudley for effecting his release, his captivity narrative seems concerned not only with representing the steadfast loyalty of "English" Protestant captives in the face of French Catholic captors, but also with warning and teaching a royal governor who is openly dismissive of his colonial heritage, about appropriately "New" English cultural modes of reading and acting on current political and social crises.

While each of the narratives examined here addresses the literal and metaphorical meanings of the possible seduction of the captive woman, only Williams, the male captive, makes seduction the central obsession not just of his captivity narrative, but of every other text he writes. The possible sexual as well as the religious "seduction" of his daughter Eunice and the momentary conversion of his son Samuel by French priests may underlie some of this obsession. At the same time, the vulnerable situation of Williams's children, his son as well as his daughter, appear only to exacerbate a fantasy about the authority of New English "fathers" that is as ongoing as it is rigid. Framing Williams's narrative with an examination of all his published works, the seventh chapter argues that Williams's representation of his textual relegation to, and his identification with, the position of the passive woman captive illustrates how the productive tensions of second generation men's ambivalence can no longer be sustained by the third generation. Once the "sons'" fantasy of replacing the captive woman in order to maintain their ambivalent relation to the "fathers'" authority becomes manifested, religious captivity representing the passive woman who is taken away, "forced" into relation with other cultural groups, and providentially restored, can neither be used, imagined nor written in quite the same way.

The literal as well as textual replacement of the orthodox woman by the man in Williams's work thus demonstrates the end of late seventeenth century Puritan captivity's capacity as a cultural form to produce and sustain a particular kind of ambivalent male identity. The trajectory of these early captivities, from Mary Rowlandson's narrative through that of John Williams, as they at once express and give shape to certain New Englishmen's responses to historical and political transformations in fatherly authority, is the story that this study sets out to tell.

To argue that religious captivity of this kind "ends" with John Williams is obviously not to claim that captivity narratives do not continue to be written, read, and indeed, as is the case with those considered here, republished. But the desires that other captivities engage can best be understood not by conceiving them as somehow carrying forward a project unwittingly begun as project by Mary White Rowlandson's narrative, but by reading their structures and their representations of the captive in relation to the particularities and exigencies of their own particular historical moments of writing, editing, prefacing, and publication. Even to the extent that the four texts considered here are republished, they are republished from the vantage point of those looking back and choosing retroactively to read the desires of their own cultural and temporal moments into the textual products of the past.

Cotton Mather's ecclesiastical history of New England—Magnalia Christi Americana—the great works of Christ in America—while written throughout the 1690s, is edited and published during the early Dudley years. Many have shown how it takes the shape of a colossal jeremiad, marking in its overarching structure and themes the rise and fall of New England's wilderness "errand," but few have considered the fact that the Magnalia concludes with Mather's addition of Decennium Luctuosum, his history of the Indians wars, to the body of the larger text . The central dramatic thrust of Decennium Luctuosum, representing at it does a series of contacts, conflicts, and captivities, can be argued to lie in the captivity narrative of Hannah Dustan.

The Coda to this book briefly explores how repositioning Dustan's narrative as an emotional, if not literal, "end" to the entire Magnalia might lead us to reread Mather's ecclesiastical history as ambivalent, but newly and differently so. Clearly, while the Magnalia claims to honor and defend the colonial "fathers," it does so by burying them in a text conceived as their "monument" rather than as an ongoing conversation with them in the present. Although his epic seems thus to entomb the authority of the Puritan "fathers" on one level, it reengages the question of authority in another, imperial scene, the contours of which his generation has watched emerge in the violent European and colonial wars of the "sorrowful decade." Reread in the light of Hannah Dustan's story, the Magnalia's "ending" both supports the theory offered here about particular colonial men's uses of women's captivity narratives, and points to a transformation in the emotional attitude and the male identity which these adaptable representations had once helped so productively to sustain.

Teresa A. Toulouse is Professor of English at Tulane University and author of The Art of Prophesying: New England Sermons and the Shaping of Belief.