The document that follows is the fruit of an experiment. To broaden the scope of this issue of PMLA, Heather Dubrow, coordinator of the special topic, suggested that the journal invite a group of academics to meet during the 1994 MLA convention to discuss the status of evidence. The participants, who were chosen to represent a range of specialities and interests, wrote preliminary statements that were circulated in the group before the meeting. Some of the questions raised during the discussion arose from these statements.

Because textual scholarship, having no definable Fach, or subject matter, is an exemplary postmodernist antidiscipline, it can serve as a site for testing the epistemological assumptions and protocols of the current debate over the status of evidence and the nature of proof. Since text is “authority” and “original” and yet also “network” and “tissue” and since forensics encompasses both hard physical “facts” and the rhetorical formulation that gives such data interpretive meaning, a textual forensics will codify and test the interaction of the phenomenological world and the hermeneutic analysis of its evidence. With procedural and conceptual links both to scientific empiricism and to rhetorical strategies for persuasion, textual scholarship becomes a vehicle for anatomizing the postmodernist breakdown of the master narratives, including that of evidence and proof.

When feminist critics of the 1970s rediscovered “The Yellow Wallpaper,” they constructed an interpretation of the story and the history of its publication and reception. Subsequent critics lent authority to an emerging set of accepted “facts”: nineteenth-century audiences read the tale as a ghost story rather than as a critique of the sexual politics of marriage; Gilman fought valiantly against hostility from the entrenched hierarchy of male editors who refused to publish her work; and irate male physicians censured the story once it appeared. By reexamining the documentary evidence on which those “facts” are based, we examine the role that ideology plays in gathering and interpreting evidence. Gilman's story serves as a fine but certainly not a unique example of how scholarship is as grounded in historical biases as the literature it seeks to illuminate.

In the field of medical imaging, theory, technique, and rhetoric converge to produce knowledge. Historical taboo and cultural belief in the fragility of life have protected the interior of the human body from the scientist's prying eyes; nevertheless, in the modern period (since about 1540), the production of medical knowledge has depended on the unveiling of physical detail. Recent work in the sociology of science—notably Bruno Latour's concept of the theater of proof—has questioned this epistemology. Latour argues that scientific knowledge can be produced by superimposing data that create an effect of reality. To illuminate traditional strategies for constructing convincing accounts of hidden biological processes, I examine texts by Andreas Vesalius, William Beaumont, and Wilhelm Conrad Röntgen. I then discuss an advertisement for a contemporary medical-imaging device that, by foregrounding the superimposition of diagnostic data, provides a useful counterexample to the constructed objectivity of the earlier texts.

Contemporary historians' definition of what constitutes historical evidence has expanded dramatically since the mid-1970s, but so have their doubts about the interpretation of evidence, especially for difficult and evanescent sources such as collective memory. As the ambitions of historical practice have challenged the limits of interpretation, the use of novel sources has necessitated recourse to “foundationalist” assumptions, without which errors in historical witnesses cannot be detected or evaluated.

Popular American tales of women's escapes from Roman Catholic convents were important manifestations of the virulent anti-Catholicism of the 1830s and 1850s. These stories also reveal how questions of evidence were imbricated with the woman question in nineteenth-century American culture. “Fictional” and “nonfictional” versions of these narratives attempt to prove their veracity, using a common standard of evidence and shared methods of authentication, documentation, and corroboration—including a reliance on their Protestant audience's reading history. Yet the multiple voices and forms and the visual, as well as verbal, rhetoric that the telling of the escaped nun's story entails work to destabilize feminine spiritual, religious, and moral authority. The escaped nun's intertextual story expresses and contains a cultural anxiety about young Protestant women and their influence in the remaking of American Protestant religious practices.

It is often said that the Shoah cannot be represented. But viewers can learn from Claude Lanzmann's film Shoah how the events referred to as the Shoah produce a reference to themselves. The film presents a montage of encounters, of unanticipated details discovered in filming, of instances selected and arranged as openings to other encounters not included in Shoah. Through careful choreography, the film invites the return of signs from the past. Shoah offers a way of seeing and invites encounters that—like the filmmaker's arrangements—are open to the force of evidence as that evidence constrains interpretative response. Lanzmann's stance, an openness to significance that precedes interpretation, provides a model for a viewer's relationship to Shoah and its material, to the otherness that addresses the audience through the film. In tracing what can be traced and in filming what can be witnessed, Shoah can turn spectators into producers in the work of bearing witness.