ISSA Proceedings 2002 – Can Testimonies Constitute Proof?
In a recent book about Holocaust survivors and their role in establishing the truth about the so-called Final Solution, the author, Kelly Oliver, prefaces her argument with the following case narrative: one of the few survivors of a Nazi camp known to have been the site of a unique event – a rebellion of the inmates – included in her testimony about the camp a description of the four liquidation rooms and the four afferent chimneys. However, several archival sources about this particular camp and the activity that took place at it explicitly refer to five chimneys, instead of four. The difficulty raised by the discrepancy between the survivor’s testimony and other kinds of documents is quite serious. In fact, it constitutes a dilemma at several distinct levels: moral, philosophical, and, I will argue in this paper, rhetorical.
It has been suggested that the woman might be an impostor, precisely because her testimony is contradicted by other evidence. Obviously, such a response is predicated on the assumption that “other evidence” carry more weight that an isolated testimony. And whereas psychiatrists on the other hand have tried to explain why a survivor’s memory can present some inaccuracies while still being largely reliable, there are also historians for whom a personal narrative about the Holocaust is qualitatively preferable to other kinds of evidence – in this case, archival ones. Indeed, in an essay on the perception of history in modernity, Phillippe Ariès has argued that perhaps the most salient feature of historiography after World War II is the increasing popularity of testimonies as genre and means of argumentation. By means of proposing a definition of testimonies, Aries emphasizes their generic distinctiveness from memoirs: while the latter merely convey a private experience, the former defines experience as eminently private. In the age of modernity, with all its incomprehensible and unaccountable atrocities, says Ariès, history can begin to make sense insofar as individual human beings own up to it, by taking responsibility for it as their own (regardless where a specific event happened, in Eastern Europe or in Somalia, Israel or China).
History, on this account, moves into the realm of the private at least from an ontological and moral standpoint. For the discovery and understanding of this kind of history, testimonies play a crucial role: yet even so, their relevance is inextricably connected to their reliability: even in the aftermath of Hayden White, Louis Mink and others’ efforts to establish the thoroughly constructed nature of the past and the disciplinary proximity between historiography and literary or rhetorical discourse, claiming responsibility for the past assumes that are certain procedures which can guarantee a modicum of accuracy of the reconstruction. The recent discussion among Holocaust scholars concerning the validity of some testimonies shows the need for establishing criteria to evaluate the reliability of historical testimonies. But in more general terms, what is ultimately at stake in disagreements about the role of testimonies is their status as means of argumentation, or, to put it differently, the context in which testimonies can constitute proof. Yes, relevance should not be confused with accuracy – as some historians are quick to point out. And while Ariès argues for the moral and ethical relevance of testimonies, historians who reject them invoke the difficulty or sometimes impossibility of determining their accuracy. But what this binary leaves out, thus becoming locked into a opposition, is a third aspect connected to testimonies: their persuasiveness.
This aspect is rhetorical, and if properly explored and accounted for, it can contribute to a better understanding of the other two. This is what I set out to do in my paper: probe into the rhetorical function of testimonies in order to articulate a space of analysis for those testimonies that are particularly difficult to accept, either because of their improbable content or because their accuracy cannot be satisfactorily established. In my argument I incorporate some recent observations from the French literature on this topic, but I am also drawing extensively from historical sources, specifically the British Enlightenment and its treatment of the question of miracles. I am certainly not trying to make a case for historical or cultural continuity – that the 18th century view of testimony can or should be applied to contemporary debates in Holocaust studies. I use arguments proposed by Enlightenment philosophers as an invention tool in my own argument, as a way of getting at the problematic of testimonies. I also find the Enlightenment, particularly in Britain, a precious repository of relevant insights, in light of the fact that it had to deal a lot with improbable testimonies on very important and worth exploring topics. From this point of view at least, I see no major discontinuity between the Enlightenment and our era. In fact, there is an important similarity between miracle-reports recorded in the 18th century and Holocaust survivors’ testimonies, insofar as both pose the same rhetorical problem in terms of their plausibility, both make claims about a state-of-affairs that is, for different reasons, impossible to accept. With respect to survivors’ testimonies I am, of course, thinking of what many scholars have described as a crisis of understanding: the survivor has seen and experienced things that are so horrendous that they cannot be comprehended or communicated.
French sociologist Renaud Dulong recommends that an analytic framework for testimonies be based on a dialectic of trust and suspicion that would unfold in a succession of phases, from assessing the witness’s ability of perceiving a situation to which he or she will later testify, to determining the efficiency of memory retention, and finally to evaluating the capacity to render in language a narrative of the event in question. In each of these stages, suspicion or trust can tip the balance. But in more specific terms, dialectic of trust and suspicion proposed by Dulong varies from one community to another: in stable, cohesive communities, in which trusting another’s word is both a consequence and a reinforcement of the interdependence, compatibility, and even similitude of the individuals as members of that community, one can speak of an “institution of testimony.” In a non-democratic society, on the other hand, where there is corruption, surveillance, and a public discourse characterized by deception and unfulfilled promises, generalized suspicion leads to a fundamental mistrust in the other and his experiences. According to Paul Ricoeur, the amount of trust a community is willing to place in one of its individual members ultimately becomes a defining factor for the security of the community in question. Yet the way in which the group acknowledges an individual’s testimony also depends on the congruence between what the testimony has to communicate on a given issue, and what is already known, expected or imagined about that issue. It is this link that explains, in Ricoeur’s view, the tragic solitude of “historic witnesses,” those people who have been in situation or have experienced events that are completely unlike what represents the experience of most people. Their testimony evades what Melvin Pollner has called “mundane reason,” insofar as the “reality” it purports to articulate is incongruent with the reality that is shared and accepted. In the British Enlightenment, testimonies were deemed an important means of argumentation particularly with regards to problematic subjects. In the investigation of particularly challenging and mysterious scientific phenomena, a testimony given by a reliable gentleman with regards to some specific aspect of that phenomenon was considered sufficient to establish the validity of the result as well as to legitimate a specific procedure that had been used by the gentleman in question.
In his Social History of Truth, Steven Shapin has shown convincingly that testimonies played a fundamental role in the consolidation of trust, and that trust in its turn was the cementing factor for the creation of an intellectual and scientific community in late 17th and early 18th century Britain. Shapin cites in his study an impressively long tradition of scholars who invoked testimonies to support especially difficult to accept statements about both reality and its perception or experience by human beings: Boyle, Glanvill, Hooke, Sprat, Stilingfleet, Tillotson, Locke, Wilkins all relied on eyewitnessing as a way of discovering new aspects of reality. But if the many and famous names present in Shapin’s enumeration are no doubt significant, the absentees are no less important. In this context, it is informative to find David Hume as the notable exception: in an essay about the miracles, the Scottish philosopher used the very unreliability of testimony as an argument for his thesis, which was that miracles and similar wondrous phenomena do not exist. Hume’s position on this matter was very unpopular in the larger context of British Enlightenment, as was the diatribe he offers in this text against testimonies. Interestingly enough, though, while rejecting the epistemic relevance of testimonies, Hume also admitted that on the issue of miracles in particular but also in all kinds of other situations in general, reports from so-called eyewitnesses tend to be a very effective means of argumentation. In Part 2 of his essay “Of Miracles,” he bemoaned the fact that “the generality of mankind” are only too eager to listen to testimonies and accept them uncritically, especially when the issue in question is hard to believe or should normally seem preposterous to most people. Thus, in his effort to dismiss testimonies as acceptable proof for the existence of miracles, Hume in fact offers an explanation for why they could (mistakenly) pass for proof. Part 1 of his essay establishes the logical impossibility of miracles, using arguments based on the fundamentals of his skeptical philosophy. In Part 2, on the other hand, the author engages in some detail the common wisdom on the topic, which at that time was, as I mentioned before, an argument for the existence of miracles based primarily on testimonies from reliable sources.
Posing the problem of miracles in terms of their historical credibility, Hume argues that when a person compares a miracle-report to his already existing knowledge about the world, the inevitable disparity is resolved in favor of the latter: what we should always trust, says Hume, is our experience, and when it differs radically from what the eyewitnesses claims, we should dismiss the testimony. For those of Hume’s contemporaries included on Shapin’s list, the difficulty that arises when a testimony is especially difficult to accept because of its implausibility was easily solved (or perhaps explained away) through recourse to the ethos of the witnesses: a testimony, in other words, was to be accepted if it came from a reliable person (and this reliability was in most cases a matter of social status). But according to Hume’s skeptical philosophy, no person should be more trustworthy than others when it comes to bold statements about the nature of reality, simply because all human beings are ultimately wrong insofar s they experience reality in a way that has nothing to do with “the real things themselves” (Buckle 94). Insisting upon the limitations of reason, skepticism rejected the scholastic doctrine about the ability of the attentive observer to gain knowledge of essences. What a human being experiences is the result of a “temporally extended encounter with the world,”which represents no guarantee that the world has always been a certain way, or that it will remain that way. How can one evaluate, then, the fit between what we think about the world, and how the world is, between appearance and essence? To this question, in a text called “Skeptical Doubts Concerning the Operations of Understanding,” Hume presents an answer that allows him to acknowledge the distinction between appearance and essence without, however, making too much epistemic fuss about it. Such a conception certainly doesn’t seem to leave much room for genuine discoveries, but then again, for Hume discoveries were only an illusion anyway. We may never come to know or understand the so-called essences or the “ultimate causes” of reality, says Hume, but we form nevertheless reliable beliefs about the world, based on custom, or habits of the mind. The iteration of certain experiences allows us to find patterns in nature, and such patterns represent the closest we can come to essences. The repeated transformation of water from one state to another as a result of temperature, for instance, allows natural philosophy to formulate physical laws that can further our understanding and explanation of nature. But outside patterns that emerge as a result of the mind being in the habit of observing the same phenomenon again and again, skepticism does not allow for anything else. According to a fairly common view for the 18th century, a criterion of likelihood can only be established, as Shapin has shown, in juxtaposition with the familiar, with already existing knowledge and information.
By definition, a miracle is an occurrence that does not conform with the state-of-affairs as known up until the moment of its manifestation, and hence cannot be accounted for by principles or laws that have been formulated based on that state-of-affairs. By necessity, then, a miracle cannot make sense to a mind that operates with patterns and regularities. Witnesses who testify to the existence of miracles ought to be considered crazy or liars. Their testimony doesn’t prove anything. And yet, as Hume points out both in “Of Miracles”and in his Treatise of Human Understanding, “miracle-stories can and do affect us… and lead us to give credit to them as if they involve a degree of justification, or causal reasoning, which is actually missing.” The credibility of such stories is explained by Hume as the consequence of the psychological mechanism of persuasion. The mental operation through which miracles are processed is described by Hume as following: Whenever presented with information that radically throws into question all the existing schemas, the mind abandons its principles and is ready to adopt a new rule:”when any thing is affirmed utterly absurd and miraculous, it (the mind) the more readily admits of such a fact, upon account of that very circumstance, which ought to destroy all its authority.” In other words, there is a sort of novelty threshold, according to this explanation, beyond which the mechanism of belief is reversed, and the most implausible ideas acquire an odd plausibility precisely because they are so shocking and unexpected. With this account, Hume in fact introduces a new category of verisimilitude, markedly different from the one functioning in the existing rhetorical theory of the age. It is a verisimilitude founded upon pathos, rather than mimesis. What do I mean by this? According to 18th century rhetoric, verisimilitude was an important quality of discourse because it facilitated comprehension and persuasion: to use Barbara Warnick’s definition of the term, “that which has the quality of vraisemblance appears to be true to listeners or readers because it conforms to their past experience and to their expectations of what is normal and natural”. Such a conception of verisimilitude is mimetic because it “arises from the congruence of facts and events in a text, the completeness with which circumstances are reported, and the general familiarity or believability of portraits” (40). Hume’s deployment of the plausibility of implausible accounts (the oxymoron is deliberate) invokes the role of emotions: “The passion of surprise and wonder, arising from miracles, being an agreeable emotion, gives a sensible tendency towards the belief of those events, from which it is derived. Testimonies about miracles are persuasive because they trigger an emotional response of wonder, which is the natural reaction, according to Hume’s entire philosophy, to unfamiliar events. Furthermore, the Humean treatment of passions recognized their epistemic dimension: although normally controlled by beliefs, emotions can in their turn stimulate and even generate beliefs. For example, a coward will readily believe the warning of a danger, because of his emotional disposition to be easily frightened (cf. Treatise, Book 1). Likewise, a melancholy person will quickly believe bad news, because she is already in a sad mood. The emotion, in such cases, finds an easy access into the imagination, and “diffusing itself over our idea of the affecting object, makes us form that idea with greater force and vivacity, and consequently assent to it.” A similar emotional mechanism of belief operates in literary productions and is responsible for the verisimilitude of fictional representations. As argued in another Humean writing, his essay “Of Tragedy,” the poet or playwright uses “bold poetic figures and frequent appeal to emotions” to “make things happen,” to give life to an imaginary lion and even make us afraid of it.
Testimonies about miracles are explicitly likened by Hume to poetry and drama (cf. Treatise); hence, their persuasiveness is steeped in pathos and poetic eloquence. In the case of literary productions, the danger that a reader would mistake fiction for reality is relatively low, according to Hume, because there is a tacit agreement between poets and their readers that literature institutes its own reality, “a poetical system of things.” And even though sometimes a poet can go mad, “in the warmth of poetical enthusiasm,” and have visions of the figments of his own imagination (Treatise, 450), when this happens, reasons must and be summoned “to dissipate the illusions of poetry.” But in the case of miracle-testimonies, the emotional appeal and verisimilitude is more insidiously efficient, because it is of an entirely different epistemic kind. The emotion triggered by miracle-reports is surprise or wonder, which, if we follow Hume’s argument, can instill the belief that miracles exist in a person who is already gullible or curious by nature. But wonder occupies, in Hume’s account, a special epistemic place, because of its paradoxical nature: wonder appears when the mind is confronted with uncertainties, inexplicable phenomena, or mysterious situations, and hence feels an uneasiness caused by “its wavering in the choice of its objects” (453). This uneasiness is resolved only when the mind can form a belief, which then serves to “inliven and infix” the formerly inexplicable phenomenon, in the future preventing “all kind of hesitation and uncertainty about it.” At the same time, however, this “solution” to what in fact constitutes an epistemic crisis is so “agreeable,” that the mind seeks out uncertainty so that it can then amuse itself with its resolution.
Hume’s explanation of wonder is not different from earlier accounts, offered by Descartes and Spinoza. In his Passions of the Soul Descartes defined wonder as the emotional response to first encounters with objects found “to be very different from what we formerly knew, or from what we supposed that it ought to be” (358). To Spinoza, wonder was a mode of conception, which provided epistemic access to concepts for whose understanding no existing category is available (in Greenblatt, 23). In wonder, “the mind comes to a stand, because the particular concept in question is so new that for a moment at least it is alone, unsystematized, an utterly detached object of rapt attention.” In the Cartesian treatment, this standstill of the wondering mind is only temporary: the initial state of perplexion can be overcome “by the application of our understanding which our will determines to a particular attention and reflection,” thus allowing the mind to categorize the novelty.
Wonder, then, constitutes a sort of cognitive therapy, insofar as it prevents the mind from becoming lazy, entrenched in its habits and thus incapable of progress. In the moment of “rapt attention”it so aptly creates, wonder operates a re-negotiation of boundaries, a rapprochement between the realms of experience and imagination, and a redefinition of the possible. It is in this capacity that wonder had gradually come to represent, from Aristotle to the Enlightenment, “an almost inevitable component of the discourse of discovery,” as Stephen Greenblatt has pointed out (25).
The persuasive force of improbable testimonies, therefore, is predicated upon an emotional mechanism. More importantly, the emotion triggered by such testimonies also acts as a mechanism of invention: it alerts the mind to what has escaped it so far, opening up new horizons as well as allowing them to be assimilated with the existing epistemic schemata. Hume’s philosophy was well suited for such a view of testimonies, insofar as it afforded emotions a role in the formation of beliefs. Granted, the philosopher’s concern – that the “generality of mankind” would not be able to differentiate between a reason and imagination, emotional and rational beliefs – also makes him reluctant to accept testimonies as proof in a logical or philosophical argument. But the fact that he acknowledged the persuasiveness of testimonies in common discourse (and also an explanation for it) shows him aware of their function as proof in a rhetorical argument.
In conclusion, the analytic framework that Hume’s account helps us to formulate will be focused on two clusters of questions:
* How do we define access to reality? How do we set limits to our understanding of reality?
* What role do emotions play in cognition? What specific emotions are triggered and/or stimulated by testimonies? How do such emotions affect the audience that responds to the testimony?
Obviously, Hume’s own answers to these questions need not be applicable to another historical period or community, or at least they should not be appropriated uncritically. Instead, Dulong’s observation that testimonies exist in a dialectic of trust and suspicion that is specific to a given community at a given time can be used as a general assumption based on which answers to the afore mentioned questions will be sought out and suggested. The frame of analysis I am proposing is especially designed to deal with the fact that improbable testimonies challenge us to ponder the meaning of “improbability” in a specific context and for specific audiences. But without a set of heuristics, without an interrogatory framework, it is difficult to make sense of specific responses in particular situations. It is a rhetorical understanding of testimonies, grounded in questions of the sort that I have tried to propose and inspired or guided by historical research into the matter that can also provide a space of negotiation and argumentation for historians and philosophers confronted with such difficult issues as accuracy of content, moral relevance, epistemic status, ontological function, etc.
Ariès, Philippe. Le temps de l’histoire. Editions du Seuil, Paris. 1986.
Buckle, Stephen. Hume’s Enlightenment Tract: The Unity and Purpose of An Enquiry Concerning Human Understanding. Oxford University Press, Oxford. 2001.
Descartes, Rene. Passions of the Soul. Trans. Stephen Voss. Hackett Publishers, Indianapolis. 1989.
Dulong, Renaud. Le temoin oculaire: les conditions sociales de l’attestation personelle. Editions de l’Ecoles des hautes etudes en sciences sociales, Paris. 1998.
Greenblatt, Stephen. Marvelous Possessions: The Wonder of the New World. Clarendon Press, Oxford. 1991.
Hume, David. Of Miracles. Intro by Anthony Flew. Open Court, La Salle Ill. 1985.
Hume, David. Treatise of Human Nature. Ed. L. A. Selby-Bigge. Cambridge University Press, New York. 1978.
Oliver, Kelly. Witnessing: Beyond Recognition. The University of Minnesota Press, Minneapolis. 2001.
Pollner, Melvin. Mundane Reason: Reality in Everyday and Sociological Discourse. Cambridge University Press, New York. 1987.
Ricoeur, Paul. La mémoire, l’histoire, l’oubli. Editions du Seuil, Paris. 2000.
Shapin, Steven. A Social History of Truth. Civility and Science in Seventeenth Century England. The University of Chicago Press, Chicago. 1994.
Warnick, Barbara. The Sixth Canon: Belletristic Rhetorical Theory and Its French Antecedents. The University of South Carolina Press, Columbia. 1993.