In their seminal 1981 work on courtroom performance, Reconstructing Reality in the Courtroom, Lance Bennett and Martha Feldman propose a theory purporting to show “how ordinary means of telling and interpreting stories are used in trials to assess the credibility of competing claims.” Key among their insights is that courtroom dynamics are a function of more than lawyers attempting to convince lay jurors in the presence of a judge. What I hope to show over the next few installments in this series is that—without disagreeing with Bennett and Feldman (and even some of their critics)—the matter is rather more complicated than they posit. Ultimately, I want to focus on two things: first, the shifting roles that the various actors play throughout the course of any trial; and second, that most trials do not present jurors with a binary choice between diametrically opposing versions of “what really happened.”

Bennett and Feldman’s baseline insight is that legal judgments are based on the plausibility (not necessarily the truth) of the stories that are told at trial and, more important, that plausibility is a function of “the structural ambiguities linking the key elements of the narrative together.” Thus conceived, jurors make their decisions as auditors of a specific sort: “stories become the layperson’s theory for mapping a set of particulars onto a set of legal rules and deciding how convincing the fit is.” And this mapping is made easier within the story framework because a story is a natural human way to organize large amounts of information, even when the information is received in disjointed packets: at trial, “Once the basic plot outline of a story begins to emerge it is possible to integrate information that is presented in the form of subplots, time disjunctures, or multiple perspectives on the same scene.”

Bennett and Feldman posit that trial stories are built on a set of structural properties that guide jurors in:

1. Locating the central action that is to be explained by the story.
2. Connecting the various symbols in the story through the use of interpretive rules that establish the relations among the supporting information surrounding the central action.
3. Evaluating the connections for internal consistency, completeness, and their collective implication for the central action.

For an illustration of this three-part framework, we can look to one of legal philosopher Neil MacCormick’s favorite objects of analysis, R v. Smith, popularly known as the “Brides in the Bath Murders” case that is well-known to students of evidence and forensic pathology. MacCormick uses the case to explain what it is that makes some narratives cohere and thereby gain plausibility. This fits rather happily with Bennett and Feldman’s theory, which orbits around the “central action” of the story. In Smith, the defendant, George Smith, had a sordid past, one littered with petty crimes, at least seven bigamous marriages, many women fleeced of their life savings, and—finally—murders:

The appellant was indicted for the murder of M. who had been discovered in her bath after having gone through a ceremony of marriage with him. At the trial, evidence was given that subsequently to the death of M. two other women had died in their baths in similar circumstances after having gone through marriage ceremonies with the appellant. Evidence was also given of a consultation between the appellant and a solicitor concerning, inter alia, the effect in law of a voluntary settlement made by M., and whether the trustees could buy an annuity without M’s permission.

Here, the central action is the death in the bathtub. The lawyers’ work at trial is to put on evidence that congeals around the central action in the form of an explanatory narrative. Had the admitted evidence in the case been limited to the fact of a single death, then Smith probably would have gone free (indeed, there probably would have been no case brought). But that was not the case, as we see as his counsel scrambles to explain away the evidence in closing argument:

The prosecution have been driven to prove, for the purpose of their case, not only that the prisoner murdered Miss Mundy, which is the only charge upon the indictment but that he murdered all three of those women. … The calling of evidence as to the other two cases impliedly proves that the evidence in that of Miss Mundy is not conclusive. Is there one of you who, if Miss Mundy’s case stood alone, and you knew nothing of the prisoner’s antecedents, would say he was satisfied that the prisoner was guilty of her murder? I venture to think no one would dare to take on himself that awful responsibility.

MacCormick asks us to consider why the multiple-drowning evidence was so damning. The answer lies in human experience. “A man is to be pitied if he loses one wife by drowning in a bath, to be somewhat suspect if he loses two in this way, and to come under the gravest suspicion if he loses three.” Why? Because a healthy person doesn’t usually drown in a bathtub. Nor does a woman often die on the heels of her spouse checking with a lawyer to ascertain whether he would profit by her immediate death. An alternative—demonstrated with expert testimony in Smith—is that a woman immersed in a bathtub can be drowned with no signs of violence if her feet are quickly pulled up. And to this, Smith’s lawyer could only bemoan the rising eminence of science as a discipline germane to the law (and, in a nice rhetorical gambit, blame it on the Americans!): “The calling of expert medical evidence by the prosecution has opened the door to the worst form of Americanism in the administration of British justice.”

MacCormick goes on to ask why the conclusion “Smith killed his wife” better follows from the premises (1) wife one died in a bathtub and Smith was close by, (2) wife two died in a bathtub and Smith was close by, (3) wife three died in a bathtub and Smith was close by, and (4) before the decedent died, Smith checked into inheriting her money then “All three wives died by accident.” It’s not that the murder conclusion follows as a matter of deductive logic; rather, it’s that the murder conclusion is more plausible than the accident conclusion based on common understandings of probabilities. In other words, the murder conclusion coheres with the factual predicates to form an explanatory narrative in a way that the accident conclusion does not. Certainly, if Smith had offered credible alibi evidence or expert testimony suggesting that all three bathtubs had a design defect that had caused many drownings, then ambiguity and attendant reasonable doubt might have crept in. But he did not, and so—reverting to the third element of Bennett and Feldman’s framework—the consistency and completeness of the network of connections bearing on the central action all tilted in favor of the Crown’s case.

As our investigation of trial performance continues, that third property will emerge as most important. Why? The answer lies, I think, in our desire for drama to answer questions of who, what, when, where, why, and how.

Randy D. Gordon is the Office Managing Partner of the Dallas office of Duane Morris LLP. He is executive professor of law and history at Texas A&M University. His new book project, “Everyday Lawyering: Where Imagination and Morality Meet Advocacy and Logic,” from which this series is partially excerpted, is currently under review at an academic press.

Reprinted with permission from © ALM Media Properties LLC. All rights reserved.