Relative plausibility and its critics
| Published date | 01 April 2019 |
| Date | 01 April 2019 |
| DOI | http://doi.org/10.1177/1365712718813781 |
| Subject Matter | Articles |
Article
Relative plausibility and its critics
Ronald J Allen
Northwestern University
Michael S Pardo
University of Alabama School of Law
Abstract
Within legal scholarship there is a tendency to use (perhaps overuse) “paradigm shift” in ways
far removed from the process famously described by Thomas Kuhn. Within the field of evi-
dence, however, a phenomenon very similar to a paradigm shift, in the Kuhnian sense, is
occurring. Although not on the scale of the transformation from Newtonian to Einsteinian
physics or other tectonic shifts in science, the best understanding of juridical proof is shifting
from probabilism to explanationism. For literally hundreds of years, proof at trial was assumed
to be probabilistic. This assumption was given sustained scholarly attention and support
beginning with the 1968 publication of John Kaplan’s path-breaking article that generated a rich
literature explaining virtually all aspects of juridical proof as probabilistic, from the basic nature
of relevancy through the processing of information to the final decision about the facts.
Although probabilism quickly became the dominant paradigm, some analytical difficulties were
detected quite early (“anomalies” or “irritants” in the words of Kuhn), beginning with L.
Jonathan Cohen’s demonstration of certain proof paradoxes. These were extended by Ronald
Allen, who also demonstrated the incompatibility of Bayesian reasoning with trials and pro-
posed an analytical alternative. Again a complex literature ensued with the defenders of the
dominant paradigm attempting to explain away the anomalies or to shield the probabilistic
paradigm from their potentially corrosive effects (in what in fact on a very small scale is pre-
cisely what Kuhn explained and predicted with respect to paradigm shifts in science). Over the
last two decades, these anomalies have become too irritating to ignore, and the strengths of
the competing paradigm involving explanatory inferences (referred to as the relative plausibility
theory) have become too persuasive to dismiss. Thus the paradigm shift that the field is now
experiencing.
We provide here a summary of the relative plausibility theory and its improvement on the
probabilisticparadigm. As Kuhn noted, not everybody gets on boardwhen paradigms shift; there
are holdouts, dissenters, and objectors. Three major efforts to demonstrate the inadequacies of
relative plausibility have recently been published. We analyze them here to demonstrate that
their objections are either misplaced or unavailing, leaving relative plausibility as the best
Corresponding author:
Ronald J Allen, Northwestern University, 633 Clark St, Evanston, IL 60208, USA. 357 E. Chicago Ave, Chicago, IL 60611, USA.
Email: rjallen@northwestern.edu
The International Journalof
Evidence & Proof
2019, Vol. 23(1-2) 5–59
ªThe Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1365712718813781
journals.sagepub.com/home/epj
explanation of juridical proof. It is interesting to note that two of the three critiques that we
discuss actually agree on the inadequacies of the probabilistic paradigm (they provide alterna-
tives). The third concedes that explanationism may provide a better overall account of juridical
proof but tries to resuscitate a probabilistic interpretation of burdens of proof in light of one
particular analytical difficulty (i.e., the conjunction problem, which arises from the fact that proof
burdens apply to theindividual elements of crimes, civil claims, and defenses rather than a party’s
case as a whole). In analyzing the alternative positions proposed by our critics, we demonstrate
that their accounts each fail to provide a better explanation than relative plausibility.
Keywords
evidence, proof, epistemology, probability, inference to the best explanation
Introduction
The “relative plausibility” explanation of the structure of proof at trial is the primary competitor to an
explanation that is more robustly pr obabilistic.
1
Under both explanations, the primary objective in
typical civil cases is to determine which party more likely deserves to win, within the confines of
resource constraints and some evidentiary rules that pursue values other than truth.
2
In criminal cases
and in atypical civil cases, the primary objective is also the same under both explanations: to skew errors
away from defendants for policy reasons.
3
This is done by raising the standard of proof that must be
achieved to “beyond a reasonable doubt” in criminal cases and “clear and convincing evidence” in
atypical civil cases.
4
It is in this general sense that some authors refer to probability as the generally
accepted approach to proof.
5
The disagreement between the relative plausibility and the probabilistic
explanations is not over whether the legal system is to some extent attempting to regulate errors; of
course it is. Rather the disputes are about, first, the process through which the conclusion is reached that
one party’s case is good enough for a verdict, and second, the nature of that judgment.
6
The “relative
plausibility” theory explains the process as involving reasoning put to the effort of deciding the relative
plausibility of the parties’ various explanations of the evidence and the contested events in civil cases; in
criminal cases the analogue is determining whether the prosecution has a plausible case, and if so,
1. Throughout this article, we use “relative plausibility” and “explanatory account” interchangeably to refer to our theory of proof,
which depends on the relative plausibility of explanations. See Michael S. Pardo & Ronald J. Allen, Juridical Proof and the Best
Explanation, 27 Law & Philosophy 223 (2008). Explanations and explanatory reasoning (i.e., “abduction”) provide the epis-
temological foundation to the theory first developed by Prof. Allen, which focused on the relative plausibility of stories. See
Ronald J. Allen, Factual Ambiguity and a Theory of Evidence, 88 Nw. U. L. Rev. 604, 606 (1994); Ronald J. Allen, The Nature
of Juridical Proof, 13 Cardozo L. Rev. 373 (1991).
2. See, e.g., Fed. R. Evid. 407–11.
3. In re Winship, 397 U.S. 358, 364 (1970); Addington v. Texas, 411 U.S. 418, 424 (1979).
4. For a discussion of these policy goals and standards of proof, see Richard S. Bell, Decision Theory and Due Process: A Critique
of the Supreme Court’s Lawmaking for Burdens of Proof, 78 J. Crim. L. & Criminology 557 (1987).
5. See, e.g., Lisa Kern Griffin, Narrative, Truth, and Trial, 101 Geo. L. Rev. 281, 304–05 (2013); William Twining, Rethinking
Evidence: Exploratory Essays 72–73 (1990). See also Riordan v. Kempiners, 831 F.2d 690, 698 (7th Cir. 1987) (“All evidence
is probabilistic”).
6. For example, do such judgments rigorously follow the axioms of conventional probability so that all ambiguity counts against
the person with the burden of persuasion (in other words, is the judgment about whether the plaintiff’s case is more likely true or
false?), or is the question instead whether the plaintiff’s explanation is more plausible than the defendant’s (with suitable
modifications for other standards higher than a preponderance of the evidence)? Under the first interpretation, the issue is
whether Xor not-X is more probable (with Xþnot-X summing to 1.0); under the second, is it about whether Xor Yis more
plausible.
6The International Journal of Evidence & Proof 23(1-2)
whether the defendant does as well (even if less plausible than the prosecution’s).
7
The more robustly
probabilistic explanation, by contrast, sees trials as a process of serial updating probabilistic judgments
on the question of whether one party’s contentions are true or not.
8
The primary disagreement, in other words, is empirical: what is the best explanation of juridical
proof?
9
It is important to resolve this disagreement, if possible, because only with a clear understanding
of what the system is actually doing can the process be effectively critiqued. Once it is clear what the
system is doing, one can decide whether and how it can be improved. But a complexity lurks here, which
is that the American legal system is itself enormously complex. There are 51 or so separate “legal
systems” in operation (actually many more than that) with millions of cases each year.
10
Moreover, the
“American legal system” is not a static entity created with a single purpose in mind; it is to a large extent
an organic, adaptive process, (or, in older but perhaps more familiar Hayekian terms, a grown rather than
a made system).
11
This limits the ability to specify simple theories or explanations that will capture all
the data. There will be outliers. The objective is thus not to identify a single explanation to which there
are no exceptions, but instead to attempt to capture to the extent possible the essence of the object under
inquiry.
From this perspective, our primary aim is to understand the general nature of juridical proof. We
focus on burdens of proof, and how they structure the proof process, as a lens through which to observe
the legal system, but burdens of proof are only a lens. What is being observed is the entire litigation
process, which includes burdens of proof as one crucial component. This task is empirical: what is the
best explanation of the data, where “the data” are o bservations of how the American legal system
structures proof at trial? A second task is to then consider whether the empirically true is normatively
appropriate, in light of the goals of the legal system. In the context of juridical proof, these goals include
7. Pardo & Allen, supra note 1. In general, explanations are evaluated based on criteria such as consistency, coverage, simplicity,
coherence, consilience, and fit with background knowledge. Peter Lipton, Inference to the Best Explanation (2d ed. 2004);
Gilbert Harman, Change in View: Principles of Reasoning 65–75 (1986); Tania Lombrozo, Explanation and Abductive
Inference, in The Oxford Handbook of Thinking and Reasoning 260 (Keith J. Holyoak & Robert G. Morrison eds., 2012). We
refer to this process throughout this article as “holistic,” but all we mean by the use of that term is that explanatory con-
siderations structure the process of proof as indicated in the text. In determining the existence and relative persuasiveness of
plausible explanations, any cognitive tool at all may be employed, as indicated above, and employed from the very discrete
level of analyzing the relationship between quite discrete evidentiary proffers and specific propositions to the appraisal of the
overall explanation itself considered as a whole. We note this because some confusion has crept into the pertinent literature as
to the significance of stories, as though story-telling were one explanation of trials, leading to the critique that a good story
may still be false. However that may be, it has nothing to do with the relative plausibility theory, which deals with expla-
nations, not stories. To be sure, a story such as a chronological narrative can provide an explanation, but the story does not
persuade independent of the evidence at trial. It should be obvious that “telling a good story” at trial but failing to provide
evidence of its truth is a recipe for disaster. For more on this, see the text at note 186, infra.
8. See Part I for an overview of this explanation.
9. It is important to note that the significance of this project extends well beyond the trial because the process of proof affects a
wide array of issues throughout the systems of civil and criminal litigation. These issues include settlements and pleas; various
discovery issues; which cases make it to trial in the first place and which are terminated pre-trial; and which verdicts are
upheld or overturned. For an overview of the roles played by evidence and proof in non-trial issues, see Michael S. Pardo,
Some Remarks on the Importance of Evidence Outside of Trials, 36 Rev. Lit. 443 (2017). Perhaps most prominently, the
standards of proof determine issues such as summary judgment and judgment as a matter of law in civil cases and the suf-
ficiency of the evidence in criminal cases, each of which depends on what reasonable juries could find based on the evidence
and the standards. See Fed. R. Civ. P 50, 56; Anderson v. Liberty Lobby, Inc. 477 U.S. 242, 248, 252 (1986); Reeves v.
Sanderson Plumbing Prods., Inc., 530 U.S. 133, 149–50 (2000); Fed. R. Crim. P. 29; Jackson v. Virginia, 443 U.S. 307, 319
(1979).
10. See, e.g, National Center for State Courts, The Landscape of Civil Litigation in State Courts 6. N. 36 (2015), available at:
http://www.ncsc.org (“In 2013, litigants filed approximately 16.9 million civil cases in state courts compared to 259,489 civil
cases filed in U.S. District Courts.”). The same report estimates over half a million trials (32,000 in a sample of five percent of
the cases). Id. at 25.
11. See Ronald J. Allen, Taming Complexity: Rationality, the Law of Evidence, and the Nature of the Legal System, 12 Law,
Probability & Risk 99 (2013).
Allen and Pardo 7
To continue reading
Request your trial