The primary aim of this chapter is to explain the nature of evidential reasoning, the characteristic difficulties encountered, and the tools to address these difficulties. Our focus is on evidential reasoning in criminal cases. There is an extensive scholarly literature on these topics, and it is a secondary aim of the chapter to provide readers the means to find their way in historical and ongoing debates.

In place of the traditional epistemological view of knowledge as justified true belief we argue that artificial intelligence and law needs an evidence-based epistemology according to which scientific knowledge is based on critical analysis of evidence using argumentation. This new epistemology of scientific evidence (ESE) models scientific knowledge as achieved through a process of marshaling evidence in a scientific inquiry that results in a convergence of scientific theories and research results. We show how a dialogue interface of argument from expert (...) opinion, along with its set of critical questions, provides the argumentation component of the ESE. It enables internal scientific knowledge to be translated over into a wider arena in which individual nonexpert citizens and groups can make use of it. The external component shows how evidence is presented and used in a legal procedural setting that includes fact-finding, weighing the credibility of expert witnesses, and critical questioning of arguments. The paper critically reviews the standards of admissibility of scientific evidence using the ESE. (shrink)

Legal probabilism is the view that juridical fact-finding should be modeled using Bayesian methods. One of the alternatives to it is the narration view, according to which instead we should conceptualize the process in terms of competing narrations of what happened. The goal of this paper is to develop a reconciliatory account, on which the narration view is construed from the Bayesian perspective within the framework of formal Bayesian epistemology.

This paper defends the heretical view that, at least in some cases, we ought to assign legal liability based on purely statistical evidence. The argument draws on prominent civil law litigation concerning pharmaceutical negligence and asbestos-poisoning. The overall aim is to illustrate moral pitfalls that result from supposing that it is never appropriate to rely on bare statistics when settling a legal dispute.

This review article of Stumer (The presumption of innocence: evidential and human rights perspectives. Hart Publishing, Oxford, 2010) explores the concept, normative foundations and institutional implications of the presumption of innocence in English law. Through critical engagement with Stumer’s methodological assumptions and normative arguments, it highlights the narrowness of common lawyers’ traditional conceptions of the presumption of innocence. Picking up the threads of previous work, it also contributes to on-going debates about the legitimacy of reverse onus clauses and their compatibility (...) with European human rights law and general principles of criminal jurisprudence. (shrink)

This paper examines Lockie’s theory of libertarian self-determinism in light of the question of prediction: “Can we know (or justifiably believe) how an agent will act, or is likely to act, freely?” I argue that, when Lockie's theory is taken to its full logical extent, free actions cannot be predicted to any degree of accuracy because, even if they have probabilities, these cannot be known. However, I suggest that this implication of his theory is actually advantageous, because it is able (...) to explain and justify an important feature of the practices we use to determine whether someone has acted culpably: our hostility to the use of predictive evidence. (shrink)

The “beyond a reasonable doubt” standard of proof, currently used in criminal trials, is notoriously vague and undermotivated. This paper discusses two popular strategies for justifying our choice of a particular precise interpretation of the standard: the “ratio-to-standard strategy” identifies a desired ratio of trial outcomes and then argues that a certain standard is the one that we can expect to produce our desired ratio, while the “utilities-to-standard strategy” identifies utilities for trial outcomes and then argues that a certain standard (...) maximizes expected utility. I argue that both strategies fail on their own terms, by requiring us to perform calculations that we simply cannot perform. No version of either strategy can be performed by jurors or legislators in our actual epistemic position, in which, since we do not know which of the defendants in our trial system are genuinely innocent and which are genuinely guilty, we cannot determine the extent to which our trial system tends to produce evidence that misleadingly incriminates the innocent or misleadingly exonerates the guilty. But we would need to determine this in order to perform the calculations required by any possible version of the ratio-to-standard or utilities-to-standard strategies. I then suggest some empirical reasons to be pessimistic about the evidence produced by our actual trial system. The upshot is that the “beyond a reasonable doubt” standard lacks a clear interpretation and rationale, nor do we have a promising way to identify an alternative. This is the trouble with standards of proof. (shrink)

Explanationist accounts of rational legal proof view trials as a competition between explanations. Such accounts are often criticized for being underdeveloped. One question in need of further attention is when guilt is proven beyond a reasonable doubt in criminal trials. This article defends an inference to the best explanation -based approach on which guilt is only established BARD if the best guilt explanation in a case is substantially more plausible than any innocence explanation, and there is no good reason to (...) presume that we have overlooked evidence or alternative explanations that could realistically have exonerated the defendant. This is a comparative account, which I argue is better suited for arriving at accurate verdicts than the non-comparative ‘no plausible alternative’ account that many explanationists tacitly assume. Furthermore, this account is not susceptible to the most important arguments against IBE in criminal trials or to arguments against other, non-explanationist interpretations of the BARD standard. I use a case study to illustrate how this account provides meaningful guidance for decision makers in criminal trials. (shrink)

In criminal trials the state must establish, to a particular standard of proof, the defendant's guilt. The most widely used and important standard of proof for criminal conviction is the ‘beyond a reasonable doubt' standard. But what legitimates this standard, rather than an alternative? One view holds the standard of proof should be determined or justified – at least in large part – by its consequences. In this spirit, Laudan uses crime statistics to estimate risks the average citizen runs of (...) being violently victimised and falsely convicted. He argues that since the former risk is higher, and the aggregate harms are worse, the standard of proof should be substantially lowered. He presents a formula for calculating the preferred standard. In this article I outline various ways Laudan's uses of crime statistics are flawed, and explain how he substantially overestimates risks of victimhood and underestimates costs of false convictions. I also explain why his formula is mistaken, and illuminate consequences Laudan neglects. I conclude that, even if consequences determine the appropriate standard of proof, Laudan's arguments fail to show the standard is too high. I conclude by suggesting that the inadequacies of Laudan's reasoning might be good news for consequence-based justifications of the standard of proof. (shrink)

When the state aims to prevent responsible and dangerous actors from harming its citizens, it must choose between criminal law and other preventive techniques. The state, however, appears to be caught in a Catch-22: using the criminal law raises concerns about whether early inchoate conduct is properly the target of punishment, whereas using the civil law raises concerns that the state is circumventing the procedural protections available to criminal defendants. Andrew Ashworth has levied the most serious charge against civil preventive (...) regimes, arguing that they evade the presumption of innocence. After sketching out a substantive justification for a civil, preventive regime, I ask what Ashworth’s challenge consists in. It seems that there is broad disagreement over the meaning and requirements of the presumption of innocence. I thus survey the myriad possibilities and extract two claims that have potential bearing on preventive regimes. One claim is that of substantive priority—the criminal law comes first when assessing blame. This is the claim at the root of objections to pretrial detention based on consideration of the crime charged. The second strand of argument is one of procedural symmetry. This is the concern that with respect to citizen/state relations, certain procedures are required, including, for example, proof beyond a reasonable doubt as to the offense or defense. Having extracted these claims, I then assess their applicability with respect to the preventive regime defended. I first conclude that the criminal law must share blame and censure with other fora, and thus, the criminal law only has substantive priority when criminal proceedings have been instituted. I then survey whether procedural symmetry is required, specifically assessing whether the preventive regime I defend requires proof beyond a reasonable doubt. My tentative conclusion is that proof beyond a reasonable doubt is warranted. (shrink)

Many philosophers have pointed out that statistical evidence, or at least some forms of it, lack desirable epistemic or non-epistemic properties, and that this should make us wary of litigations in which the case against the defendant rests in whole or in part on statistical evidence. Others have responded that such broad reservations about statistical evidence are overly restrictive since appellate courts have expressed nuanced views about statistical evidence. In an effort to clarify and reconcile, I put forward an interpretive (...) analysis of why statistical evidence should raise concerns in some cases but not others. I argue that when there is a mismatch between the specificity of the evidence and the expected specificity of the accusation, statistical evidence—as any other kind of evidence—should be considered insufficient to sustain a conviction. I rely on different stylized court cases to illustrate the explanatory power of this analysis. (shrink)

Smith argues that, unlike other forms of evidence, naked statistical evidence fails to satisfy normic support. This is his solution to the puzzles of statistical evidence in legal proof. This paper focuses on Smith’s claim that DNA evidence in cold-hit cases does not satisfy normic support. I argue that if this claim is correct, virtually no other form of evidence used at trial can satisfy normic support. This is troublesome. I discuss a few ways in which Smith can respond.

Suppose one hundred prisoners are in a yard under the supervision of a guard, and at some point, ninety-nine of them collectively kill the guard. If, after the fact, a prisoner is picked at random and tried, the probability of his guilt is 99%. But despite the high probability, the statistical chances, by themselves, seem insufficient to justify a conviction. The question is why. Two arguments are offered. The first, decision-theoretic argument shows that a conviction solely based on the statistics (...) in the prisoner scenario is unacceptable so long as the goal of expected utility maximization is combined with fairness constraints. The second, risk-based argument shows that a conviction solely based on the statistics in the prisoner scenario lets the risk of mistaken conviction surge potentially too high. The same, by contrast, cannot be said of convictions solely based on DNA evidence or eyewitness testimony. A noteworthy feature of the two arguments in the paper is that they are not confined to criminal trials and can in fact be extended to civil trials. (shrink)

Arguments on legal evidence rely on generalizations, that link a certain circumstance to a certain hypothesis and warrants the claim that the circumstance makes the hypothesis more probable. Some generalizations are acceptable and others are unacceptable. A generalization can be unacceptable on at least four different grounds. A false generalization is unacceptable because membership in the reference class does not increase the probability of the hypothesis. A non-robust generalization is unacceptable because it uses a reference class that is too heterogeneous. (...) A biastriggering generalization is unacceptable because decision makers are inclined to overestimate the evidentiary value of membership in the reference class. A discriminating generalization is unacceptable because it puts members in the reference class in an unfair disadvantage. Research funded by the Swedish Research Council. (shrink)

El trabajo explora la controversia entre las concepciones atomistas y holistas del razonamiento probatorio con el objeto de evaluar críticamente sus contribuciones desde la perspectiva de una teoría normativa de la justifi cación de la decisión judicial sobre los hechos. La autora reconoce a las aproximaciones holistas el mérito de haber iluminado la relevancia semántica de la integración de las proposiciones a probar en un relato global del caso y la relevancia justifi cativa de la integración explicativa entre cada una (...) de ellas y el conjunto de pruebas disponibles. Pero a la vez sostiene que, por razones epistémicas y jurídicas, en la arquitectura de la justifi cación probatoria cada hecho principal debe mantenerse atomísticamente diferenciado como objeto de prueba y respecto de cada elemento de prueba debe analizarse en forma individualizada su fi abilidad y la solidez de la inferencia que lo conecta con un hecho principal. (shrink)

I defend a deontological social contract justification of punishment for free will deniers. Even if nobody has free will, a criminal justice system is fair to the people it targets if we would consent to it in a version of original position deliberation (OPD) where we assumed that we would be targeted by the justice system when the veil is raised. Even if we assumed we would be convicted of a crime, we would consent to the imprisonment of violent criminals (...) if prison conditions were better than the state of nature but deterring enough to prevent the state of nature. And even if we assumed we would be accused of a crime, we would consent to an evidentiary standard low enough to allow a conviction rate sufficient to prevent the state of nature. After explaining this justification, I contrast it with the public health quarantine justification defended by Derk Pereboom and Gregg Caruso, and argue that it provides a stronger response to the using criminals as mere means objection. (shrink)

The standard of proof in criminal trials should require that the evidence presented by the prosecution is robust. This requirement of robustness says that it must be unlikely that additional information would change the probability that the defendant is guilty. Robustness is difficult for a judge to estimate, as it requires the judge to assess the possible effect of information that the he or she does not have. This article is concerned with expert witnesses and proposes a method for reviewing (...) the robustness of expert testimony. According to the proposed method, the robustness of expert testimony is estimated with regard to competence, motivation, external strength, internal strength and relevance. The danger of trusting non-robust expert testimony is illustrated with an analysis of the Thomas Quick Case, a Swedish legal scandal where a patient at a mental institution was wrongfully convicted for eight murders. (shrink)

Existing discussions of legal proof address a host of apparently disparate questions: What does it take to prove a fact beyond a reasonable doubt? Why is the reasonable doubt standard notoriously elusive, sometimes considered by courts to be impossible to define? Can the standard of proof by a preponderance of the evidence be defined in terms of probability thresholds? Why is statistical evidence often insufficient to meet the burden of proof? -/- This paper defends an account of proof that addresses (...) each of these questions. Where existing theories take a piecemeal approach to these puzzles, my theory develops an insight that unifies them—namely, the thesis that legal proof requires knowledge. Although this thesis may seem radical at first, I answer several common objections to it, and I argue that it does better than several competing accounts when it comes to making sense of our intuitions about what legal proof requires. (shrink)

In the field of Evidence Law the relation between truth and law has been a somewhat non- debated topic in the past years. It is a given that such a connection exists and, it is understood as the notion of the ascertainment of the truth of disputed questions of fact through legal evidence. The thesis that I have reconstructed in this paper has been presented by Prof. Jordi Ferrer who grounds the connection between truth and law in the role of (...) the legal system as a tool for guiding the behavior of the governed. I call this thesis the theory of the structural clause of truth. In this paper I shall present a possible reconstruction of this theory, I shall raise two objections against it and I shall present five possible interpretations of the relation between truth and law. (shrink)

Why can testimony alone be enough for findings of liability? Why statistical evidence alone can’t? These questions underpin the “Proof Paradox” (Redmayne 2008, Enoch et al. 2012). Many epistemologists have attempted to explain this paradox from a purely epistemic perspective. I call it the “Epistemic Project”. In this paper, I take a step back from this recent trend. Stemming from considerations about the nature and role of standards of proof, I define three requirements that any successful account in line with (...) the Epistemic Project should meet. I then consider three recent epistemic accounts on which the standard is met when the evidence rules out modal risk (Pritchard 2018), normic risk (Ebert et al. 2020), or relevant alternatives (Gardiner 2019 2020). I argue that none of these accounts meets all the requirements. Finally, I offer reasons to be pessimistic about the prospects of having a successful epistemic explanation of the paradox. I suggest the discussion on the proof paradox would benefit from undergoing a ‘value-turn’. (shrink)