Quaestio facti. Revista Internacional sobre Razonamiento Probatorio
Quaestio facti. International Journal on Evidential Legal Reasoning
Sección: Conjeturas y refutaciones
2025 l 9 pp. 225-239
Madrid, 2025
DOI: 10.33115/udg_bib/qf.i9.23151
Marcial Pons Ediciones Jurídicas y Sociales
© Paul Roberts
ISSN: 2604-6202
Recibido: 11/05/2025 | Aceptado: 25/06/2025 | Publicado online: 30/06/2025
Editado bajo licencia Reconocimiento 4.0 Internacional de Creative Commons

Standards and Methods of Proof: An English Perspective on Della Torre’s Comparative Legal History

Paul Roberts*

Abstract: This Comment responds to Jacopo Della Torre’s recent article advocating Taking the Evolution of the Standards of Proof for Criminal Conviction Seriously. Having summarised Della Torre’s central thesis and clarified (and largely endorsed) his contextual approach to investigating the criminal standard of proof, I offer—from the perspective of an English common lawyer—some methodological caveats, conceptual reservations and minor textual corrections to Della Torre’s illuminating comparative legal history.

KEYWORDS: standards of proof, beyond reasonable doubt, intime conviction, forensic epistemology, comparative criminal procedure, comparative legal history

SUMMARY: 1. METHODS OF INQUIRY.— 2. (MIS)UNDERSTANDING STANDARDS OF PROOF.— 3. CONCLUDING REMARKS: THE BURDENS OF COMPARATIVE LEGAL SCHOLARSHIP.— BIBLIOGRAPHY

In an erudite and wide-ranging contribution to this Revista, Jacopo Della Torre leverages the analytical power of comparative legal history to illuminate contemporary debates surrounding the standard of proof for criminal convictions. At the invitation of the Editors, I am pleased to have this opportunity to comment on Della Torre’s thought-provoking article. The following remarks are of two broad kinds. The first section of this Comment addresses methodological issues in comparative legal scholarship, largely expressing agreement with Della Torre’s general approach, but with a few caveats and clarifications for further consideration. In the second section, I turn to practical questions of procedural jurisprudence and institutional practice in criminal adjudication. With the disciplinary agenda and biases of an English lawyer, my thoughts on these issues will embroider upon, and diverge somewhat, from Della Torre’s exposition. I will also suggest some minor exegetical corrections and refinements.

1. Methods of Inquiry

Della Torre (2025) begins with a puzzle in Evidence law theory, which he dramatises as “a bitter dispute between those who, for legal and moral reasons, consider it possible and/or desirable to establish objective or intersubjective thresholds of proof, and those who, for various reasons, oppose this thesis” (p. 156) 1. The debate, encapsulated, is between objectivists and subjectivists regarding standards of proof. Much of the—extensive and burgeonin—literature engaged in this debate adopts an analytical approach preoccupied with conceptual definition and speculative theoretical modelling, increasingly informed by philosophical, in particular epistemological, perspectives 2. Della Torre takes a markedly different tack, tracing the historical evolution of European criminal procedures and grounded in institutional practices. From this vantage point, we can see that common law and civilian jurisprudence share an Enlightenment intellectual framework and an even older theological tradition, both of which continue to inform contemporary criminal trial practice. On the surface, it may appear that the common law’s iconic standard of proof “beyond reasonable doubt” (“BARD”) is a world away from the civilians’ intime conviction. But dig a little deeper and the commonalities become more striking than any superficial terminological differences. As Della Torre convincingly shows,

the problems of the standard of proof were in many ways the same, as were the lexicon, the positions and the solutions…. [M]oral certainty and the BARD are concepts that have deep transversal roots in the culture—theological, philosophical and, later, legal—of the entire West (2025, p. 177).

Jurisprudence is predominantly normative and analytical, with a corresponding tendency to being unhistorical. More nuanced understanding of the historical facts reveals that the medieval Schoolmen were not as callous or dim-witted as they can be made to seem in twenty-first century potted retrospectives 3; and that the transition to Enlightened law reform across Europe, inaugurated by the French revolutionaries at the birth of modernity (Johnson, 1991; Hobsbawn, 1988), was neither sudden nor complete 4. Della Torre (2025) observes that “the clash between more subjective and more objective notions of the standards of proof is not a contemporary novelty, but rather a historically unresolved issue» (p. 163). Moreover, medieval theologians developed the concepts of “moral certainty” and reasonable doubt that would eventually pass through centuries of Roman-canon law into modern European criminal jurisprudence (Whitman, 2008; Shapiro, 1991; Waldman, 1959). Formalistic, quasi-mathematical approaches to assessing evidential weight could not withstand the combined force of rationalist critiques and the realisation that, in practice, jurisprudential formalism merely licensed untrammelled judicial discretion in the assessment of evidence. So «free evaluation» became the new Enlightenment mantra. Crucially, however, the French revolutionaries assumed that lay juries on the English model would be entrusted with performing such evaluations, so that the common sense of ordinary citizens would anchor criminal adjudication in objective (or at least conventional, “inter-subjective”) epistemic standards. What actually transpired, for much of western Europe, was that professional judges edged out juries as the arbiters of fact and in many legal systems replaced them entirely. In the process, the institutional meaning of intime conviction (and its terminological equivalents in other European languages) was transformed from a community standard of evidence appraisal into a measure of subjective personal belief, allowing rampant judicial discretion back into the trial process and precipitating further rounds of critical debate and reform.

[A]s was to be expected, when systems based on professional judges or mixed courts broke the original link between intimate conviction and the jury, the need gradually arose to structure a theory of evidence, free from legal proof, but nevertheless based on respect for the rules of logic, science and reason (Della Torre, 2025, p. 183).

Although the broad outlines of this story are well-known 5, Della Torre’s compact retelling makes for fascinating reading, not least form a common lawyer’s perspective. Whereas comparative analysis of criminal procedure often proceeds at the level of shallow system-wide institutional comparisons, and is frequently reductive in taking some version of French legal process as the standardised model of civilian criminal trial, Della Torre provides more granular jurisprudential analysis accommodating doctrinal and terminological variations characteristic of Italian, German and other national legal systems. This attention to jurisprudential detail resonates with my own preferred method of “common law comparativism” and its programmatic plea for sensitivity to the differences within the common law family of procedural systems 6. Such differences become more prominent, and significant in practice, as the comparative lens focuses more precisely on detailed institutional artefacts such as juridical conceptions of standards of proof. A second major theme in this narrative is the enduring salience of cultural transplants in criminal procedure. Having briefly summarised how “the BARD has now taken on a global dimension… codified in the criminal procedure codes of many national legal systems” (Della Torre, 2025, p. 192), Della Torre concedes (not without a whiff of regret) that

all the success that the standard of reasonable doubt in particular is enjoying at the political and jurisprudential level in many legal systems… makes it prohibitive to think that it too can really be abandoned, at least in the short term, in favour of another formula (2025, p. 198).

Albeit that the global diffusion of “beyond reasonable doubt” may be driven as much, one suspects, by Hollywood cultural tropes as by Strasbourg judgments, this is clearly another important and valuable illustration of legal cosmopolitanism (Roberts, 2010) to add to the Evidence scholar’s travelogue 7.

Whenever we consciously undertake comparative legal scholarship, we also implicitly embody and enact it (Nelken, 1995; 2000). Cultural frames of reference ingrained within disciplinary education and professional identities are often difficult to see, much less to externalise and subject to reflexive critical scrutiny. There is no escaping the fact that I am an English common lawyer, and Della Torre is an Italian civilian. Some implications of these different perspectives will become more apparent in the next section. The antidote to the perspectival nature of comparative scholarship is not to deny cultural conditioning, which would be methodologically suspect even if it were psychologically possible, but rather to recognise and try to understand and manage or mitigate its influence. In a time when national differences in criminal justice are too often politicised, and even weaponised, with motivations antithetical to scholarship (and to justice) the analytical significance of standpoint is worth underlining as a valuable methodological heuristic.

A third notable feature of Della Torre’s framing of the issue, viewed from a common law perspective, is a taxonomical assumption that he does not even acknow­ledge, much less defend: namely, his exclusive focus on criminal procedure in addressing standards of proof. Steeped in a juridical and pedagogical tradition in which criminal procedure and civil procedure are distinct disciplinary domains with their own legislated codes, institutional frameworks and professional experts, the rationale for treating proof standards for criminal conviction as a discrete topic might seem self-evident to a civilian jurist. In the common law world, by contrast, Evidence law is conventionally conceptualised as a generic, “trans-substantive” discrete sub- part of procedural law, and the topic generally known as “Burdens and Standards of Proof” would, by default setting, embrace private law “civil” actions as well as criminal trials. As it happens, I believe that there are compelling arguments for treating criminal evidence as a distinctive disciplinary domain separate from civil procedure (Roberts, 2020; 2022b). But I am not in this regard an orthodox common lawyer. Della Torre’s implicit decision to exclude private law from his history of standards of proof might be a source of puzzlement to some common law readers, who might then be inclined to interpret his article as an exercise in institutional legal history or comparative criminal procedure rather than Evidence law stricto sensu.

Legal history can explain how, and why, current institutional structures came into being, and—to the extent that established trends tend to continue and outcomes are path—dependent – it is a useful tool in predicting future developments. Careful historical inquiry may correct prevailing misconceptions and provide an antidote to legal scholarship’s characteristic amnesia and ahistoricism. Della Torre’s exposition skilfully employs each of these virtues to telling effect. However, historical inquiry cannot settle normative (evaluative) questions; and to the extent that conceptual analysis reflects normative choices, it cannot decisively arbitrate between concepts, either. It is not strictly necessary to undertake any historical research to pose the normative question: how should we define the standard of proof in criminal adjudication? Indeed, as already stated, this is precisely the point of departure for much analytical jurisprudence addressing legal standards of proof. When in the final section of his article Della Torre shifts gears from historical retrospective and contemporary cosmopolitan survey to questions of institutional design and reform, he is, in a methodological sense, changing the subject.

2. (Mis)understanding Standards of Proof

There are multiple aways of conceptualizing and thinking about evidence law and criminal procedure, and my version will inevitably reflect a common law perspective. This insight should recommend a threshold commitment to methodological pluralism, openness to alternative viewpoints, and a measure of humility in propounding one’s own worldview. The best version of evidence law will be best for me and those who share my theoretical and practical ambitions and concerns. Those with different objectives, institutional reference points or disciplinary affiliations may prefer, and possibly require, different approaches. What follows is offered in the spirit of shared, reflective but unapologetic comparative discussion, without any aspiration to disciplinary imperialism or pretention to cultural superiority.

Having avoided the trap of conflating British and American approaches to standards of proof into an homogenized “common law” tradition, Della Torre (2025) organises his narrative in terms of an “English divergence” form “the US experience” (p. 185-190). Setting aside the historical detail that it was the American colonists who originally “diverged” (rebelled) from English law in 1776, this framing makes sense if one takes “beyond reasonable doubt” as the orthodox common law standard of proof for criminal convictions, inasmuch as English law has lately rejected (what Americans call) BARD in favour of instructing juries that they must be “sure” of guilt before convicting (Roberts, 2022b, §6.4). Yet this sequencing sacrifices strict historical accuracy, because—as Della Torre’s own exposition methodically demonstrates—BARD was long-predated by concepts of “moral certainty” and looser conceptions of rational (“reasonable”) doubt and it was only comparatively recently that orthodox legal formulations were distilled into doctrinal boilerplate and backed by prescriptive judicial authority. In Della Torre’s own pithy encapsulation:

In time, the BARD formula, which was preferred to the original lexicon also for terminological reasons—since it did not contain direct references to “morality” but to human “reasonableness”—was used to indicate the level to be reached for a criminal conviction…. [W]hile moral certainty and BARD were originally two formulas used to refer to a single standard of decision-making, over time the latter concept gradually separated from its original core and took on an autonomous scope and meaning, not least because it was seen as more easily adaptable to contemporary society (2025, p. 186-188)

In fact, this origin story is not uniquely American. English law has consistently maintained a somewhat flexible approach to the precise formulation of the standard of proof and, crucially, the manner of its judicial communication to juries. As the Court of Appeal routinely affirms, “[e]xperienced judges are accustomed in their courts to fashion their directions to a jury according to their experience and judgment of the jury in front of them and of the facts of the case. This flexibility is to be encouraged” 8. In terms of terminological precision, Woolmington (for England and Wales) 9 and Winship (for US federal law) 10 are not jurisprudentially identical twins. The claim that “the legal system in England and Wales has, since the late 1940s, attempted to make an important departure in this respect” (Della Torre, 2025, p. 189) is therefore not an accurate summary.

Although proof beyond reasonable doubt has been described as “the time-honoured formula” 11, and it does appear in several important legislative provisions 12, the “sure” direction is also woven into the fabric of English criminal jurisprudence 13. According to Lord Goddard CJ, “sure” represents the more authentic common law inheritance 14 and, besides, semantic quibbles have been regarded as an unwelcome distraction from the more important business of getting juries to understand that criminal convictions require an appropriately exacting standard of proof 15. The phrase “beyond reasonable doubt” appears to have entered the (reported) judicial lexicon during the mid-nineteenth century. It turns up in a judgment of Kindersley VC in Way v East (1853) 16 in relation to the validity of a trust deed; and in several other civil cases in the 1860s and 70s 17. Courts, counsel and case reporters routinely employed the alternative language of “moral certainty” during the eighteenth and earlier part of the nineteenth centuries, right up to the 1860s 18, when it seems to peter out. A regular system of criminal appeals was not established in England and Wales until 1908 19, and there are consequently few reported criminal cases before that time. So it is not entirely surprising that the first reported references to proof “beyond reasonable doubt” in English jurisprudence concern fact-finding by judges in courts of equity, not criminal trials as one might expect today. Stephen’s influential history of English criminal law, published in 1883, asserts as an established and unremarkable fact that, “[i]f the commission of a crime is directly in issue in any proceedings, civil or criminal, it must be proved beyond reasonable doubt… This is otherwise stated by saying that the prisoner is entitled to the benefit of every reasonable doubt” (1996, p. 438) 20. The same standard was applied without argument or objection in a 1900 reported case concerning contempt of court 21. This orthodoxy prevailed for the remainder of the twentieth century, yet without silencing dissenting voices such as Lord Chief Justice Goddard’s propounding a revisionist interpretation of common law history. The programmatic campaign against “beyond reasonable doubt” directions only begins in earnest in the early 2000s, spearheaded by the Judicial College (formerly, the Judicial Studies Board) and backed up with precedential muscle by senior members of the Court of Appeal 22. Whilst some judges have used “sure” and references to “reasonable doubt” more or less interchangeably, the orthodox position in English law is now that judges should preferably avoid the phrase “beyond reasonable doubt” entirely 23. If complete avoidance is not viable, for example because counsel has already used the phrase in argument or the jury during its deliberations requests greater definitional clarity, trial judges should say only that “being satisfied beyond a reasonable doubt mean[s] the same as being sure” 24 and add nothing further. The temptation to editorialise should be resisted, since trying to say more is liable to land trial judges in hot water on appeal, and might even result in an otherwise perfectly sound conviction being quashed because it is no longer procedurally “safe” 25.

This notable shift in English criminal trial practice raises a host of theoretical, jurisprudential, practical and even constitutional questions that cannot be fully canvassed here. But several further clarifications are in order. Having accurately paraphrased Roberts and Zuckerman (Roberts, 2022b) in asserting that “the use of this term [‘sure’] is not so much intended to change the level of the standard of proof with regard to the past… but rather to use a locution that is considered easier for juries to understand” (Della Torre, 2025, p. 189), Della Torre continues:

The problem, however, is that the new English-language formulation of the standard of proof for a criminal conviction is no less vague than its predecessors. This is confirmed not only by empirical studies showing that jurors find even this phrase difficult to understand, but also by the Court of Appeal of England and Wales itself, which finally admitted that ‘to define what is meant by “reasonable doubt” or what is meant by “being sure” requires an answer difficult to articulate and likely to confuse’ (2025, p. 190).

And he adds:

This approach is further confirmed at an official level by the Crown Court Compendium, which states that “it is unwise to elaborate on the standard of proof” (Judicial College, 2023, § 5 n. 3). However, as Roberts and Zuckerman have rightly pointed out, it is impossible not to notice how this solution ultimately leads to an “abdication of judicial responsibility! (Roberts and Zuckerman, 2022, p. 279), making defendants pay for the system’s inability to provide a common minimum definition of the standard for a criminal conviction (2025, p. 190).

These passages contain several misapprehensions, understandable in context, but demanding correction to avoid misleading readers, not merely in terms of my own arguments, but far more importantly, as to the correct interpretation of English law and judicial practice.

First, when the Court of Appeal warns that trial judges should avoid getting into knotty definitional questions, this is not intended as any kind of “admission” or concession, but rather an instruction to avoid pointless definitional speculation. As far as English criminal procedure law is concerned, “sure” is as precise and meaningful as it can be and needs to be for the purposes of a criminal trial. It is an ordinary English word which the jury is presumptively competent to understand and apply to the evidence in the case, with some judicial prompting and general guidance. Della Torre seems to fall into the common trap of thinking that because a normative standard is somewhat open-ended and susceptible to a variety of interpretations, this necessa­rily creates institutional problems requiring remedial attention. In fact, English legal tradition treats the process of jury trial as an acceptable surrogate for greater definitional precision. That is to say, the adjudicative process is a functional alternative to more refined (and abstruse) conceptual definitions. One can certainly debate, as a normative proposition, whether this institutional arrangement is a desirable feature of criminal trial procedure (as a practical component of idealized criminal justice); and the extent to which the normative design is successfully translated into criminal trial practice is partly an empirical question susceptible, in principle 26, to empirical investigation and assessment. But in terms of its own internal self- conception and legitimizing rationale, it needs to be understood that the preference for process over definition is part of the blueprint, not a design flaw, in the institutional architecture of English criminal procedure.

Epistemologists, economists and analytical jurists often seem unaware that in preoccupying themselves with linguistic refinements to the standard of proof they are largely ignoring the practical concerns of criminal adjudication as it is organised in England and Wales and elsewhere in the broader common law tradition. Without getting embroiled in deeper controversies 27, definitional approaches to standards of proof are flagrantly at odds with institutional realities. The standard of proof for conviction in England and Wales is communicated to criminal trial juries as a nested series of practical instructions, rather than as an isolated phrase or one-word canonical definition 28. And the standard of proof direction is itself contextualised within the trial judge’s comprehensive summing-up on the facts of the case 29 and the applicable law, nowadays typically supported by a written “route to verdict” structuring the jury’s deliberations 30. In short, even “defining” the proof standard is an explanatory process in English criminal trials rather than (merely) a prescriptive norm!

Behavioural scientists likewise traduce the adjudicative enterprise when they conduct experimental research demonstrating that ordinary people (potential jurors in criminal trials) understand phrases like “beyond reasonable doubt” in different ways, lacking consistency or standardised calibration, or report that experimental subjects invited to express “being sure” in percentage terms produce widely divergent estimates, ranging from the mid-60s to 100%. It doesn’t seem to occur to these researchers that jurors in criminal trials are never actually asked to define proof standards or to estimate numerical probabilities for reasonable doubt. They are simply instructed to deliberate together to consider whether the evidence and arguments in the case makes them sure the accused is guilty. Again, one might or might not regard this model of criminal adjudication as noble, rational or effective, but it is how criminal trials currently actually operate in England and Wales. It would be surprising if lay jurors chosen at random from the electoral register needed to be well-versed in epistemological subtleties or competent in Bayesian probabilistic reasoning in order to discharge their mandated civic function in criminal adjudication 31. Although the meaning and application of the standard of proof do occasionally generate controversy in England and Wales, in the vast majority of criminal trials taking place up and down the land on a daily basis—thousands of Crown Court trials annually—juries decide that they are “sure” of guilt, or not sure and acquit, without any apparent difficulty; just as they applied the “beyond reasonable doubt” standard with minimum fuss when that was the preferred juridical formulation. From the perspective of accumulated judicial experience, it would be more representative, and perhaps more illuminating, to investigate how such widespread consensus is routinely achieved in the absence of semantic precision, rather than purporting to infer institutional weakness from experimental subjects’ limited powers of conceptual analysis and probabi­listic reasoning 32.

When Della Torre recycles familiar criticisms of current trial practice in England and Wales, and appears to adopt them without qualification, he is taking too much on trust from system insiders with their own viewpoints and agendas 33. A notorious peril of comparative research is knowing whom to trust, given that local experts typically adopt a range of positions on contentious issues. As a general rule, it is safer to rely on insiders’ exegesis of doctrinal law than on causal propositions employing interdisciplinary research methods or normative arguments for reform. Perhaps Della Torre’s impressionability is partly explained by his familiarity with Italian criminal trial procedure, in which the factfinders are professional judges who must provide written justifications for their verdicts. Whilst the cognitive demands of rational factfinding may not differ very much as between lay juries and professional judges 34, the task of writing a reasoned judgment on the facts should be viewed as an important and challenging intellectual exercise in its own right. For this, performative purpose—ex post facto justification, rather than primary fact-finding—greater conceptual precision in definition and implementation of proof standards might well be desirable or at least desired by judicial factfinders anxious for more concrete jurisprudential guidance in performing a difficult task. Della Torre suggests that the notion of “standards of proof” should be stretched to include aspects of judicial reasoning in satisfying the standard:

I think it is important to focus on the fact that the threshold of evidential sufficiency set by the BARD rule can only be said to have been reached if a certain method of reasoning is actually followed in the judgement (2025, p. 199).

This argument is taken to imply that “the standard of proof should… indicate a useful method of reasoning for the evaluation of all types of evidence” (2025, p. 199). Although I am sympathetic to Della Torre’s preference for contextualising criminal proof within a dynamic institutional conception of criminal trial (and pre-trial) process, I would not recommend attempting such conceptual stretching, which is liable to lead to jurisprudential distortion and confusion. Della Torre canvasses “two possible drawbacks to the suggestion of interpreting standards of proof as methods of reasoning”, neither of which gets to the heart of the matter, in my opinion. Reasoning procedures are affected by many and varied factors, normative, institutional, cognitive, cultural, social and more. If the concept of a standard of proof is supposed to incorporate all of these considerations, there is no obvious limit on its imperialist ambitions. I cannot see what is to be gained by making the concept so bloated and indistinct. It makes more sense to me to keep the standard of proof tightly anchored to a normatively defined threshold of evidential sufficiency, whilst being very clear about the distinction between the jurisprudential standard and the institutionally mandated means of satisfying it. A standard of proof is one thing; a comprehensive theory of forensic fact-finding quite another, and a great deal more 35.

Finally, an exegetical clarification. I am afraid that Della Torre has misattributed to Roberts and Zuckerman (Roberts, 2022b) an argument that is canvassed in our pages, but is not a position we endorse. It is, in fact, the opposite of our view. The full passage from which Della Torre’s quotation was extracted reads:

Commentators who proceed from the assumption that the task is to construct a better definition of the prosecutor’s standard of proof in criminal trials almost inevitably regard the law as deficient. Moreover, the judicial strategy of saying as little as possible for fear of making matters worse, originally championed in English law by Lord Goddard CJ and now endorsed by the Judicial College, will appear from this perspective as an abdication of judicial responsibility (Roberts and Zuckerman, 2022, p. 279).

We then briefly summarise, and query, some superficially attractive arguments and data ostensibly reinforcing criticism of English law’s disregard for more punctilious definitions of standards of proof, before setting out our own, more supportive position:

Commentators criticizing judicial directions on the prosecutor’s standard (and burden) of proof have paid too much attention to definitional and conceptual nuance in the choice of words and too little attention to the practical task of explaining to juries how and why their verdict should honour the foundational values of criminal adjudication. Viewed in this light, judicial unwillingness to be drawn into interminable definitional wrangles, and even the occasional off-colour remark to the effect that choices between particular standards might not have much practical significance in the end, are not as silly or hypocritical as they might appear to the casual reader. Whatever words are chosen to express the criminal standard, the jury must be brought to appreciate its function in protecting the innocent from the profound moral harm of wrongful conviction (Roberts, 2022b, p. 280).

Once again, the process is key, and definitions very much secondary 36. The best way of explaining to jurors that, how and why, the standard of proof for a criminal conviction is very exacting is to remind them, preferably in so many words, that it is profoundly wrong and oppressive to convict an innocent person of a criminal offence and that the evidence in the case must be compelling before they are entitled 37 to convict the accused. Analogies to other important life decisions are at best imperfect proxies to spelling out the jury’s civic duty to test the evidence against the highest practical standard of epistemic warrant; at worst, they may operate to cloud and dilute the force of the criminal standard of proof and thereby erode the presumption of innocence 38. At all events, trying to find the magical formula to compress all of these resonant ideas into a single word or phrase is a wild goose chase. The meaning of the proof standard, as Della Torre intimates, is located in the wider institutional context of criminal adjudication. But it does not follow that the juridical standard itself must be stretched to accommodate the reasoning procedures prescribed for satisfying it.

3. Concluding Remarks: The Burdens of Comparative Legal Scholarship

Della Torre’s valuable article demonstrates major strengths of comparative legal history as applied to criminal procedure. Whereas conventional Evidence law on the common law model tends to be ahistorical, doctrinal, insular and static, comparative reconstruction of the historical evolution of legal institutions paints a more holistic and dynamic picture of criminal adjudication capable of correcting widespread misapprehensions and offering fresh insights – so long as we are not tempted by the “genetic fallacy” of assuming that historical antecedents dictate, rather than merely influencing, institutional destiny.

Della Torre’s exposition also exhibits some of the perils of comparative legal scholarship. When researching a foreign legal system, it is difficult to assess which local experts should be trusted, and in respect of what kinds of information or claim. Sources are easily misread and meaning sometimes gets lost in translation. Whilst I cannot help noticing such distortions, especially when I am the author being misrepresented, they are incidental to Della Torre’s central narrative and barely detract from its cogency and illumination. Like many native English speakers, my own foreign language skills are rudimentary. I can only marvel at the ability of scholars, like Della Torre, who are able to operate in more than one professional language. First and foremost, comparative criminal procedure must be a collaborative and inclusive conversation in which we continuously learn from each other, even though, inevitably, it will sometimes be necessary to quibble in order to set the record straight.

Bibliography

Brodowski, D., Fissell, B. and Roberts, P. (2024). Pretext, Deception and Entrapment in Criminal Investigations. In K. Ambos, A. Duff, A. Heinze, J. Roberts and T. Weigend (eds.), Core Concepts in Criminal Law and Criminal Justice (volume 3). Cambridge University Press.

Brook, C. A., Fiannaca, B., Harvey, D., Marcus, P., Pomerance, R. and Roberts, P. (2021). A Comparative Examination of Police Interrogation of Criminal Suspects in Australia, Canada, England and Wales, New Zealand, and the United States. William & Mary Bill of Rights Journal, 29(4), p. 909-1008. https://scholarship.law.wm.edu/wmborj/vol29/iss4/2/

Dahlman, C., Stein, A. and Tuzet, G. (2021). Philosophical Foundations of Evidence Law. Oxford University Press.

Damaška, M. (1997). Rational and Irrational Proof Revisited. Cardozo Journal of International and Comparative Law, 5(25), p. 25-40.

Damaška, M. (2018). Evaluation of Evidence: Pre-Modern and Modern Approaches. Cambridge University Press.

Darbyshire, P., Maughan, A. and Stewart, A. (2001). What Can We Learn from Published Jury Research? Findings for the Criminal Courts Review. Criminal Law Review, p. 970-979

Della Torre, J. (2025). Taking the Evolution of the Standards of Proof for Criminal Conviction Seriously. Quaestio facti. International Journal on Evidential Legal Reasoning, 8, p. 155-216. https://doi.org/10.33115/udg_bib/qf.i8.23112

Ellison, L. and Munro, V. E. (2015). “Telling tales”: exploring narratives of life and law within the (mock) jury room. Legal Studies, 35(2), p. 201-225. https://doi.org/10.1111/lest.12051

Epps, D. (2015). The Consequences of Error in Criminal Justice. Harvard Law Review, 128(4), p. 1065-1151.

Franklin, J. (2001). The Science of Conjecture: Evidence and Probability before Pascal. Johns Hopkins University Press.

Helm, R. K. (2024). How Juries Work. Oxford University Press.

Hobsbawn, E. (1988). The Age of Revolution: 1789–1848. Abacus.

Johnson, P. (1991). The Birth of the Modern: World Society 1815-1830. HarperCollins.

Keane, A. and McKeown, P. (2019). Time to abandon “beyond reasonable doubt” and “sure”: The case for a new direction on the criminal standard and how it should be used. Criminal Law Review, p. 505-527.

Kotsoglou, K. N. (2020). Proof beyond a context-relevant doubt. A structural analysis of the standard of proof in criminal adjudication. Artificial Intelligence and Law, 28(1), p. 111-133.

Marcus, P. (2013). Judges Talking To Jurors in Criminal Cases: Why U.S. Judges Do It So Differently From Just About Everyone Else. Arizona Journal of International and Comparative Law, 30, p. 1-64. https://scholarship.law.wm.edu/facpubs/1660

Nelken, D. (1995). Disclosing/Invoking Legal Culture: an Introduction. Social and Legal Studies, 4(4), p. 435-452. https://doi.org/10.1177/096466399500400401

Nelken, D. (ed.) (2000). Contrasting Criminal Justice. Ashgate.

Picinali, F. (2015). The threshold lies in the method: instructing jurors about reasoning beyond reasonable doubt. International Journal of Evidence and Proof, 19(3), p. 139-153. https://doi.org/10.1177/1365712715571511

Picinali, F. (2022). Justice In-Between: A Study of Intermediate Criminal Verdicts. Oxford University Press.

Roberts, P. (2010). Confronting the Challenges of Cosmopolitan Criminal Jurisprudence. Inaugural Lecture, University of Nottingham, 10 September 2010

Roberts, P. (2020). Adrian Zuckerman’s New Evidence Scholarship. In R. Assy and A. Higgins (eds.), Principles, Procedure, and Justice: Essays in Honour of Adrian Zuckerman. Oxford University Press.

Roberts, P. (2022a). Beccaria Now: (Re)reading On Crimes and Punishment. In A. du Bois-Pedain and S. Eldar (eds), Re- Reading Beccaria: On the Contemporary Significance of a Penal Classic. Hart.

Roberts, P. (2022b). Roberts and Zuckerman’s Criminal Evidence (3rd ed.). Oxford University Press.

Roberts, P. (2023). Theorising Evidence Law. Oxford Journal of Legal Studies, 43(3), p. 629-649. https://doi.org/10.1093/ojls/gqad007

Roberts, P. (2024). Review of Picinali: Justice In-Between: A Study of Intermediate Criminal Verdicts. In Criminal Law and Criminal Justice Books. https://clcjbooks.rutgers.edu/books/justice-in-between/.

Roberts, P. (2025). Reasonable Doubts?: Rationalising Appellate Review of Witness Credibility in Criminal Trials. Canadian Criminal Law Review, 28(1), p. 21-58.

Shapiro, B. J. (1991). Beyond Reasonable Doubt and Probable Cause: Historical Perspectives on the Anglo-American Law of Evidence. California University Press.

Stephen, J. F. (1881). A Digest of the Law of Evidence (4th edition) Macmillan & Co.

Stephen, J. F. (1996) [1883]. A History of the Criminal Law of England (volume 1). Routledge.

Thomas, C. (2013). Avoiding the Perfect Storm of Juror Contempt. Criminal Law Review, p. 483-503. https://discovery.ucl.ac.uk/id/eprint/1390245/1/Chery_Thomas_article__final.pdf

Tuzet, G. (2020). Assessment Criteria or Standards of Proof? An Effort in Clarification. Artificial Intelligence and Law, 28(1), p. 91-109.

Waldman, T. (1959). Origins of the Legal Doctrine of Reasonable Doubt. Journal of the History of Ideas, 20(3), p. 299-316. https://doi.org/10.2307/2708111

Weisselberg, C. D. (2017). Exporting and Importing Miranda. Boston University Law Review, 97, p. 1235-1291. https://www.bu.edu/bulawreview/files/2017/08/WEISSELBERG.pdf

Whitman, J. Q. (2008). The Origins of Reasonable Doubt: Theological Roots of the Criminal Trial. Yale University Press.

Legislation and Jurisprudence

AG v Dean and Canons of Windsor (1860) 11 ER 472, 8 HLC 369.

Criminal Appeal Act 1968, c. 19. https://www.legislation.gov.uk/ukpga/1968/19/contents

Criminal Justice Act 1988, c. 33. https://www.legislation.gov.uk/ukpga/1988/33/contents

Criminal Practice Directions 2023. https://www.judiciary.uk/wp-content/uploads/2023/04/Criminal-Practice-Directions-2023-1.pdf

Domestic Abuse Act 2021, c. 17. https://www.legislation.gov.uk/ukpga/2021/17/contents

Enohin v Wylie (1862) ER 924; 10 HLC 1.

Ferguson v R [1979] 1 WLR 94, PC

Forbes v Meer Mahomed Tuquee (1870) 20 ER 614, 13 Moore Ind App 438.

In re Winship, 397 US 358 (1970).

Judicial College (2024). The Crown Court Compendium (Part I: Jury and Trial Management and Summing Up). https://www.judiciary.uk/wp-content/uploads/2024/09/Crown-Court-Compendium-Part-I-July-2024.pdf

Police and Criminal Evidence Act 1984, c. 60. https://www.legislation.gov.uk/ukpga/1984/60/contents

R v Adams (No.2) [1998] 1 Cr App R 377, CA.

R v Ayre [2025] EWCA Crim 255.

R v Gray (1900) 82 Law Times 534 (16 June).

R v Kritz [1950] 1 KB 82, CCA.

R v Majid [2009] EWCA Crim 2563.

R v Smith (Michael William) [2012] EWCA Crim 404.

R v Summers (1952) 36 Cr App R 14, CCA.

R v Yap Chuan Ching (1976) 63 Cr App R 7, CA.

Road Traffic Act 1988, c. 52. https://www.legislation.gov.uk/ukpga/1988/52/contents

Serious Crime Act 2015, c. 9. https://www.legislation.gov.uk/ukpga/2015/9/contents

Thakoor v Rai (1865) 19 ER 941, 10 Moore Ind App 183.

Way v East (1853) 61 ER 635, 2 Drewry 44.

Woolmington v DPP [1935] AC 462, HL.

  1. * Professor of Criminal Jurisprudence, University of Nottingham, UK; Adjunct Professor of Law, CUPL, Beijing, PRC.

  2. 1 “Intersubjective” is the preferred term for sufferers of objectivity-phobia, but these two terms are typically synonymous, in my observation.

  3. 2 Generally, see Dahlman et al. (2021) and Roberts (2023).

  4. 3 Cf. Damaška (1997; 2018).

  5. 4 On transitional aspects of Enlightenment criminal procedure, see Roberts (2022a).

  6. 5 One engaging version of the tale is told by James Franklin (2001).

  7. 6 Cf. Brook et al. (2021) and Brodowski et al. (2024).

  8. 7 Also see, eg, Weisselberg (2017).

  9. 8 R v Majid [2009] EWCA Crim 2563, [1].

  10. 9 Woolmington v DPP [1935] AC 462, 481–2, HL.

  11. 10 In re Winship, 397 US 358 (1970).

  12. 11 Ferguson v R [1979] 1 WLR 94, PC (Lord Scarman).

  13. 12 Including Police and Criminal Evidence Act, 1984, s.76(2), regulating the admissibility of confessions. Also see, eg, Domestic Abuse Act, 2021, s.70; Serious Crime Act, 2015, s.72; Criminal Justice Act, 1988, ss.133 and 141; Road Traffic Act, 1988, s.5A(5).

  14. 13 Thus, in Ferguson v R [1979] 1 WLR 94, 99, Lord Scarman affirmed: “It is generally sufficient and safe to direct a jury that they must be satisfied beyond reasonable doubt so that they feel sure of the defendant’s guilt. Nevertheless, other words will suffice, so long as the message is clear. In the present case, the jury could have been under no illusion. The importance of being sure was repeatedly emphasised”.

  15. 14 ‘If a jury is told that it is their duty to regard the evidence and see that it satisfies them so that they can feel sure when they return a verdict of Guilty, that is much better than using the expression “reasonable doubt” and I hope in future that that will be done. I never use the expression when summing-up. I always tell a jury that, before they convict, they must feel sure and must be satisfied that the prosecution have established the guilt of the [accused]’: R v Summers (1952) 36 Cr App R 14, 15, CCA.

  16. 15 “It would be a great misfortune, in criminal cases especially, if the accuracy of a summing-up were made to depend upon whether or not the judge… had used a particular formula… It is not the particular formula of words that matters; it is the effect of the summing up”: R v Kritz [1950] 1 KB 82, 89, CCA (Lord Goddard CJ).

  17. 16 “It appears to me that there is such a body of evidence as is quite sufficient to convince a sound and temperate judgment beyond reasonable doubt that there did exist at the time of the execution of the deed of 1839, and during all the subsequent time, a design among the parties to the deed that the payment of the annuity should not commence till the death [of the testator]” (Way v East (1853) 61 ER 635, 643; 2 Drewry 44, 67)

  18. 17 Forbes v Meer Mahomed Tuquee (1870) 20 ER 614, 13 Moore Ind App 438; Thakoor v Rai (1865) 19 ER 941, 10 Moore Ind App 183; AG v Dean and Canons of Windsor (1860) 11 ER 472, 8 HLC 369.

  19. 18 See eg Enohin v Wylie (1862) ER 924, 934; 10 HLC 1, 24.

  20. 19 The Court of Criminal Appeal, the predecessor of today’s Court of Appeal (Criminal Division), was established by the Criminal Appeal Act 1907.

  21. 20 Note, however, that the cited authority for this assertion is Stephen’s own Digest of the Law of Evidence (1881)!

  22. 21 R v Gray (1900) 82 Law Times 534, 536 (16 June).

  23. 22 See eg R v Majid [2009] EWCA Crim 2563, [11]: “Judges are advised by the Judicial Studies Board, as they have been for many years, to direct the jury that before they can return a verdict of guilty, they must be sure that the defendant is guilty”

  24. 23 Crown Court Compendium (Judicial College, 2024 5.3), citing R v Desir [2022] EWCA Crim 1071.

  25. 24 R v Majid [2009] EWCA Crim 2563, [15].

  26. 25 Convictions are quashed on appeal when they are regarded as “unsafe” in the light of procedural errors or new information: Criminal Appeal Act 1968, s.2 (as amended).

  27. 26 There are practical constraints on researching real juries, which is presumptively forbidden, but not impossible: cf. Thomas (2013). Moreover, empirical researchers have employed a variety of stratagems and experimental proxies to build up a very substantial corpus of data on jury decision-making: Helm (2024); Ellison and Munro (2015); Darbyshire et al. (2001).

  28. 27 Not to mention wildly implausible claims for employing the standard of proof as a kind of institutional lever for engineering a desired scheme of trade-offs between convictions and acquittals. For a sense of the cascade of causal variables demanding consideration, and the complexity of their interactions, see Epps (2015). Some of this modelling is very sophisticated and theoretically adept: see eg Picinali (2022). But I remain unpersuaded by this procrustean methodology: Roberts (2024).

  29. 28 The canonical approach is now distilled in the Crown Court Compendium (Judicial College, 2024, 5.8): “[T]the jury should be directed as follows: (1) It is for the prosecution to prove that D is guilty. (2) To do this, the prosecution must make the jury sure that D is guilty. Nothing less will do. (3) It follows that defence does not have to prove that D is not guilty… [T]his is so even [when] D has given/called evidence”.

  30. 29 This is an important point of contrast between criminal trial procedure in England and Wales and judicial approaches in at least some US state jurisdictions, where judicial comment on the facts is widely avoided for fear of compromising the jury’s independent assessment of the evidence (Marcus, 2013).

  31. 30 Criminal Practice Directions 2023, Part 8.5.3: “A route to verdict, which poses a series of legal questions the jury must answer in order to arrive at a verdict, may be provided as part of the written directions. Each question should tailor the law to the issues and evidence in the case. The route to verdict may be presented (on paper or digitally) in the form of text, bullet points, a flowchart or other graphic”. See eg R v Ayre [2025] EWCA Crim 255, [51] (observing that “The route to verdict, carefully crafted by the judge, correctly took the jury through the questions which they had to consider”); R v Smith (Michael William) [2012] EWCA Crim 404, [32].

  32. 31 The Court of Appeal expressly refutes any such expectation: R v Adams (No.2) [1998] 1 Cr App R 377, CA.

  33. 32 One experimental red flag should have been the number of people who translate the meaning of “proof beyond reasonable doubt” as “100% certain” – a standard which is strictly impossible to achieve, and should logically result in 100% acquittals. Plainly, criminal juries are not using the same terminology or methods as epistemologists or mathematicians!

  34. 33 Including Keane and McKeown (2019).

  35. 34 Bench trials in common law jurisdictions provide a useful point of comparison: for a detailed worked example, see Roberts (2025).

  36. 35 To similar effect, see Tuzet (2020).

  37. 36 To similar effect, see Picinali (2015) and Kotsoglou (2020).

  38. 37 Juries are not required to convict merely on the basis of adequate epistemic warrant. A jury may extend its “equity” to defendants who, though technically guilty on the facts, do not, in the jury”s estimation, deserve criminal censure and punishment.

  39. 38 Cf. R v Yap Chuan Ching (1976) 63 Cr App R 7, CA.