Quaestio facti. Revista Internacional sobre Razonamiento Probatorio
Quaestio facti. International Journal on Evidential Legal Reasoning
Sección: Conjeturas y refutaciones
2024 l 6 pp. 201-235
Madrid, 2023
DOI: 10.33115/udg_bib/qf.i6.22888
Marcial Pons Ediciones Jurídicas y Sociales
© Federico Picinali
ISSN: 2604-6202
Recibido: 23/02/2023 | Aceptado: 22/06/2023 | Publicado online: 04/09/2023
Editado bajo licencia Reconocimiento 4.0 Internacional de Creative Commons
Evidential reasoning, testimonial injustice and the fairness of the criminal trial*
Federico Picinali
ABSTRACT: The article argues that the assessment of the relevance and of the probative value of an item of evidence is susceptible to an evaluation on moral grounds (such as fairness), rather than just to an evaluation on epistemic grounds (such as accuracy). In particular, the article shows that an assessment of relevance and of probative value is unfair, and renders the trial unfair, when this assessment instantiates epistemic injustice of the testimonial kind; and that it instantiates such an injustice when, due to identity prejudice against a social group to which one of the parties in the proceedings belongs, the evidence is assessed without considering the experience and stock of knowledge of this party. The article offers several examples of this phenomenon. The upshot is that higher courts, whose role includes checking that proceedings have been fair, should dirty their hands more readily than they are currently doing with the evidential reasoning of the first-instance adjudicator. However, the focus should be on preventing unfairness, rather than treating it.
KEYWORDS: evidential reasoning, testimonial injustice, fairness, criminal trial, relevance, probative value.
SUMMARY: 1. INTRODUCTION.—2. FUNDAMENTALS OF EVIDENTIAL REASONING.—3. TESTIMONIAL INJUSTICE IN EVIDENTIAL REASONING: 3.1. Testimonial Injustice as Prejudiced Dismissal of a Party’s Stock of Knowledge.—4. TESTIMONIAL INJUSTICE AS UNFAIRNESS.—5. CONCLUSION.—6. APPENDIX.—7. BIBLIOGRAPHY.
«Justice, being aligned with truth as final and singular, hears multiple voices struggle for its attention in the name of fairness, and then settles on one version of the “statement of facts” as the voice of the real»
Scheppele (1994, p. 1021-1022).
Traditional evidence scholarship assumes that any given assessment of the relevance and of the probative value of an item of evidence may be inaccurate, but it may not be unfair 1. Fairness, or the lack thereof, are not predicated of assessments of relevance and of probative value, which are unreflectively taken to be epistemic endeavours devoid of a moral dimension. Such a dimension is, instead, recognised in the overall question of the admissibility of the evidence, of which the assessments of relevance and—to an extent—of probative value are components. Therefore, fairness, or the lack thereof, are predicated of the choice to admit or to exclude a given item of evidence; or of the trial as a whole, depending on which of these options was chosen by the trial court.
This understanding of the nature of the assessments of relevance and of probative value is also apparent in the case law, and indeed it contributes to explaining why the English and Welsh Court of Appeal and, especially, the European Court of Human Rights (ECtHR) tend to defer to the evidential assessments made by lower courts 2. Because the assessments of the relevance and of the probative value of an item of evidence are seen as purely «factual» matters, they are also seen as falling outside the remit of institutions whose primary task is to elucidate and enforce legal norms, including the right to a fair trial provided by art. 6 of the European Convention on Human Rights (ECHR) 3. To be sure, the ECtHR has, on rare occasions, scrutinised and censured the factual conclusions of national courts, claiming that they had compromised the fairness of the trial 4. These isolated decisions, though, betray problematic views about evidential reasoning that appear to be widespread in the legal community. First, the view that, while evidential reasoning—which has, at its heart, assessments of relevance and of probative value—may render the trial unfair, it cannot itself be unfair. Second, the view that evidential reasoning may render the trial unfair only if the reasoning is inaccurate, that is, if its conclusion is not warranted by the epistemic material that is available to the reasoner. In other words, the view is that when evidential reasoning renders the trial unfair, unfairness is always mediated by the inaccuracy of the reasoning; unfairness may not, instead, be caused by, or reside in, other properties of the reasoning. Since, allegedly, it is only on rare occasions that the evidential reasoning of a national court is so inaccurate as to compromise trial fairness, it is only on rare occasions that the ECtHR delves into the national courts’ assessments of relevance and of probative value.
In this article, I question the received understanding of assessments of relevance and of probative value. While I accept the ECtHR’s view that these assessments may render the trial unfair, I also contend that the assessments themselves can be unfair, and that their impact on the fairness of the trial may be due to their intrinsic unfairness, rather than merely to their inaccuracy. In other words, my argument is that assessments of relevance and of probative value are susceptible to an evaluation on purely epistemic grounds (such as accuracy), but also to an evaluation on moral and moral/legal grounds (such as fairness) 5; and that, irrespective of their accuracy, they undermine the fairness of the trial when they are themselves unfair. While I am open to the possibility that there is more than one way in which an assessment of relevance and of probative value is unfair, my focus here is on unfairness due to the «testimonial injustice» that may occur in the assessment. As will be discussed shortly, relying on the seminal work of Miranda Fricker, testimonial injustice is a form of «epistemic injustice» that occurs «when prejudice causes a hearer to give a deflated level of credibility to a speaker’s word» (Fricker, 2007, p. 1). Jasmine Gonzales Rose (2021, p. 387-389) and Abenaa Owusu-Bempah (2022a, p. 148) have already drawn the connection between epistemic—in particular, testimonial—injustice and racist evidential practices 6. Developing their insight, my goal here is to provide a general—that is, applicable beyond the case of racism—theoretical framework for diagnosing testimonial injustice in assessments of relevance and probative value, and for understanding its legal significance. This unifying diagnostic lens serves as a bridge between two strands of literature: on the one hand, the fairly large philosophical scholarship on epistemic injustice; on the other, the growing legal scholarship on discriminatory and oppressive evidential practices in the criminal process.
Here is the plan of the work. In section 2, I clarify the notions of «relevance» and of «probative value» and I present the standard view about how assessments of these evidential attributes are to be carried out. In section 3, I clarify the notions of «epistemic injustice» and of «testimonial injustice», and I argue that assessments of relevance and of probative value instantiate testimonial injustice when, due to prejudice against a social group to which a party in the proceedings belongs, these assessments are carried out without giving appropriate consideration to generalisations reflecting the experience of this party qua member of that group. I offer some examples of such testimonial injustice drawn, in part, from the existing literature on discriminatory and oppressive evidential practices in the criminal process. My focus will be on the testimonial injustice suffered by defendants and by complainants, leaving aside other participants in the proceedings. In section 4, I argue that testimonial injustice is problematic from the very institutional perspective of the criminal trial. More precisely, I argue that a testimonially unjust assessment of relevance and of probative value is unfair—and renders the trial unfair—insofar as one understands fairness as depending on whether a party is given the opportunity to participate in the proceedings and, in particular, in the enterprise of fact finding. Notably, this understanding of fairness is in line with the ECtHR case law and the mainstream literature. Section 5 offers some concluding remarks, including a brief discussion of concrete measures to prevent testimonial injustice in evidential reasoning.
A terminological note. In the article I will often use the term «adjudicator» without distinguishing between magistrate, judge and jury. This is because the problem discussed here concerns both professional and lay fact finders; moreover, it concerns both the institution of the judge and that of the jury in a jury trial. Consider that, in a jury trial, both judge and jury assess the relevance and probative value of an item of evidence. While the judge is in charge of the question of admissibility and, therefore, makes a preliminary decision on the relevance of the evidence, the jury is free to find irrelevant an item of evidence that the judge has admitted. Also, while it is for the jury to make a final assessment of the probative value of the evidence, the judge may well have to make a fairly precise assessment of probative value when deciding whether there is a case to answer, but also when deciding on the admissibility of a particular item of evidence. The latter case occurs, for example, when the judge has to determine whether the item’s probative value «outweighs» its potential prejudicial effect (such that the evidence can be admitted); or, in case of incriminating hearsay evidence, when the judge has to determine whether there exist counterbalancing factors that, having considered the probative value of the evidence, would preserve a sufficient quantum of confrontation, should the evidence be admitted. Importantly, while for the most part I will follow this terminological choice, when discussing possible solutions to the problem of testimonial injustice I will clearly distinguish between measures that are targeted at the judge, deciding on admissibility, and measures that are targeted at the jury, deciding on evidential sufficiency.
I have written the article with the English and Welsh criminal justice system in mind, and the article contains references to English and Welsh institutions and legislation. This notwithstanding, I expect the theoretical component of the article to be applicable also elsewhere.
The starting point of my analysis is a critique of an aspect of what William Twining has called the «rationalist tradition of evidence scholarship». The rationalist tradition is a way of thinking about evidence and legal adjudication that has its roots in the empiricism of thinkers such as Francis Bacon, John Locke and John Stuart Mill and that has informed the modern and contemporary evidence law scholarship, as well as the common law model of adjudication. The tradition can be described in terms of the endorsement of a set of assumptions, including assumptions about the possibility of knowledge in the context of adjudication, about the nature of inferential reasoning, and about the goals of adjudication 7. Rather than providing a comprehensive account of the set, I shall draw attention to a particular assumption concerning the material that adjudicators should rely upon when assessing items of evidence and reaching factual conclusions.
Assessing an item of evidence involves determining its relevance and its probative value with respect to a given hypothesis. The locus classicus on the notion of relevance is found in the House of Lords’ decision in DPP v Kilbourne (1973, at 756), according to which «[e]vidence is relevant if it is logically probative or disprobative of some matter which requires proof […] [R]elevant (i. e. logically probative or disprobative) evidence is evidence which makes the matter which requires proof more or less probable». Probative value, instead, is the extent to which relevant evidence alters the probability of the «matter which requires proof». It follows that evidence cannot be relevant if it has no probative value and, hence, that an assessment of relevance is always also an assessment of probative value, even if it may stop at the recognition that the evidence has probative value, without specifying how much. These basic and, by and large, uncontested notions will suffice for the purposes of this article. Indeed, the focus here is not on elucidating the notions themselves, but on the process for assessing relevance and probative value.
In assessing the relevance and the probative value of an item of evidence, the adjudicator is likely to resort to one or more generalisations. A generalisation can be defined as a statement that, explicitly or implicitly, takes the conditional form «if A then B», where A and B represent distinct types of facts. Importantly, the connective «then» need not stand for a necessary implication. It stands for a numerical probability, or a probabilistic hedge, expressing the extent to which the occurrence of a fact of type A is indicative of the occurrence of a fact of type B 8. Consider, for example, the generalisation according to which an individual who flees from the scene of the crime as the police approaches is likely to have committed the crime. Flight from the scene of the crime as the police approaches corresponds to A in the conditional formula. Guilt corresponds to B. Finally, the connective «then» is spelled out in terms of the likelihood of guilt, given flight. This generalisation may come into play when the adjudicator assesses the relevance and the probative value of the defendant’s flight from the scene of the crime as the police was approaching. For this fact to be relevant evidence, it has to make guilt more or less probable than it would be if this fact had not been observed. So, relying on the above generalisation, the adjudicator estimates the probability of guilt given the defendant’s flight—or the posterior probability of guilt—and assesses whether it differs from the probability of guilt prior to considering the defendant’s flight—or the prior probability of guilt. If, and only if, the two probabilities differ, the evidence is relevant. The greater the difference between the two probabilities, the greater the probative value of the evidence. Obviously, if the posterior probability of guilt is greater than the prior probability of guilt, the evidence is incriminating; if it is smaller, the evidence is exculpatory 9.
But where do generalisations such as that just employed come from? The answer that the rationalist tradition gives to this question is found in one of the assumptions characterising this school of thought. «Judgments about probabilities»—including, therefore, assessments of the relevance and of the probative value of an item of evidence—«have, generally speaking, to be based on the available stock of knowledge about the common course of events…» (Twining, 2006, p. 76). This is to say that it is in «the available stock of knowledge»—i. e., the available set of known propositions—that the adjudicator should find the generalisations with which to assess an item of evidence. The use of the definite article before «stock of knowledge» may be read as betraying the second-order assumption that, in a given society, there is only one stock of knowledge or, at least, that there is a stock of knowledge that is to be privileged for the purposes of adjudication, due to some unspecified reason. It would not be an exaggeration to suggest that something akin to this second-order assumption indeed informs prominent works within the rationalist tradition 10. Needless to say, it is a problematic assumption to make 11.
Perhaps this will strike the reader as an uncharitable interpretation of the works in the tradition. After all, the very empirical philosophy from which the rationalist tradition allegedly derives, insists that knowledge is based on experience rather than on purely theoretical musings; and the fact that experience is not constant across a society and can vary dramatically from one individual to another, as well as from one social group to another, is certainly not lost on (it could not possibly be lost on) scholars in the tradition such as Twining and his co-authors 12. And yet, the rationalist tradition does not seem to have paid sufficient attention to this fact. In particular, it has failed to investigate the ethical implications that the variability of the stock of knowledge across a society can have for the enterprise of fact finding; hence, it has failed to appreciate that evidential reasoning can be unfair precisely because of such variability. This article is an attempt to address these failures.
While in the next section I offer further examples of the phenomenon, here is an initial illustration of the variability of the stock of knowledge across a society, and of the effect that it may have on evidential assessments 13. It is unlikely that in England and Wales a random White middle-class person experienced unwarranted abuse or violence at the hands of the police. Sadly, such an experience is fairly common among Black working-class youth 14. After all, that of England and Wales is by and large a racist criminal justice system, where the police force is disproportionately White and is itself institutionally racist 15. Now, a random White middle-class person asked to assess the relevance and probative value of a Black working-class youth’s flight from the crime scene as the police approaches is likely to rely on a generalisation akin to that considered earlier, such that they may well find this evidence relevant and fairly incriminating; or, at the very least, they may conclude that the suspect had something to hide from the police (perhaps, the involvement in some other illegal behaviour). After all, the assessor’s experience qua member of the group of White middle-class people suggests that there is little reason to fear injustice from the police: if someone flees from the police this is probably due to their having done something illegal. A Black working-class youth (perhaps any Black individual) asked to make the same assessment is unlikely to resort exclusively to that generalisation, since the generalisation does not give an exhaustive account of their experience qua member of the social group of Black people and, in particular, of Black working-class youngsters. They may also take into consideration the competing generalisation according to which, irrespective of their involvement in crime, a Black working-class youth is likely to run away from the police for fear of suffering an injustice. The assessor may, therefore, reach a different conclusion about the relevance and the probative value of flight 16.
The rationalist tradition’s apparent obliviousness to, or insufficient interest for, the variability of the stock of knowledge across a society may well have contributed to the correspondence between the stock of knowledge normally resorted to in court and the stock of knowledge dominating in the societies where the tradition has developed and thrived; societies such as the American and the English and Welsh, both shaped by conflictual phenomena such as racism, sexism, classism, and ableism. I am referring here to the stock of knowledge of White, able-bodied, middle- or upper-class, men. The claim that this has been the dominant stock of knowledge in criminal fact finding is not new, and has indeed been at the hearth of both feminist and critical race theory critiques of the law of evidence for the last thirty years or so 17. What interests me now is not to provide a further defence of this claim. Rather, I advance a distinct argument, which is also premised on the variability of the stock of knowledge across a society. Put succinctly, the argument is that depending on which stock of knowledge is considered in the assessment of relevance and probative value, a party in the proceedings may suffer testimonial injustice and, consequently, unfair treatment. My focus here is on the defendant and the complainant. I leave aside the illustration, and the assessment of the legal consequences, of the testimonial injustice that may be suffered by other participants, such as defence counsel, the prosecutor, and non-complainant witnesses. The following passage by Kim Scheppele (1994, p. 1011) encapsulates the background to the problem I address:
Each of us every day makes innumerable assessments of the real. Such assessments are generally unproblematic, either because our estimates do not disappoint us or because we are not confronted with the costs of our errors. Most of the time, each of us lives in the «real world» without having to explain or question exactly what it is that is real about it […] But after something has gone wrong, wrong enough for a lawsuit, the «real world» we constructed without systematic effort must be reconstructed systematically for a court. And then people encounter the problem of making the obvious features of their own social situation visible to others who may have quite different life experiences and quite different ways of imagining the real. For people with experiences that are likely to be comprehensible to judges and juries, such demonstrations are much easier than for people who are not similar to those who sit in judgment.
My concern here is with cases in which a party in the proceedings does not meet the challenge of «making visible» to the adjudicator their socially dependent experience and stock of knowledge and, hence, their interpretation of the world—in particular, of the evidence. The problematic scenarios that I deal with, though, are ones in which the responsibility for the challenge not being met does not lie with the party; rather, the challenge is not met due to the prejudice that the adjudicator holds against the party in virtue of the party’s membership in a given social group. This «identity prejudice» (Fricker, 2007, p. 27) is at the hearth of the notion of testimonial injustice, to which I now turn.
With this section, I introduce the notion of epistemic and, in particular, of testimonial injustice, I explain how this form of injustice may occur during the assessments of the relevance and of the probative value of an item of evidence, and I offer some examples of this phenomenon. In section 4, instead, I clarify the wrong involved in testimonial injustice, showing that, when the injustice occurs in said assessments, this wrong acquires institutional significance, since it also involves a violation of the right to a fair trial.
Epistemic injustice occurs when someone is «wronged in their capacity as an epistemic subject» (Fricker, 2017, p. 53) 18, that is, as a subject capable of epistemic tasks and attitudes that are central to human nature and essential to human existence. These include drawing inferences about past and future events, knowing, providing knowledge to others, and participating in interpersonal epistemic practices. There is a variety of forms of epistemic injustice 19. Epistemic injustice is «distributive» when it consists in an unfair distribution of epistemic goods and services—such as education, information, and access to expert advice—so that some individuals are unjustly hindered in their development and agency qua epistemic subjects 20. Epistemic injustice is «hermeneutical» when it consists in a member of a marginalised social group being unable to make visible and explain to others their social experience, since the collective hermeneutical resource is unfit for this communication, having been impoverished as a result of the identity prejudice held by the dominant social groups against the marginalised group 21. The phrase «identity prejudice» signifies a spurious judgment relating to the identity of a social group, that has some resistance to counter-evidence and has the character of systematicity, in that it tracks members of the relevant social group through different dimensions of their social lives 22. Identity prejudice may be positive or negative, but it is the negative kind that is at play in epistemic injustice: this is essentially a spurious association between members of a social group and one or more negative attributes (e. g. women are irrational and fickle) 23. Finally—but this is not meant as an exhaustive taxonomy 24—epistemic injustice is «testimonial» when a speaker suffers a credibility deficit due to the hearer holding an identity prejudice against a social group to which the speaker belongs 25. A credibility deficit, or deflation, consists in the speaker being afforded less credibility than is warranted by the available evidence that they are telling the truth. A familiar example of testimonial injustice is that of students who, due to sexist prejudice, give a credibility deficit to a female teacher (perhaps considering her less credible than her male counterparts).
Notice that the negative attribute involved in the identity prejudice at play in testimonial injustice need not be an epistemic attribute—e. g. the lack of trustworthiness. As we will see (consider the hypothetical Rap lyrics below), it may well be non-epistemic—e. g. the propensity to act violently. What matters for testimonial injustice is that there is a causal connection between the prejudice—whether or not it involves an epistemic attribute—and the deflated credibility judgment 26. Notice also that the lack of intentionality is an essential character of testimonial injustice 27. Indeed, in testimonial injustice the hearer misjudges the credibility of the testimony due to identity prejudice 28. If, instead, the hearer made a correct assessment of the testimony’s credibility, but intentionally treated (e. g. presented to others) the testimony as non-credible, the hearer would not have incurred a misjudgement. They would have deliberately harmed the speaker’s epistemic status. The intentional case too is a case of epistemic injustice, by any reasonable account of this notion. It is not, however, a case of testimonial injustice.
Here, I am concerned with testimonial injustice. As it will soon become evident, though, I adopt an extensive notion of this form of epistemic injustice. Whereas testimonial injustice normally involves testimony, understood as an utterance or a written statement that is meant to contribute to an interpersonal epistemic endeavour, I encompass within this notion also cases in which the «speaker» is not actually communicating anything. They do have a story to tell, but their telling of the story is pre-empted; the story is a priori dismissed by the «hearer» without it being heard 29.
Having clarified the notion of testimonial injustice, we are now in the position to understand how it may affect evidential reasoning. In the assessments of relevance and of probative value, testimonial injustice occurs when the stock of knowledge that a party in the proceedings has qua member of a social group is ignored or discounted due to the adjudicator’s identity prejudice against that group (or against another group to which the party belongs) 30 and, as a result, the party’s argument about the relevance and the probative value of an item of evidence—argument that relies on such stock of knowledge—receives a credibility deficit. To clarify, the plausibility of the party’s argument in question derives from the stock of knowledge and, hence, from the experience, of the party qua member of a social group, such that ignoring or discounting this stock and experience amounts to assigning a deflated credibility level to the argument. The credibility level is deflated because the stock of knowledge is a genuine reflection of the experience of a social group to which the party belongs and, reflecting the experience of a social group rather than the idiosyncratic experience of the party, it is cognisable and verifiable by the adjudicator with reasonable effort. Therefore, this socially determined stock of knowledge warrants assigning a higher credibility level to the party’s argument than the level assigned as a result of ignoring or discounting it 31.
In fact, it is irrelevant whether the party has actually advanced an argument about relevance and probative value, built on generalisations that are borne out of their experience. What matters is that these generalisations, and the argument they support, are a genuine reflection of the experience of the party qua member of a social group and that, due to identity prejudice against that group, they are not considered—or they are discounted—by the adjudicator when making the assessments of relevance and of probative value. Whether the communication of the party is actual or potential, then, the adjudicator’s prejudiced disregard for the distinctive stock of knowledge of the party (a stock of knowledge that the adjudicator should be aware of, insofar as it reflects the relevant experience of a social group) amounts to unjustly deflating the credibility of arguments about relevance and probative value that the party were to put forward based on such stock of knowledge. These arguments are dismissed, if put forward, or outright pre-empted. Either way, they are bound to have little to no basis from the perspective of an adjudicator who disregards the experience of the party.
There is an important caveat to make with respect to the conditions, set out in italics at the start of this section, for the occurrence of testimonial injustice in assessments of relevance and probative value. These are not necessary conditions, but sufficient conditions. Take the case of a participant in the proceedings (say, a non-complainant witness) who gives a testimony that is not based on their stock of knowledge qua member of a particular social group (say, a purely perceptual testimony to the effect that they saw an event take place). If this witness suffers a credibility deficit due to identity prejudice, testimonial injustice occurs; and it clearly is an instance of testimonial injustice that affects evidential reasoning (in particular, the probative value assigned to the words spoken by the witness). Nonetheless, in this article I focus on cases in which testimonial injustice is produced through ignoring or discounting the stock of knowledge of a party qua member of a particular social group, given that the party’s testimony is indeed based on such stock of knowledge. I focus on cases of this kind because, in a more palpable way than the case in the example just given, they evidence and reinforce the problematic second-order assumption highlighted in section 2 and, consequently, they constitute a more pernicious obstacle to achieving a fact finding process that is informed by the lived experience of a plurality of social groups, rather than by that of one social group only. Moreover, I already pointed out that in this article I focus on the testimonial injustice suffered by the defendant and the complainant, as opposed to other participants. This is because, as I will discuss later, the defendant and the complainant are owed opportunities to participate in the enterprise of fact finding; and undermining these opportunities, as testimonial injustice does, is relevant to whether evidential reasoning and the trial were fair under art. 6 ECHR 32. Since my aim is to show that evidential reasoning can be unfair due to testimonial injustice, I need not look beyond the cases of the defendant and of the complainant 33.
Below I offer three examples of the phenomenon of testimonial injustice in evidential reasoning. In the first two, the wronged party is the defendant; in the third, it is the complainant. In the Appendix, I offer a few more examples. Because all these examples present similar dynamics, three of them suffice for the purposes of the article. The examples in the Appendix, though, are useful to show that testimonial injustice may realistically occur in a wide range of cases. It is important to stress that, being examples of testimonial injustice, the cases below do not involve the deliberate disregard of the distinctive experience of a party. If the adjudicator recognised this experience as distinctive and useful for the purposes of fact finding, but deliberately disregarded or downplayed it, the result would not be a misjudgement of the credibility of the party’s testimony. Rather, the adjudicator would be intentionally dismissing, or pre-empting, the testimony of the party, thus deliberately harming the party’s epistemic status. I pointed out earlier that cases of deliberate harm fall outside the contours of testimonial injustice, but I will have more to say about them later.
Silence
The defendant, a Black youth, remained silent during police interview. At trial, they claim self-defence, thus making it permissible for the court to instruct the jury that they can use the defendant’s silence at interview, followed by their claim of self-defence, as incriminating evidence (see section 34 of the Criminal Justice and Public Order Act 1994) 34. Some jurors consider the behaviour of the defendant to be relevant and fairly incriminating evidence. They do so on the ground of the generalisations according to which an innocent defendant has nothing to hide and is, therefore, likely to put forward their defence at the earliest opportunity, especially considering that doing so may prompt a decision not to charge them. In other words, in the eyes of these jurors the defendant may well have fabricated their late defence. Because of racism in law enforcement and because of the underrepresentation of Blacks in the police, though, a Black young suspect may mistrust the police and see it as an oppressive force which, whether deliberately or by disposition, works on the basis of the racist narrative according to which Black youth are prone to misconduct and crime. Therefore, even if innocent, with a genuine defence, such suspect may have no desire to cooperate with the police and may use silence as an instrument of both defiance and self-preservation against—as Susie Hulley and Tara Young put it—«the racialised misinterpretation of talk» 35. Nevertheless, due to identity prejudice against Black people, in assessing the relevance and the probative value of the defendant’s silence, some jurors are not considering the generalisation, borne out of Black people’s experience, according to which, irrespective of their involvement in crime, a Black youth is likely to remain silent at police interview as a form of defiance and self-preservation. Notice that this generalisation supports an argument to the effect that the defendant’s silence is irrelevant: not taking into account the generalisation, means giving a deflated level of credibility to this argument. The jurors at issue are not deliberately disregarding, after having acknowledged them, the distinctive perspective of the defendant qua member of the Black minority and the resulting generalisation about silence. Rather, because of their racial prejudice, they are wilfully ignorant about lived experiences other than that of Whites. They are, therefore, oblivious to the phenomenon of racism in law enforcement or, at least, they are oblivious to its scale and impact on Black people, thus ignoring or discounting the possibility that, when it comes to interactions with the police, the perspective of a Black person may differ from that of a White person.
Rap lyrics
The defendant, a Black youth, is tried for a crime of street violence and the adjudicator considers it relevant and fairly incriminating that the defendant penned rap lyrics with a vivid violent content, including references to weapons, drugs and gangs. In drawing this conclusion, the adjudicator relies on the generalisation that someone who writes rap lyrics of this kind is likely to be in a gang and has a propensity to act violently towards other people, in particular, members of other gangs or crews. The adjudicator, though, ignores information about the stylistic rules and the topoi of the genre of rap. According to such rules and topoi, verbal violence, references to gangs and weapons, exaggeration, and bragging about committing or planning to commit violent acts are all commonplace, irrespective of an artist’s actual involvement in crime, and are often part of the artist’s attempt to construct a credible persona, with the aim of being more appealing in a music market that rewards the appearance of street authenticity 36. Moreover, far from being evidence of a propensity towards physical violence, the deployment of these topoi is often an outlet for frustration that may well serve as a substitute for, or a curb on, physical violence 37. Of course, there may be cases in which, say, due to the significant correspondence between the words of the lyrics and the crime charged 38, the temporal proximity between the two, and/or the fact that the lyrics mention details of the crime that are not in the public domain, it would be reasonable to consider the lyrics relevant and incriminating 39. However, in order to identify these cases, one would need to consider the perspective of the defendant qua member of the rap art community (precisely, the rules and topoi of the genre of rap) and to exclude that it provides an innocent explanation for the violent lyrics; that is, to exclude that it shows the lyrics to be irrelevant. The adjudicator in this hypothetical does not give appropriate consideration to this perspective because the generalisation linking violent lyrics to gang membership and physical violence resonates with their identity prejudice, which associates Black youth with street violence 40. This generalisation, therefore, chimes with a racist narrative of the case that the adjudicator independently endorses, thus providing an explanation of the evidence with which the adjudicator is intuitively comfortable and satisfied. An argument for irrelevance, based on the disregarded rules and topoi, would be given a deflated credibility level if it were put forward
Rape myth
In a trial for rape, the adjudicator concludes that the complainant’s claim that she was not consenting to the sexual act has very little probative value with respect to the issue of guilt. The adjudicator reaches this conclusion on the basis of a generalisation according to which, for rape to have occurred, the complainant must have put up some sort of resistance, such as crying for help or fighting back 41. Because apparently no such resistance was deployed by the complainant—in particular, no one in the vicinity heard any scream nor did the body of the complainant show any sign of struggle—the adjudicator concludes that sexual intercourse was probably consensual, that is, that the complainant’s words have little credibility 42. The generalisation in question is one of several, so called, «rape myths». These can be described as generalisations stating, or implying, that a female complainant in a trial for a sexual offence is not credible due to facts about the complainant and her behaviour (e. g. the lack of resistance) that are, by and large, irrelevant to her credibility 43. So understood, a rape myth is essentially an identity prejudice against women (female complainants, in particular), since it boils down to a spurious association between members of the group and a negative attribute—precisely, lack of credibility on certain matters 44. Notice that reliance on the rape myth about consent and resistance—indeed, on any rape myth, as just defined—is tantamount to ignoring the stock of knowledge of the complainant qua member of the social group of women (and, possibly, also qua member of that of women victims of sexual offences). This stock of knowledge indicates that sexual intercourse can be—and often is—non-consensual even in the absence of verbal or physical resistance. The complainant’s testimony is given a deflated credibility level due to such stock of knowledge being displaced by the rape myth. Notice that this is the same as giving a deflated credibility level to an argument, based on the disregarded stock of knowledge, to the effect that the complainant’s words («I did not consent!») are relevant and incriminating evidence.
With these examples in mind, and before moving to discuss the relationship between testimonial injustice and trial fairness, it is worth making some additional remarks on the nature of testimonial injustice, against the backdrop of the criminal trial. Notice that, for such injustice to occur, it does not matter whether the party that receives a deflated credibility judgement is or is not telling the truth. What matters is that, due to identity prejudice, their testimony is considered less credible than it should—or is pre-empted. A party who gives an inaccurate account—a party who lies even—is still wronged if the assessment of their credibility does not depend exclusively on indicators of accuracy, being also influenced negatively by identity prejudice against a social group to which the party belongs. Testimonial injustice occurs precisely when identity prejudice, which is not a reason for or against assigning a given level of credibility, makes it the case that the level of credibility assigned to the testimony by the adjudicator is lower than it should be given such reasons. In the examples given, the level of credibility is deflated because the experience of the party qua group member—which gives plausibility to the party’s argument on relevance and probative value—is ignored or discounted; and because, being this experience genuine and reasonably accessible to the adjudicator, it warrants assigning a higher credibility level to the party’s argument.
Notice also that for testimonial injustice to be avoided there is no need that the party’s testimony be believed or accepted by the adjudicator. Testimonial injustice realises «epistemic exclusivity», in that it unjustly restricts the epistemic material that is effectively relied upon for the purposes of assessments of relevance and probative value. Its avoidance requires «epistemic inclusivity», not belief or acceptance. Take Silence and Rap Lyrics. The point is not that in such cases the evidence at issue should always be considered irrelevant. Rather, the point is that if the evidence is found to be relevant and incriminating, this should be notwithstanding that the stock of knowledge of the defendant qua socially situated individual, and the arguments about relevance and probative value that can be built on such stock of knowledge, are properly taken into account. If accuracy is a goal, the stock of knowledge of the defendant should be taken into account, because it is useful epistemic material in the assessment of the relevance and of the probative value of the evidence. After all, the evidence in a criminal trial often consists in the defendant’s behaviour and in the traces left by their behaviour. The defendant, therefore, is in a privileged epistemic position for what concerns the interpretation of such evidence, and they may well be able to provide an innocent explanation for it on the basis of their socially determined experience (e. g. «being Black, I remained silent during police interview because I feared that the White interviewer would misinterpret my words»; or «as other rap artists, I wrote violent rap lyrics to express my frustration at social inequality and to appeal to the music market»). If the defendant’s stock of knowledge is not considered, then, this may affect the accuracy of the conclusions reached. If it is not considered due to identity prejudice, though, the risk of inaccuracy is not the only problem: testimonial injustice occurs 45. I have spoken in terms of a stock of knowledge «not being considered», «not being taken into account», «being dismissed», «being disregarded» or «being discounted». It should be clear that testimonial injustice occurs both when, due to identity prejudice, the adjudicator is unaware of a party’s distinctive stock of knowledge and when, due to identity prejudice, the adjudicator is aware of it, but does not give to this stock of knowledge appropriate consideration. Either way, the party’s testimony, being based on such stock of knowledge, receives a deflated credibility judgment.
An important question concerns which tools to deploy in order to avoid testimonial injustice and realise, instead, epistemic inclusivity in assessments of relevance and probative value. A fruitful avenue would be to develop an account of the epistemic virtue(s) that an adjudicator should exercise, perhaps along the lines of Miranda Fricker’s (2007, ch. 3-5) account of the epistemic (and ethical) virtue of «testimonial justice» 46. An additional challenge is that of identifying ways through which adjudicators can acquire and hone such virtue(s) 47. While this may sound unhelpfully vague, in the Conclusion I will identify concrete measures to prevent testimonial injustice in the particular institutional context of the criminal trial. The rich literature on discriminatory and oppressive evidential practices leaves little doubt about the fact that the problem that I am considering is real 48. Understanding the problem through the correct theoretical framework is the necessary starting point for detecting it in practice and for formulating any solution; and the task of this section was, precisely, to provide such an understanding.
When someone suffers testimonial injustice, they are wronged in their (typically human) capacity as an epistemic subject; in particular, they suffer the harm of not being recognised as capable of providing knowledge by contributing to a particular epistemic exchange. This harm is intrinsic to testimonial injustice, since it resides in the deflation of the credibility judgment that is characteristic of such injustice. But further, extrinsic, harms may also be produced. The wronged individual may suffer practical consequences as a result of having their contribution to the epistemic exchange unjustly ignored or discounted. For example, an innocent defendant may be convicted because their exculpatory account is dismissed due to identity prejudice; or they may receive a harsher sentence because mitigating factors that are part of this account are ignored or discounted, as a result of the account not being taken seriously. Also, in case of repeated testimonial injustice the wronged individual may lose confidence in their intellectual abilities, with the ramifications that this may have for their intellectual development 49. What makes these harms wrongs—that is, what makes the phenomenon described a form of injustice—is their being caused by the identity prejudice of the hearer. If the speaker suffered these harms as a result of bad luck or of their own fault—say, because they happened to instantiate indicators of inaccuracy which do not, in fact, track the truth of their particular testimony 50—they would not be wronged by the hearer and, hence, they would not suffer testimonial injustice 51.
When evaluated against the legal framework of the criminal trial, the testimonial injustice occurring in assessments of relevance and probative value involves a distinctly legal wrong: it infringes the right to a fair trial. For my purposes it is unnecessary to determine whether this is an additional wrong with respect to those just described, occurring in the specific context of the trial, or is, instead, the intrinsic wrong of testimonial injustice that, in such a context, acquires legal relevance and name. What matters is that a legal wrong is involved: testimonial injustice is problematic from the very institutional perspective of the criminal trial. To show that testimonial injustice is a form of unfairness, though, I need to clarify what trial fairness consists of; that is, what is the content of the defendant’s right to a fair trial.
John Jackson and Sarah Summers (2012, p. 195-196) convincingly reconstruct the ECtHR’s interpretation of trial fairness as «demand[ing] that those accused of criminal offences have the opportunity to challenge the evidence levelled against them by the prosecution and to present their own evidence in adversarial proceedings» 52. In the ECtHR’s understanding, then, the defendant’s right to a fair trial centres around participation, which involves being in the condition to follow the development of the proceedings, but also having the opportunity to contribute, in particular, to the epistemic enterprise of fact finding 53. Indeed, a typical battleground for the question of fairness is represented by cases where the national court admitted evidence obtained improperly; and in such cases a factor to which the ECtHR gives prominence in deciding on the overall fairness of the trial is precisely whether the defendant was given the opportunity to challenge the admissibility of the evidence. 54
The ECtHR is not alone in recognising that participation is at the heart of the notion of trial fairness. Antony Duff (2018, p. 304) writes that «[t]he rights that are taken to bear on the fairness of a trial include “participatory”, as well as “protective”, rights: accused persons have a right to be effectively heard, and this seems crucial to the fairness of the proceedings» 55. In a similar vein—but with an important twist to which I will soon return—Stefan Trechsel (2018, p. 33) writes that, «in the context of “fair proceedings” [the term “fair”] means equitable, balanced, giving each side an equal chance to present its point of view». That participation is considered central to fairness should come as no surprise. After all, it is generally accepted that the right to a fair trial, enunciated in the title and in the first paragraph of art. 6 ECHR, is a «superordinate concept» (Trechsel, 2018, p. 22) which includes the specific guarantees provided by the third paragraph of the article. These guarantees are all about giving the defendant the opportunity for effective participation in the trial.
There is another crucial, and rarely spelt out, consideration to make with respect to the content of the defendant’s right to a fair trial. The gist is that the trial cannot be fair—and, hence, the defendant’s right cannot be respected—if the defendant is the only party in the proceedings who is given the opportunity for effective participation. This is due to the pluralistic nature of the concept of «trial fairness». Let me elaborate. As suggested by Trechsel’s words, a fair trial is a trial in which suitable treatment is given to all parties, as opposed to one party only. Because a trial is a contest between parties, for the trial to be fair all parties involved must be treated fairly, equitably. Now, equitable treatment is by no means identical treatment. Rather, it is treatment that is commensurate to the legitimate interests that each party has at stake in the contest, interests that are clearly not constant across the parties (e. g., the defendant, but not the complainant, risks their freedom and property). To treat a party equitably, then, is to give this party what is owed to them, given their legitimate interests. It follows that for a contest between parties to be fair, all parties must receive the treatment they are owed, given their legitimate interests 56. As soon as a party gets more, or less, than they are owed, unfairness ensues. This is because the correspondence between treatment and a party’s legitimate interests works for some party only, if at all. Some party will be unjustly advantaged or disadvantaged, such that the contest cannot be called fair.
These considerations support Trechsel’s (2018, p. 33) emphatic conclusion that «[f]airness, as a matter of course, concerns all participants in the game […] it would be grotesque, absurd, completely illogical to claim fairness for one participant alone» 57. To recapitulate, then, fairness depends on the treatment of all parties and on the interests that each party has at stake: it requires that each party is given what they are owed given their legitimate interests. Moreover, the treatment that is crucial to fairness consists in giving the opportunity for meaningful participation in the trial and, in particular, in the epistemic enterprise of fact finding. This is true for the defendant, but also for the complainant. After all, it is through participation that both parties can voice, operationalise, and attempt to secure, their legitimate interests. Importantly, while both the defendant and the complainant must be afforded the opportunity for meaningful participation, fairness does not demand that they be afforded identical opportunities. This is because the legitimate interests with which fairness is concerned are not constant across the parties.
Having articulated a pluralistic conception of fairness, the next step is to show that if the defendant or the complainant suffer testimonial injustice due to the adjudicator’s evidential reasoning, the trial is unfair. But there is an important clarification to make first. I am aware that the ECtHR has not fully and wholeheartedly extended to the complainant the right to a fair trial. In fact, the Court has formally recognised a right of the complainant to a fair trial only in the case in which the complainant has brought a civil action within the criminal proceedings, as is allowed in some civil law systems 58. The ECtHR, though, has maintained that the fairness of the proceedings depends on «balancing» the interests of the defendant with those of the complainant (indeed, the witnesses and also the public) 59, apparently embracing a pluralistic conception of fairness akin to that just sketched 60. And the interests of the complainant that are at issue for the Court no doubt call for giving this party the opportunity to be heard; that is, to participate in the fact-finding enterprise through presenting their side of the story 61. Now, for the purposes of my argument it matters not whether the complainant, like the defendant, has a right to a fair trial—that is, an actionable prerogative to a fair contest—or mere protected interests that are relevant to trial fairness—that is, prerogatives that are not actionable by the complainant, but that nonetheless should be respected, through providing appropriate treatment to the complainant, for the trial to be fair 62. Either way, under a pluralistic conception of fairness a trial cannot be fair if the complainant is not given the treatment that they are owed, given the legitimate interests they have at stake. Indeed, in these circumstances, whoever has a right to a fair trial will suffer a breach of said right: they will be denied that to which they have a right. So, even if the complainant had no right to a fair trial, the fact that they are not given the treatment they are owed—in particular, the opportunity for meaningful participation—means that the trial was unfair, and that the defendant suffered a breach of their right to trial fairness. While possibly counterintuitive for some readers, this strikes me as the inevitable implication of a pluralistic conception of fairness 63.
As seen in the previous section, testimonial injustice occurs in assessments of relevance and of probative value when the stock of knowledge of a party—be it the defendant in Rap lyrics or the complainant in Rape myth—is not given appropriate consideration due to identity prejudice, such that the testimony of this party concerning the evidence at issue is discounted or pre-empted. The party is, therefore, hampered in their ability to participate in the proceedings—and, in particular, in the enterprise of fact finding—through having their word and their experience unjustly discounted, if not altogether dismissed as unworthy of consideration or credit. The injustice resides, of course, in the intrinsic harm that is suffered by the party (they are given a deflated credibility level), coupled with the identity prejudice that causes the harm. In the institutional context of the criminal trial, though, this injustice acquires the attribute of unfairness, since the defendant and the complainant suffering testimonial injustice are denied treatment that is surely owed to them, given the interests they have at stake: precisely, the opportunity for a meaningful—if more modest, in case of the complainant—participation in the enterprise of fact finding. A trial where testimonial injustice occurs is an unfair trial 64.
I pointed out earlier that testimonial injustice involves a misjudgement of the credibility of a testimony, not the intentional dismissal of a testimony whose credibility has been correctly assessed. I have also pointed out that cases of intentional dismissal are cases of epistemic injustice nonetheless. I add here that if the adjudicator intentionally dismisses the account of a party in the context of assessments of relevance and probative value, they are not treating this party fairly since they are intentionally denying this party the role of a participant in the proceedings. Once the notion of fairness is elucidated, this conclusion is straightforward. One may ask why I have focused specifically on cases of testimonial injustice, sidelining cases of deliberate injustice. Miranda Fricker theorised testimonial injustice because she felt that episodes of unjust misjudgement of credibility were pervasive, but their moral significance was not adequately understood, such that they would often go unnoticed by wrongdoers, observers and even victims 65. Similarly, while cases of intentional dismissal of a party’s testimony are probably rare and straightforwardly wrong, cases of misjudgement of the credibility of a party’s testimony due to identity prejudice are probably more frequent, and their injustice and relationship to fairness may go unnoticed or misunderstood without a theoretical diagnosis of the phenomenon. While, in this article, I have not offered any data as to the frequency of such misjudgements, I have explained how they can occur in assessments of relevance and probative value and why, when they do occur, they generate unfairness.
In assessing the relevance and the probative value of an item of evidence, an adjudicator who disregards or discounts the stock of knowledge of a party qua member of a social group, due to identity prejudice against that group, commits testimonial injustice. Indeed, the behaviour of the adjudicator results in the fact that the party’s argument about the evidence, based on such stock of knowledge, is pre-empted—that is, a priori dismissed—or in any case receives a deflated credibility judgement—that is, is found less credible than it should be found, given the available indicators of accuracy. These are harms to the epistemic status of a party—that is, harms that the party suffers in her capacity as a knower and a provider of knowledge in the context of the trial. They are unjust harms, since they result from identity prejudice.
If the assessment of relevance and of probative value can unjustly harm a party’s epistemic status—in particular, if it can unjustly hinder the party’s participation in the enterprise of fact finding—then such an assessment can be unfair. Not only are assessments of relevance and of probative value susceptible to being evaluated on moral—and, indeed, moral/legal—grounds rather than purely on grounds of accuracy, but also their impact on trial fairness is not always mediated by their inaccuracy, contrary to what courts seem to believe. In other words, an assessment of relevance and of probative value can generate unfairness—in fact, can itself be unfair—quite irrespective of whether it is inaccurate. Let me clarify. Of course, if testimonial injustice occurs, the level of credibility assigned to the testimony is by definition inaccurate, in that it is lower than is warranted by the evidence. However, this need not mean that the unfairness of an assessment of relevance and of probative value, and the resulting unfairness of the trial, are always mediated by the assessment’s inaccuracy. Consider that assigning a level of credibility to an argument concerning relevance and probative value may be just a part of the overall assessment of the relevance and probative value of the evidence. The overall assessment may, indeed, involve several such arguments, put forward by different parties or independently elaborated by the adjudicator. It is possible that, notwithstanding that testimonial injustice—and, hence, inaccuracy—occurred in assigning a level of credibility to a party’s argument, the overall assessment of the relevance of the item of evidence is serendipitously accurate: its conclusion is warranted by the epistemic material available to the assessor. And yet, the assessment, and the trial, would be unfair due to the testimonial injustice: this injustice is alone sufficient to generate unfairness.
The conclusion that assessments of relevance and of probative value can be unfair, and can render the trial unfair, naturally calls for a more interventionist role of higher courts in the scrutiny of the adjudicator’s evidential reasoning. In principle, insofar as the Court of Appeal and the ECtHR are judges of the fairness of the proceedings, they should more readily concern themselves with evidential matters. Of course, the practical obstacles of an increased evidential scrutiny by these courts should not be downplayed. These obstacles relate, in particular, to the difficulty of scrutinising evidential reasoning and conclusions that, in a system such as the English and Welsh, are rarely articulated in writing. Can trial transcripts, judicial summing up and jury instructions—perhaps coupled with empirical evidence on the commonness of a certain identity prejudice across different demographics—offer a sufficient basis to determine whether the judge or the jury have committed testimonial injustice? 66 A study of this and related practical questions lies beyond the scope of this article.
These practical obstacles, though, may not be as consequential as it seems at first: once a problem is identified and understood, prevention can play a more incisive role than treatment; and preventing the unfairness studied in this article need not require that higher courts play a large part. I mentioned earlier that developing—and imparting to adjudicators, be they lay or professional—a virtue-ethical and virtue-epistemological account of adjudication is a fruitful preventative avenue to pursue. 67 But there are also readier preventative measures to consider. Because testimonial injustice in evidential reasoning may be committed both by professional and by lay adjudicators, one must identify measures that are tailored to each type of adjudicator, being mindful of the respective institutional roles. Here I restrict my attention to the institutional context of the jury trial, and I start with possible measures targeted at the jury.
A radical instrument for preventing testimonial injustice by the jury is the enactment of ad hoc statutory rules that exclude, or restrict the admissibility of, the evidence in question, such that the jury will not hear the evidence or will only hear it when the risk that they will commit testimonial injustice in assessing it is sufficiently contained. In fact, it is arguable that the much-debated section 41 of the Youth Justice and Criminal Evidence Act 1999 68, which restricts the admissibility of sexual history evidence, is not just an instrument to safeguard the accuracy of the verdict and to prevent secondary victimisation due to stress and loss of privacy; it is also an instrument to prevent secondary victimisation due to testimonial injustice committed by the jury in the assessment of the evidence. While the introduction of exclusionary rules has been defended in the case of rap lyrics, both here and overseas 69, given the loss of sound epistemic material that such rules can produce, they should be treated as a last resort. A less radical way of preventing testimonial injustice by the jury is to rely on judicial discretion to exclude the evidence in relation to which jurors might commit such injustice. In deciding on the admissibility of an item of evidence, the judge should consider the risk that testimonial injustice—therefore, unfairness—will occur in the assessment of this evidence, if admitted. Take Rap lyrics. Insofar as the judge is the guardian of the fairness of the proceedings, they should exclude the lyrics if unfairness is sufficiently likely to originate from how these will be assessed by the jury. Indeed, section 78(1) of the Police and Criminal Evidence Act 1984 may be interpreted so as to require the judge to exclude evidence in the presence of a substantial risk that the jury will commit testimonial injustice—and, hence, cause unfairness—in its assessment. No doubt, this would be a difficult and contentious evaluation for the judge to make, especially when they consider that the evidence has significant probative value; but an evaluation not altogether different from some of the evaluations that are currently part of the question of admissibility. It is also worth considering and testing the possibility that judicial instructions play some debiasing role for what concerns jurors’ reasoning. Should this be the case, the strategy of admitting the evidence while issuing debiasing instructions would have greater appeal than the options just considered: it would contain the risk of testimonial injustice without excluding useful epistemic material. The Crown Court Compendium includes sample instructions on rape myths that are premised on the idea that instructions have debiasing capability 70. Perhaps this is wishful thinking. If it isn’t, further instructions of this kind would have to be devised and promoted.
Let’s now consider measures to prevent professional judges from committing testimonial injustice in their evidential reasoning, with particular regard to their decisions on admissibility. A promising, and overdue, one is to increase the diversity of the judiciary: a more diverse judiciary would probably limit the impact that identity prejudice may have on the judicial assessment of the evidence 71. Yet another preventative measure targeted at judges could be to encourage or require them to offer, either orally in court or in writing, a sufficiently detailed articulation of the reasoning underlying their decisions on admissibility. This measure should be especially steadfast for evidence of such a kind that the risk of unfairness in evidential reasoning is marked. For decisions on the admissibility of bad character evidence and of sexual history evidence, the Criminal Procedure Rules 2020 include a requirement that the court announces their reasons at a public hearing 72. But a more specific incentive or requirement, perhaps with broader application, should be considered. From his 2017 review on racial disparities in the criminal justice system, David Lammy MP (2017, p. 6) drew «the key lesson […] that bringing decision-making out into the open and exposing it to scrutiny is the best way of delivering fair treatment». Whether or not this is indeed the best route to fairness, Lammy seems right in saying that having to justify a decision to others «both deters and exposes prejudice or unintended bias» (p. 6). There is a last measure worth considering, targeted both at professional and at lay adjudicators. This is to demand that adjudicators regularly take an Implicit Association Test 73, in order for them to learn (or be reminded of) which implicit biases they harbour 74. I assume here that they will harbour some—we are all likely to—and the point of this exercise would not be to chastise them for this. Rather, research has shown that awareness of one’s implicit biases increases one’s ability to disengage them in cognitive tasks carried out relatively soon after the awareness is acquired 75.
The focus on prevention brings me back to the epigraph. There is a sense in which criminal justice must settle on «one voice». If by «voices» we mean the verdicts that the adjudicator may hand down, it is true that the adjudicator has to choose a single voice as the voice of justice. If, however, by «voices» we mean the experiences, stocks of knowledge, perspectives of the parties, then the epigraph is mistaken. At least, it is mistaken as a statement about what criminal justice can, and should, achieve—as opposed to what it currently achieves. Philosophical work on epistemic injustice has shown that epistemic oppression—that is, any unjust exclusion that prevents individuals or groups from participating in interpersonal epistemic enterprises—is pervasive and perhaps impossible to eradicate completely 76. We harbour prejudices. As a result, in the cognition that is involved in decision problems and enterprises with a plurality of participants, we unwittingly exclude the perspective of others, even when we value and pursue inclusion. But include we can and include we do, if often only partially. An epistemic enterprise such as the criminal trial is a context where inclusion is relatively easy to achieve. The parties are limited in number and, hence, the different experiences that the adjudicator should cognise and consider in their assessment of relevance and probative value are limited as well. The challenge is to make these experiences visible to, and salient for, the adjudicator; to disengage the non-deliberate prejudiced process that may lead the adjudicator to ignore or discount some of them. Once this process is disengaged, and assessments of relevance and probative value involve reasoning that is epistemically inclusive, justice can indeed speak a plurality of voices 77.
Asylum seeker
The defendant is an asylum seeker. Two years after submitting their asylum application, they are still waiting for the decision of the competent authorities and they are, therefore, forbidden from working 78. They are on trial for supplying a controlled drug and for possession of a controlled drug with intent to supply. The prosecution’s case rests on the testimony of an eyewitness, on drugs (in a quantity allegedly inconsistent with personal use) found in the accommodation that the defendant shares with fellow asylum seekers, and on a substantial amount of cash found under the defendant’s mattress. With respect to the cash, the prosecution argues that it is indicative of drug dealing, and they support their argument with the generalisation that such an amount of cash is more likely to be found in the possession of a drug dealer than of an innocent member of the public 79. The adjudicator agrees with this argument, therefore finding the cash relevant and incriminating. They reach this conclusion without considering—or without taking seriously—the possibility that the cash is the remuneration that the defendant received for illegal work carried out while waiting for a decision on their asylum application and that, therefore, the cash is irrelevant to the hypothesis of guilt. The practice of working illegally is common amongst asylum seekers, given the prohibition to work and the very small allowance they receive from the state 80. The adjudicator does not consider appropriately the possibility of illegal work, because they ignore or discount the perspective of the defendant qua member of the group of asylum seekers, due to identity prejudice against this group. To clarify, whether consciously or unconsciously, the adjudicator associates asylum seekers with a propensity for criminality and dishonesty, such that the narrative of drug dealing represents for them an intuitively fitting explanation for the evidence, and any alternative account is indiscernible or unworthy of careful consideration. An argument for irrelevance, based on the disregarded experience of asylum seekers, would be given a deflated credibility level, if it were put forward.
Homeless
The defendant is a homeless and unemployed person who spends their day and night on the high street. During the Christmas period, they are found in possession of packaged toiletries, first-aid products, and confectionery for a total value of more than thirty pounds. The prosecution argues that the defendant stole these products from a local corner shop, treating the defendant’s possession of them as relevant and incriminating evidence of theft. The only other prosecution evidence is the testimony of the shop owner, who is known to have an uneasy relationship with homeless people. The prosecution ground their argument about the relevance and probative value of possession on the generalisation that someone who cannot afford relatively expensive products, and is found in possession of them, is likely to have acquired them illegally. The adjudicator finds the argument persuasive. In fact, the stated generalisation does not take into account the possibility that the products have been donated. Notably, in the Christmas period charities collect parcels (often shoeboxes) from donors, containing essential and other goods, and distribute them to people in need, including the homeless. It is, therefore, quite common for a homeless person around the country to receive (and to count on receiving) one such parcels. The adjudicator does not consider—or give due weight to—this possibility because, due to identity prejudice against the homeless, they ignore or discount the experience of the defendant qua member of the group of homeless people. To clarify, whether consciously or unconsciously, the adjudicator associates the homeless with a propensity for laziness, free-riding and dishonesty, such that the narrative of theft represents for them an intuitively fitting explanation for the evidence, and any alternative account is indiscernible or unworthy of careful consideration. An argument for the irrelevance of possession (indeed, for innocence), based on the disregarded experience of the homeless, would be given a deflated credibility level, if it were put forward.
Learning Disability
The complainant in a trial for sexual assault has a learning disability affecting speech. As a result, during their testimony their speech is broken, they take long pauses before starting to answer a question, and they use vague words when asked to recount specific details of the alleged crime 81. The adjudicator finds the testimony not credible, relying on a generalisation to the effect that a testimony delivered in the manner described is likely to be a lie or, at the very least, to be an account based on imprecise and patchy recollection. They reach this conclusion disregarding, or not giving due weight to the fact that an individual with a learning disability affecting speech may speak as the complainant did, without thereby being any less credible than someone who is not disabled. The adjudicator ignores, or discounts, the perspective of the complainant qua member of the group of people with learning disabilities, due to identity prejudice against those affected by mental disability 82. In particular, whether consciously or unconsciously, the adjudicator holds the ableist prejudice that associates mental disability with untrustworthiness, as well as with a marked incapacity to accurately encode and assess a situation that the mentally disabled person finds themselves in. Given this prejudice, the narrative according to which the complainant is misremembering or outright lying represents for the adjudicator an intuitively fitting explanation for the manner of speech displayed in the testimony. The complainant’s testimony is thus given a deflated credibility level, due to the disregarding of their perspective qua person with a learning disability. Notice that this is the same as giving a deflated credibility level to an argument, based on the disregarded experience, to the effect that the complainant’s words are relevant and incriminating.
Akala (2018). Natives: Race and Class in the Ruins of Empire. Two Roads.
Anderson, T., Schum, D. and Twining, W. (2005). Analysis of Evidence (2nd ed.). Cambridge University Press.
Arcila-Valenzuela, M. and Páez, A. (2022). Testimonial Injustice: The Facts of the Matter. Review of Philosophy and Psychology. https://doi.org/10.1007/s13164-022-00662-w
Campbell, L., Ashworth, A. and Redmayne, M. (2019). The Criminal Process. Oxford University Press.
Casey, L. Baroness (2023). Final Report. An Independent Review Into the Standards of Behaviour and Internal Culture of the Metropolitan Police Service. Retrieved August 8, 2023, from at https://www.met.police.uk/SysSiteAssets/media/downloads/met/about-us/baroness-casey-review/update-march-2023/baroness-casey-review-march-2023a.pdf.
Coady, D. (2017). Epistemic Injustice as Distributive Injustice. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Congdon, M. (2017). What’s Wrong with Epistemic Injustice? Harm, Vice, Objectification, Misrecognition. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Chalmers, J., Leverick, F. and Munro, V. (2021). Why the Jury Is, and Should Still Be, Out on Rape Deliberation. Criminal Law Review, 9, 753-771.
Cohen, L. J. (1977). The Probable and the Provable. Oxford University Press.
Cohen, L. J. (1983). Freedom of Proof. In W. Twining (Ed.), Facts in Law. Franz Steiner Verlag GmbH.
Crown Prosecution Service (2021). Drug Offences, Legal Guidance. Retrieved June 15, 2023, from https://www.cps.gov.uk/legal-guidance/drug-offences#:~:text=Possession%20with%20intent%20to%20supply,-Where%20the%20evidence&text=As%20with%20a%20simple%20possession,intent%20of%20the%20actual%20drug–
Daly, E., Smith, O., Bows, H., Brown, J., Chalmers, J., Cowan, S., Horvath, M., Leverick, F., Lovett, J., Munro, V. and Willmott, D. (2023). Myths About Myths? A Commentary on Thomas (2020) and the Question of Jury Rape Myth Acceptance. Journal of Gender-Based Violence, 7(1), 189-200.
Dennis, I. (2020). The Law of Evidence (7th ed.). Thomson Reuters.
Doak, J. (2008). Victims’ Rights, Human Rights and Criminal Justice: Reconceiving the Role of Third Parties. Hart Publishing.
Dotson, K. (2012). A Cautionary Tale: On Limiting Epistemic Oppression. Frontiers: A Journal of Women Studies, 33(1), 24-47.
Duff, R. A. (2018). Fairness in Criminal Proceedings: Concluding Thoughts and Further Questions. In J. Jackson and S. Summers (Eds.), Obstacles to Fairness in Criminal Proceedings. Individual Rights and Institutional Reforms. Hart Publishing.
Fairclough, S. (2018). Speaking up for Injustice: Reconsidering the Provision of Special Measures through the Lens of Equality. Criminal Law Review, 1, 4-19.
Fricker, M. (2007). Epistemic Injustice. Power and the Ethics of Knowing. Oxford University Press.
Fricker, M. (2017). Evolving Concepts of Epistemic Injustice. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Gerger, H., Kley, H. and Bohner, G. (2007). The Acceptance of Modern Myths about Sexual Aggression Scale: Development and Validation in German and English. Aggressive Behaviour, 33(5), 422-440.
Gonzales Rose, J. B. (2017). Toward a Critical Race Theory of Evidence. Minnesota Law Review, 101, 2243-2312.
Gonzales Rose, J. (2021). Race, Evidence, and Epistemic Injustice. In C. Dahlman, A. Stein and G. Tuzet (Eds.), Philosophical Foundations of Evidence Law. Oxford University Press.
Goss, R. (2016). Criminal Fair Trial Rights. Article 6 of the European Convention on Human Rights. Hart Publishing.
Griggs, M. (2021). Race, Rules, and Disregarded Reality. Ohio State Law Journal, 82(6), 931-951.
Home Office (2022). Permission to Work and Volunteering for Asylum Seekers. Guidance. Retrieved June 15, 2023, from https://www.gov.uk/government/publications/handling-applications-for-permission-to-take-employment-instruction/permission-to-work-and-volunteering-for-asylum-seekers-accessible-version
Hoyano, L. (2014). What is Balanced on the Scales of Justice? In Search of the Essence of the Right to a Fair Trial. Criminal Law Review, 1, 4-29.
Hoyano, L. (2015). Reforming the Adversarial Trial for Vulnerable Witnesses and Defendants. Criminal Law Review, 2, 107-129.
Hulley, S. and Young, T. (2022). Silence, Joint Enterprise and the Legal Trap. Criminology and Criminal Justice, 22(5), 714-732.
Hunter, R. C. (1996). Gender in Evidence: Masculine Norms vs Feminist Reforms. Harvard Women’s Law Journal, 19(1), 127-168.
Jackson, J. D. and Summers, S. J. (2012). The Internationalisation of Criminal Evidence. Beyond the Common Law and Civil Law Traditions. Cambridge University Press.
Jackson, J. and Summers, S. (2018). Introduction. In J. Jackson and S. Summers (Eds.), Obstacles to Fairness in Criminal Proceedings. Individual Rights and Institutional Reforms. Hart Publishing.
Jalloh, T. O. (2022). Does the Critical Scrutiny of Drill Constitute an Epistemic Injustice? British Journal of Aesthetics, 62(4), 633-651.
Johnson, A. (2022). Racial Discrimination—A Stocktake. Criminal Law Review, 2, 83-87.
Judicial College (2020). Crown Court Compendium. Part I: Jury and Trial Management and Summing Up. Retrieved June 15, 2023, from https://www.judiciary.uk/wp-content/uploads/2020/12/Crown-Court-Compendium-Part-I-December-2020-amended-01.02.21.pdf
Judicial College (2021). Equal Treatment Bench Book. Retrieved June 15, 2023, from https://www.judiciary.uk/wp-content/uploads/2023/06/Equal-Treatment-Bench-Book-April-2023-revision.pdf
Kaye, D. H. (1986). Quantifying Probative Value. Boston University Law Review, 66, 761-766.
Kinports, K. (1991). Evidence Engendered. University of Illinois Law Review, 413-456.
Lacey, N. (2022). Criminal Justice and Social (In)Justice. Retrieved June 15, 2023, from https://eprints.lse.ac.uk/116949/1/Lacey_criminal_justice.pdf
Lackey, J. (2020). False Confessions and Testimonial Injustice. Journal of Criminal Law and Criminology, 110, 43-68.
Lackey, J. (2023). Criminal Testimonial Injustice. Oxford University Press.
Lammy, D. MP (2017). The Lammy Review: Final Report. An Independent Review Into the Treatment of, and Outcomes for, Black, Asian and Minority Ethnic Individuals in the Criminal Justice System. Retrieved June 15, 2023, from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/643001/lammy-review-final-report.pdf
Law Commission (2023). Evidence in Sexual Offences Prosecutions. Consultation Paper 259, Retrieved July 27, 2023, from https://www.lawcom.gov.uk/project/evidence-in-sexual-offence-prosecutions/#:~:text=The%20consultation%20paper,-In%20our%20consultation&text=The%20proposed%20reforms%20have%20three,defendants%20receive%20a%20fair%20trial.
Lempert, R. O. (1977). Modeling Relevance. Michigan Law Review, 75(5/6), 1021-1057.
Leverick, F. (2020). What Do We Know about Rape Myths and Juror Decision Making? International Journal of Evidence and Proof, 24(3), 255-279.
MacCrimmon, M. (1991). The Social Construction of Reality and the Rules of Evidence. University of British Columbia Law Review, 25, 36-50.
MacCrimmon, M. (2001). What is Common about Common Sense: Cautionary Tales for Travelers Crossing Disciplinary Boundaries. Cardozo Law Review, 22, 1433-1460.
Medina, J. (2013). The Epistemology of Resistance. Gender and Racial Oppression, Epistemic Injustice, and Resistant Imaginations. Oxford University Press.
Ministry of Justice (2022). Diversity of the Judiciary: 2022 Statistics. Retrieved June 15, 2023, from https://www.gov.uk/government/statistics/diversity-of-the-judiciary-2022-statistics
Monteith, K. KC, Quinn, E., Dennis, A. L., Joseph-Salisbury, R., Kane, E., Addo, F. and McGourlay, C. (2022). Racial Bias and the Bench. A Response to the Judicial Diversity and Inclusion Strategy (2020-2025). Retrieved June 15, 2023, from https://documents.manchester.ac.uk/display.aspx?DocID=64125#:~:text=Racial%20Bias%20and%20the%20Bench%20was%20created%20in%20response%20to,judges%20and%20other%20legal%20professionals.
Nance, D. A. (2001). Naturalized Epistemology and the Critique of Evidence Theory. Virginia Law Review, 87(8), 1551-1618.
Nicolson, D. (2000). Gender, Epistemology and Ethics: Feminist Perspectives on Evidence Theory. In M. Childs and L. Ellison (Eds.), Feminist Perspectives on Evidence. Cavendish Publishing.
Ormerod, D. (2020). Racism in the Criminal Justice System. Criminal Law Review, 8, 659-662.
Owusu-Bempah, A. (2017). Defendant Participation in the Criminal Process. Routledge.
Owusu-Bempah, A. (2018). The Interpretation and Application of the Right to Effective Participation. International Journal of Evidence and Proof, 22(4), 321-341.
Owusu-Bempah, A. (2022a). The Irrelevance of Rap. Criminal Law Review, 2, 130-151.
Owusu-Bempah, A. (2022b). Prosecuting Rap: What Does the Case Law Tell Us? Popular Music, 41(1), 427-445.
Pattenden, R. (2009). The Standards of Review for Mistake of Fact in the Court of Appeal, Criminal Division. Criminal Law Review, 1, 15-30.
Picinali, F. (2012). Structuring Inferential Reasoning in Criminal Fact Finding: An Analogical Theory. Law, Probability and Risk, 11(2-3), 197-223.
Picinali, F. (2016a). Base-Rates of Negative Traits: Instructions for Use in Criminal Trials. Journal of Applied Philosophy, 33(1), 69-87.
Picinali, F. (2016b). Generalisations, Causal Relationships, and Moral Responsibility. International Journal of Evidence and Proof, 20, 121-135.
Pohlhaus, G., Jr., (2017). Varieties of Epistemic Injustice. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Puddifoot, K. (2021). How Stereotypes Deceive Us. Oxford University Press.
Quirk, H. and Ormerod, D. (2022). The Draft Victims Bill. Criminal Law Review, 8, 621-625.
Redmayne, M. (2003). Myths, Relationships and Coincidences: The New Problems of Sexual History. International Journal of Evidence and Proof, 7(2), 75-101.
Redmayne, M. (2015). Character in the Criminal Trial. Oxford University Press.
Registry of the ECtHR (2022). Guide on Article 6 of the European Convention on Human Rights. Right to a Fair Trial (Criminal Limb). Retrieved June 15, 2023, from https://www.echr.coe.int/documents/guide_art_6_criminal_eng.pdf
Reece, H. (2013). Rape Myths: Is Elite Opinion Right and Popular Opinion Wrong? Oxford Journal of Legal Studies, 33(3), 445-473.
Roberts, P. (2022). Roberts and Zuckerman’s Criminal Evidence (3rd ed.). Oxford University Press.
Saul, J. (2017). Implicit Bias, Stereotype Threat, and Epistemic Injustice. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Scaife, R., Stafford, T., Bunge, A. and Holroyd, J. (2020). To Blame? The Effects of Moralized Feedback on Implicit Racial Bias. Collabra: Psychology, 6(1), art. 30.
Scheppele, K. L. (1994). Manners of Imagining the Real. Law and Social Inquiry, 19(4), 995-1022.
Schum, D. A. (2001). The Evidential Foundations of Probabilistic Reasoning. Northwestern University Press.
Simon-Kerr, J. (2021). Relevance through a Feminist Lens. In C. Dahlman, A. Stein and G. Tuzet (Eds.), Philosophical Foundations of Evidence Law. Oxford University Press.
Sullivan, M. (2017). Epistemic Justice and the Law. In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Taggart, J. (2022). Vulnerable Defendants and the HMCTS Court-Appointed Intermediary Services. Criminal Law Review, 6, 432-447.
Tanovich, D. M. (2017). Regulating Inductive Reasoning in Sexual Assault Cases. In B. L. Berger, E. Cunliffe and J. Stribopoulos (Eds.), To Ensure that Justice is Done: Essays in Memory of Marc Rosenberg. Thomson Reuters.
Thayer, J. B. (1898). A Preliminary Treatise on Evidence at the Common Law. Little Brown and Co.
Thomas, C. (2020). The 21st Century Jury: Contempt, Bias and the Impact of Jury Service. Criminal Law Review, 11, 987-1011.
Toole, B. (2021). Recent Work in Standpoint Epistemology. Analysis Reviews, 81(2), 338-350.
Trechsel, S. (2005). Human Rights in Criminal Proceedings. Oxford University Press.
Trechsel, S. (2018). The Character of the Right to a Fair Trial. In J. Jackson and S. Summers (Eds.), Obstacles to Fairness in Criminal Proceedings. Individual Rights and Institutional Reforms. Hart Publishing.
Tremain, S. (2017). Knowing Disability, Differently. In In I. J. Kidd, J. Medina and G. Pohlhaus Jr. (Eds.), The Routledge Handbook of Epistemic Injustice. Routledge.
Twining, W. (2006). Rethinking Evidence. Exploratory Essays (2nd ed.). Cambridge University Press.
Waite, L. (2017). Asylum Seekers and the Labour Market: Spaces of Discomfort and Hostility. Social Policy and Society, 16(4), 669-679.
Ward, T. and Fouladvand, S. (2021). Bodies of Knowledge and Robes of Expertise: Expert Evidence about Drugs, Gangs and Human Trafficking. Criminal Law Review, 6, 442-460.
Wasserman, D. T. (1991). The Morality of Statistical Proof and the Risk of Mistaken Liability. Cardozo Law Review, 13, 935-976.
Al-Khawaja and Tahery v United Kingdom [GC], No. 26766/05 and No. 22228/06, European Court of Human Rights, 15 December 2011.
Allan v United Kingdom, No. 48539/99, European Court of Human Rights, 5 November 2002.
Behrani v Albania, No. 847/05, European Court of Human Rights, 27 May 2010.
Bryant v State, 802 N.E.2d 486 (Ind. Ct. App. 2004).
Criminal Justice and Public Order Act 1994. https://www.legislation.gov.uk/ukpga/1994/33/contents
Criminal Procedure Rules 2020 (United Kingdom). https://www.legislation.gov.uk/uksi/2020/759/contents/made
DPP v Kilbourne (1973) AC 729 (HL)
English and Welsh Victims Bill 2022 (Draft). https://www.gov.uk/government/publications/draft-victims-bill
European Convention on Human Rights, Rome, 4 November 1950. https://www.echr.coe.int/documents/d/echr/convention_eng
Human Rights Act 1998. https://www.legislation.gov.uk/ukpga/1998/42/contents
Jalloh v Germany [GC], No. 54810/00, European Court of Human Rights, 11 July 2006.
JL v Italy, No. 5671/16, European Court of Human Rights, 27 May 2021.
Khamidov v Russia, No. 72118/01, European Court of Human Rights, 15 November 2007.
Perez v France [GC], No. 47287/99, European Court of Human Rights, 12 January 2004.
Police and Criminal Evidence Act 1984. https://www.legislation.gov.uk/ukpga/1984/60/contents
R. v Horncastle [2010] 2 AC 373.
R. v Saleem [2007] EWCA Crim 1923.
SC v UK, No. 60958/00, European Court of Human Rights, 15 June 2004 (2005).
Youth Justice and Criminal Evidence Act 1999. https://www.legislation.gov.uk/ukpga/1999/23/contents
* I am very grateful to Abenaa Owusu-Bempah for her insightful feedback on an earlier draft. I have presented some of the ideas in this article at the 1st Michele Taruffo Girona Evidence Week, Universitat de Girona, 23-27 May 2022 and at the online conference La Valutazione della Prova Scientifica, organised by the Università degli Studi di Roma, 19 November 2021. I have also presented an earlier draft at the LSE Law Staff and PhD Lunchtime Seminar on 13 October 2022 and at the seminar series Thinking Like a Fact Finder, University of Bologna, 24 May 2023. I am grateful to the participants in these events for their useful questions and criticisms. Finally, I thank my law of evidence students for the constructive discussion we had in class on an earlier draft.
1 Here I am referring to the familiar notion of «logical relevance»—which I will clarify in the next section—not to the Wigmorean notion of «legal relevance», which explicitly encompasses practical, if not also moral, considerations. See Roberts (2022, p. 113-117).
2 On the self-restraint of the Court of Appeal with respect to «factual» findings of trial judges, see Pattenden (2009). On the self-restraint of the ECtHR with respect to the overall question of the admissibility of the evidence, see Jackson and Summers (2012, p. 81-83) and Goss (2016, p. 58-62).
3 As far as national courts are concerned, one should remember that both section 3 and section 6 of the Human Rights Act 1998 require them to comply with, and implement, the Convention rights; and that, according to section 2 of the Act, in performing these functions national courts «must take into account» the interpretation that the ECtHR has given to such rights.
4 See, among others, Khamidov v Russia, No. 72118/01, ECtHR, 15 November 2007 and Behrani v Albania, No. 847/05, ECtHR, 27 May 2010.
5 The literature on naked statistical evidence and that on behavioural generalisations have already investigated the possibility of immorality in evidential reasoning. See, in particular, Wasserman (1991), Picinali (2016a) and Picinali (2016b). Notably, evidential reasoning is related to, but separate from, the question concerning what to do once the evidence has been assessed (i. e., whether to convict or acquit). This question being about which course of action to take, it is straightforwardly susceptible to moral evaluation.
6 But see also the freshly published Lackey (2023), discussing cases of «agential testimonial injustice» in the US criminal justice system. Unlike the standard notion of testimonial injustice, agential testimonial injustice involves a credibility excess. See below footnote 25. In this article I will work with the standard notion only, but my considerations concerning participation and trial fairness may well apply to cases of agential testimonial injustice suffered by the defendant or the complainant.
7 See Twining (2006, p. 75-86) and Anderson et al. (2005, p. 78-87).
8 For further treatment of the notion of «generalisation», see Picinali (2012, p. 199-202), Schum (2001, p. 81-83) and Anderson et al. (2005, p. 262-280).
9 For a more detailed discussion of the assessments of relevance and of probative value in probabilistic terms, see Lempert (1977), Nance (2001, p. 1595–1599) and Kaye (1986).
10 A proper discussion of this claim would require more space than I can give to it here, but consider the works cited in the following note. Below are a few telling examples drawn from the rationalist tradition, all positing as privileged stock of knowledge that resulting from experience allegedly shared by everyone in a society. If indeed there is such a stock of knowledge, it surely is too narrow in scope to serve as an adequate basis for fact finding. The worry is that references to a «common stock of knowledge» and to «general experience» are unwittingly used to impose a particular cognitive viewpoint. See Thayer (1898, p. 265), famously stating that «[t]he law furnishes no test of relevancy. For this, it tacitly refers to logic and general experience» (italics added); Cohen (1977, p. 274-275), arguing that «[t]he main commonplace generalizations themselves are for the most part too essential a part of our culture for there to be any serious disagreement about them» and that, therefore, «on any rational reconstruction» of disagreements between members of the jury, they are not due to a disagreement on the soundness of such generalisations; Cohen (1983, p. 4), endorsing «the common belief in a universal cognitive competence whereby, given a proper presentation of all the relevant evidence about any particular factual issue, either every normal and unbiased person would come to the same conclusion about it or at worst everyone would agree that it was an issue about which the norms of proof are indeterminate and reasonable people might venture different conclusions»; and Dennis (2020, p. 65). Commenting on J. F. Stephen’s classic definition of relevance (and, by and large, endorsing that definition) Dennis writes: «Stephen’s definition requires relevance to be assessed according to the common course of events. In deciding whether something may be inferred about the existence of fact A from proof of fact B, reliance is to be placed on the common stock of knowledge about the world; in other words, on logic, common sense and general experience […] The only “laws” which identify the existence of a relationship between facts are the laws of nature and human behaviour» (italics added).
11 For criticisms of the assumption, see MacCrimmon (1991, p. 37-39; 2001, p. 1445-1448), Kinports (1991, p. 430-434), Scheppele (1994, p. 1010-1012), Hunter (1996, p. 131), Nicolson (2000, p. 21-25), Simon-Kerr (2021, p. 367-370), and Gonzales Rose (2017, p. 2280-2284, 2298-2302).
12 See Anderson et al. (2005, p. 265-266).
13 A strand of contemporary epistemological thought has developed precisely from the realisation that experience is not constant across a society. This is standpoint epistemology, which claims that «what one is in a position to know depends on facts about that person’s social identity» (Toole, 2021, p. 340).
14 An instructive read on this point is Akala (2018, ch. 7).
15 In particular, racism pervades virtually all stages of English and Welsh criminal proceedings, including the first steps of police investigation. For a recent and concise picture, see Ormerod (2020) and Johnson (2022). For a more comprehensive treatment, see Lammy (2017) and Casey (2023)..
16 For further discussion of the example of flight from the crime scene, see Gonzales Rose (2017, p. 2269-2288; 2021, p. 386-387) and Roberts (2022, p. 158-160).
17 See, among others, the works referenced in footnote n 11 above.
18 Italics in the original.
19 In this essential taxonomy I am relying on Fricker’s work. Cf. Pohlhaus (2017, p. 19-21).
20 See Fricker (2017, p. 53). Cf. Coady (2017).
21 See Fricker (2007, p. 152-161). This represents the central, or «systematic», case of hermeneutical injustice. Fricker also discusses the «incidental» case, which does not involve identity prejudice and, hence, lacks the character of systematicity.
22 See Fricker (2007, p. 27, 35).
23 Fricker (2007, p. 35) also talks in terms of a «negative identity-prejudicial stereotype». By using the concept of a stereotype, she stresses the fact that the disparaging association is widely held. Kathy Puddifoot has recently argued that something akin to testimonial injustice may be also produced by reliance on accurate associations. See Puddifoot (2021, p. 86-89).
24 Consider, for instance, Dotson’s notion of «contributory injustice», which occurs when there are different hermeneutical resources that the agent could utilise in a given interpersonal epistemic practice (including resources that would allow a member of a marginalised social group to make visible and explain to others their social experience), but the agent continues to use biased epistemic resources due to willful ignorance. Being thwarted by this behaviour, members of the marginalised group cannot contribute to shared knowledge in the epistemic practice. See Dotson (2012, p. 31-32).
25 See Fricker (2007, p. 27-28). This represents the central, or «systematic», case of testimonial injustice. Fricker also discusses the «incidental» case, which does not involve identity prejudice and, hence, lacks the character of systematicity. Cf. Medina (2013, p. 59-67)—arguing that excessive attributions of credibility are a type of, and contribute to, testimonial injustice; and Lackey (2020, p. 59-63)—arguing that credibility excesses afforded in case of a false confession are a form of testimonial injustice. For further treatment, see Lackey (2023). I will not deal with credibility excesses in this article.
26 On this point, see Jalloh (2022, p. 638). One may object that, according to the definition of testimonial injustice just offered, there wouldn’t be testimonial injustice if, absent the identity prejudice, the hearer would have assigned to the testimony a deflated credibility level anyway, even if higher than that actually assigned. Given this counterfactual, identity prejudice can be said to have caused a reduction of the assigned credibility level, but not the credibility deficit or the deflation themselves: in the counterfactual scenario too, the credibility level assigned by the hearer would have been lower than that warranted by the evidence. And yet, it seems that an injustice did occur, given that identity prejudice was responsible for the lowering of the credibility level actually assigned, compared to that assigned in the counterfactual scenario. To address this problem, one may redefine testimonial injustice such that identity prejudice need not cause a deficit but needs to cause a reduction in the credibility level, and that the resulting credibility level needs to be deflated—that is, it needs to be below that warranted by the evidence.
27 See Fricker (2017, p. 54).
28 Identity prejudice may well operate through an implicit bias, that is, an association between a social group and a stereotypical characteristic, which affects one’s cognition and reasoning and is, by and large, outside of one’s awareness and control. As Jennifer Saul (2017, p. 236-237) has perceptively shown, though, testimonial injustice and implicit bias are distinct phenomena.
29 In fact, this extensive notion is also endorsed in Fricker (2007, p. 130), writing that a significant form of injustice «occurs when prejudice […] leads to a tendency for some groups simply not to be asked for information in the first place. Now this most straightforward of epistemic exclusions—barred entry to the community of informants—is obviously […] a crucial feature of the politics of epistemic real life. The exclusion in fact marks a commonplace form of testimonial injustice: those social groups who are subject to identity prejudice and are thereby susceptible to unjust credibility deficit will, by the same token, also tend simply not to be asked to share their thoughts, their judgements, their opinions […] This kind of testimonial injustice takes place in silence. It occurs when hearer prejudice does its work in advance of a potential informational exchange: it pre-empts any such exchange. Let us call it pre-emptive testimonial injustice» (italics in the original).
30 See the hypothetical Rap lyrics below, where the ignored stock of knowledge is that of the defendant qua member of the rap art community, whereas the identity prejudice is against Black youth. Cases where the social group whose stock of knowledge is ignored and the social group targeted by the identity prejudice do not match (while perhaps overlapping) may be seen as «peripheral» cases of testimonial injustice in the context of evidential reasoning.
31 The injustice described here may have a hermeneutical component insofar as ignoring the experience of the social group means that the adjudicator also ignores hermeneutical resources that would help with the assessment of the party’s evidential argument. This occurs in the hypothetical Rap lyrics below. See infra footnote n 36.
32 One could argue that if a prosecution or defence witness, other than the complainant or the defendant themselves, suffers testimonial injustice, then the complainant’s or the defendant’s own opportunity for participating in fact finding is indirectly undermined, and unfairness arises. While reasonable, this is not an argument I will pursue here.
33 I am grateful to Rachel Herdy for pushing me to consider the apparent narrowness of my account.
34 One can construct similar examples based on ss 36 and 37 CJPOA 1994 referring, respectively, to an arrested suspect’s failure to explain to the police their possession (or the presence on their person) of objects, marks, or substances, and to an arrested suspect’s failure to explain to the police their presence at a particular place.
35 See Hulley and Young (2022, p. 728), concluding, on the basis of empirical research, that «[f]or young Black men […] remaining silent in a suspect interview offers protection against the racialised misinterpretation of talk and the potential legal harm that it exposes».
36 See Owusu-Bempah (2022a, p. 130-131; 2022b, p. 430-31). Jalloh (2022, p. 643-644) argues that the prosecution’s use of drill lyrics and videos in court is an instance of contributory injustice (see supra, n 24 for a definition of this form of epistemic injustice) precisely because the hermeneutical resources that would allow for an informed assessment of the significance of these materials are available, but they are wilfully ignored by prosecutors, who instead proffer street-illiterate and racist interpretations. To be sure, Jalloh (p. 638) also argues that the current use of drill as incriminating evidence in court is an instance of testimonial injustice.
37 See Owusu-Bempah (2022a, p. 131-132; 2022b, p. 428).
38 For a possible example, see the case of R. v Saleem (2007) and the commentaries by Redmayne (2015, p. 159-161) and by Owusu-Bempah (2022a, p. 141). In R. v Saleem (2007) the rap lyrics contained a reference to the date in which the crime charged occurred, this being the birthday of the defendant. The lyrics recited: «Im gon make history, 1stly dey gon call me mister an dey gon say I dissed ya, I hav 2 b carfull hu I talk 2 becos ur bird wil be da listner, 2ndly February 24th my birth day im gon make it ur worst day, 3rdly do I have 2 have u layin in emergency 2 have dem stitch ya» (§9). Perhaps, a better example is provided by the American case of Bryant v State (2004), where the defendant was charged with the murder of his stepmother, who was found in the trunk of the defendant’s car. The rap lyrics used in evidence stated: «[c]uz the 5–0 won’t even know who you are when they pull yo ugly ass out the trunk of my car» (p. 498).
39 Cf. Owusu-Bempah’s (2022a, p. 149) checklist for the purposes of a reasonable assessment of the relevance of rap lyrics.
40 See Owusu-Bempah (2022a, p. 135-136, 147-148; 2022b, p. 439-440). See also Ward and Fouladvand (2021, p. 454). Discussing the prosecution’s practice of relying on police officers as experts on rap lyrics and their significance, Ward and Fouladvand criticise officers for not presenting the «range of opinion» on the matter (as rule 19.4(f) of the Criminal Procedure Rules 2020 requires experts to do), offering instead a univocal «narrative that resonates with stereotypes on black criminality». This wilful ignorance of street-literate interpretations that are informed by the stylistic rules and topoi of the genre is an instance of contributory injustice that could be remedied with a more frequent use of defence rap experts. See supra footnote n 36 and Jalloh (2022, p. 647-648).
41 There is disagreement within the empirical literature about the degree of support that this generalisation enjoys. Compare Leverick (2020, p. 269)—finding that the generalisation enjoys substantial support amongst mock jurors—and Thomas (2020, p. 1001-1002)—finding that the generalisation enjoys little support amongst actual jurors.
42 The problem discussed here can also be framed in terms of the relevance and probative value of the absence of signs of resistance—rather than in terms of the relevance and probative value of the complainant’s testimony. This alternative is formal, rather than substantial, since the ultimate question is invariably that of the credibility of the complainant.
43 In fact, this definition of rape myths departs from that normally employed in the literature. See Gerger et al. (2007, p. 423), defining rape myths as «descriptive or prescriptive beliefs about rape (i. e., about its causes, context, consequences, perpetrators, victims, and their interaction) that serve to deny, downplay or justify sexual violence that men commit against women» (italics in the original). In any case, the presentation of the complainant as not credible is inherent to a rape myth, even when understood as per the normal definition. The literature on rape myths is vast, and in recent years there has been a lively academic discussion as to the popularity of such myths and the extent to which they affect jurors’ decision making. See, among others, Redmayne (2003), Reece (2013), Leverick (2020), Thomas (2020, p. 998-1005), Chalmers et al. (2021), and Daly et al. (2023).
44 Consider the recent decision of the ECtHR in the case JL v Italy, No. 5671/16, ECtHR 27 May 2021. The case involved the application from a complainant in a trial for rape, who argued that her right to private life under art. 8 of the Convention had been breached by the member state due to how the criminal proceedings were conducted. The ECtHR condemned Italy on the ground that the Italian Court of Appeal, which ultimately acquitted the defendants, ostensibly relied on generalisations linking the attire, the family situation, the sexual orientation, the sexual history, and the artistic endeavours of the complainant to a lack of credibility (see §134-143). If the argument defended in this article is correct, it should be possible to bring a successful claim of this kind also under Art 6 ECHR. See infra footnote n 62 and accompanying text.
45 Conversely, if the adjudicator failed to consider the stock of knowledge of the defendant due to non-prejudiced, accidental, oversight, their assessment of relevance and probative value may be inaccurate, but not also a case of testimonial injustice.
46 Referring to the virtue, Fricker writes that «[i]ts possession requires the hearer to reliably neutralize prejudice in her judgements of credibility […] The virtue is, most basically, a matter of one’s credibility judgements being unprejudiced. That they should be unprejudiced might result from their being prejudice-free from the start, or it might result from the influence of prejudice being somehow neutralized» (p. 92-93, italics in the original). Later, she points out that the virtue «is bound to be hard to achieve, owing to the psychologically stealthy and historically dynamic nature of prejudice» (p. 98) and that it is «a third basic virtue of truth [besides accuracy and sincerity], for the reason that it frees the hearers […] from the prejudice that would cause them to miss out on truths they may need» (p. 116-117).
47 See Sullivan (2017, p. 295, 300). See also Jalloh (2022, p. 644-649), presenting individual and structural approaches to address the epistemic injustice that is frequently involved in the evidential use of drill lyrics and videos in the criminal justice system. Finally, see Griggs (2021, p. 943-944) on the importance of an antiracist education of «insiders» to the legal system, starting from the law school classroom.
48 In addition to the literature that I have referenced so far, consider the findings of a recent report on racial bias in the judiciary, published by the University of Manchester. See Monteith et al. (2022, p. 16).
49 On the wrong of epistemic injustice, see Fricker (2007, p. 43-59 and ch. 6). Cf. Congdon (2017, p. 247-250). Reflecting on the intrinsic, or primary, harm of testimonial injustice, Fricker (2007, p. 52-54) points out that being excluded from an epistemic exchange means being denied the opportunity to settle or steady one’s mind through trustful conversation with others; and that the inability to settle the mind may produce the inability to settle aspects of one’s identity.
50 As an example of bad luck, Fricker (2007, p. 41) mentions the case of a shy speaker who «avoids looking [the hearer] in the eye, frequently looks askance, and pauses self-consciously in mid-sentence as if to work out his story». The hearer may give a low credibility judgement on the basis of this behaviour, rather than because of identity prejudice. As an example of the speaker’s own fault, Fricker (p. 42) mentions the case of «Matilda, who told such dreadful lies that her reputation justifies the hearer’s disbelief when she is exclaiming truthfully from the window that the house is on fire».
51 See Fricker (2007, p. 41-43). Cf. Dotson (2012, p. 37-41). Fricker’s view is that injustice requires culpability on the part of the speaker and that, when the harm is caused by identity prejudice, the hearer is nearly always culpable to some degree, even if they do not harm the speaker intentionally (see infra footnote n 65). As mentioned earlier, in case of intentional harm there is still epistemic injustice, but not of the testimonial kind.
52 As is well known, section 2(1) of the Human Rights Act 1998 requires English and Welsh courts to «take into account» the ECtHR’s case law. This means that the ECtHR’s understanding of the right to a fair trial carries considerable weight in national decisions. Cf. R. v Horncastle (2010, p. 432), where the Supreme Court clarified that «[t]he requirement to “take into account” the Strasbourg jurisprudence will normally result in the domestic court applying principles that are clearly established by the Strasbourg court. There will, however, be rare occasions [in which] it is open to the domestic court to decline to follow the Strasbourg decision, giving reasons for adopting this course».
53 The centrality of participation within the notion of fairness is also evident in the guidance on art. 6 produced by the Registry of the ECtHR (2022). The ECtHR has, indeed, recognised the specific right to «effective participation», treating this as the key component of the broader right to a fair trial. See Registry of the ECtHR (2022, p. 32-33) and SC v UK, No. 60958/00, ECtHR 15 June 2004 (2005), §28-29. For a critical assessment of the right to «effective participation», see Owusu-Bempah (2018).
54 See, for example, Jalloh v Germany [GC], No. 54810/00, ECtHR 11 July 2006, §96 and Allan v United Kingdom, No. 48539/99, ECtHR 5 November 2002, §42-43, 48.
55 See also Jackson and Summers (2018, p. 9). For a general study of the legitimate role of the defendant in the criminal process, see Owusu-Bempah (2017). Relying, in particular, on the privilege against self-incrimination and on the right to silence, Owusu-Bempah argues that when participation becomes the content of a duty, as opposed to a right, of the defendant, the right to a fair trial is undermined. In other words, according to Owusu-Bempah (p. 6-8, 66-73), for a trial to be fair participation has to be an option, not a requirement.
56 While not necessary, it is possible that some of the interests of the defendant compete with some of the interests of the other parties (in particular, the complainant) in a zero-sum game of sorts, such that what is owed to a party inevitably depends on what is owed to another, and that a party’s interest may need «sacrificing» (completely or partially) for the sake of achieving fairness. It is in situations such as this that the metaphor of balancing is usually deployed to describe the reasoning on which a fair distribution of prerogatives is premised. See, for example, Duff (2018, p. 308), claiming that «fairness to the accused is a matter of giving the accused what is due to him or to her, in relation to other participants in the process—a matter of weighing the accused’s claims against those of other parties, and striking a fair balance between them». See also infra footnote n 60. There are, however, contexts in which the interests of the parties do not compete in a zero-sum game. Laura Hoyano (2015, p. 107) convincingly argues that one such contexts is that of special measures for vulnerable witnesses/complainants.
57 Duff (2018, p. 308) quotes Trechsel with approval. See also Hoyano (2014, p. 24-25), writing that «[i]t is not only the defendant who can lay claim to the right to a fair trial, but all participants, and so the court has an obligation to ensure that judicial processes are conducive to a trial that is fair to all». Finally, see the Criminal Procedure Rules 2020, rule 1.1(2)(c), acknowledging that fairness is not a prerogative of the defendant only.
58 See Perez v France [GC], No 47287/99, ECtHR 12 January 2004. See also Trechsel (2005, p. 40-42).
59 See Al-Khawaja and Tahery v United Kingdom [GC], No 26766/05 and 22228/06, ECtHR 15 December 2011, §146.
60 For critical assessment of the ECtHR case law in point and, more generally, of the deployment of the metaphor of «balancing» by the ECtHR and national courts alike, see Hoyano (2014) and Campbell et al. (2019, p. 42-44). See also supra footnote n 56.
61 Notice that this opportunity is clearly central to the English and Welsh Victims Bill (Draft) 2022. Clause 2 of the Bill instructs the Secretary of State to issue a code of practice for victims implementing «the principles that victims […] (iii) should have the opportunity to make their views heard in the criminal justice process; (iv) should be able to challenge decisions which have a direct impact on them». For a brief commentary on the Bill, see Quirk and Ormerod (2022). For a theoretical discussion of victims’ participatory rights, see Doak (2008, ch. 3).
62 Trechsel argues that the complainant too should be afforded a right to a fair trial, even if they do not bring a civil action for damages in the proceedings. See Trechsel (2018, p. 34; 2005, p. 41-42). Notice that recognising that the complainant has a right to a fair trial surely need not mean that, were the defendant acquitted as a result of an unfair trial, the complainant should be granted a new trial. That the complainant has a right to a fair trial may merely mean that, in the hypothesised scenario, the complainant could bring their claim to court and, if successful, obtain compensation, as happens with other rights in the Convention. Obviously, though, the criminal justice system should focus on preventing, rather than remedying, violations of the complainant’s right.
63 Notice that, if the defendant is convicted as a result of a trial that was unfair because the complainant was denied treatment that is owed to them, it does not necessarily follow that the defendant has a ground of appeal. True, their right to a fair trial has been breached, but distinctions can be made with respect to the causes of unfairness; and only the causes pertaining to the treatment of the defendant may be considered grounds to appeal a conviction. I am also open to the possibility that not all instances of testimonial injustice suffered by the defendant warrant an appeal, notwithstanding that, as clarified in the next paragraph, they do render the trial unfair.
64 There is a complicating factor to bear in mind. The defendant’s participatory prerogatives—which are central to trial fairness—can be, and often are, exercised by proxy, that is, via the defendant’s lawyer (for critical assessment, see Owusu-Bempah (2018, p. 331-333)). It is possible that the lawyers themselves commit testimonial injustice against the defendant, and defend their client in a way that does not appropriately reflect the defendant’s experience. In such a case, although the defendant’s participatory prerogatives would apparently be exercised and the trial would apparently be fair, the defendant’s participation would be, in fact, thwarted by the testimonial injustice at the hands of their lawyer and, as a result, the trial would be, in fact, unfair. Something similar may occur in the interaction between prosecution and complainant.
65 See Fricker (2017, p. 53-55), pointing out that in her view the lack of intentionality does not entail lack of culpability on the part of the hearer. In fact, she assumes that «in connection with testimonial injustice […] prejudiced thinking is almost always culpable in some degree», given that the hearer may be «colluding with the forces of prejudice» even when they are wilfully ignorant, or may be in a «self-interested or plain lazy denial», about harbouring prejudice (p. 55).
66 Relatedly, some philosophers doubt whether the constitutive elements of testimonial injustice can be ascertained in any given case. See Arcila-Valenzuela and Páez (2022).
67 In this regard, it is instructive to consider the flaws of the main training tool for judges on the issue of equal treatment, that is, the Judicial College’s Equal Treatment Bench Book (2021). These flaws are identified and discussed in Monteith et al. (2022, p. 26-28).
68 Notably, this section, as well as other aspects of sexual offences prosecutions, are the subject of a recent consultation paper by the Law Commission. See Law Commission (2023).
69 See Owusu-Bempah (2022a, p. 150-151).
70 See Judicial College (2020, p. 20.5 ff).
71 See Sullivan (2017, p. 297) and Owusu-Bempah (2022a, p. 148). The unsatisfactory degree of diversity of the English and Welsh judiciary is evident from Ministry of Justice (2022).
72 See rules 21.5 and 22.3, respectively.
73 The test can be taken at this link https://implicit.harvard.edu/implicit/takeatest.html (accessed 15 June 2023).
74 See Sullivan (2017, p. 300) and supra footnote n 28.
75 See Sullivan (2017) and Scaife et al. (2020). The latter study shows that a blaming response to an individual’s implicit bias can reduce this individual’s IAT score in the short term and can motivate the individual to change behaviour influenced by implicit bias.
76 See Dotson (2012).
77 Of course, a more inclusive evidential reasoning is not sufficient to address all the problems that the underlying social injustice poses for criminal justice. For discussion, see Lacey (2022).
78 While the general rule is that an asylum seeker is forbidden from working, there are very limited opportunities for an asylum seeker to work legally if their claim is still outstanding 12 months after submission, and the applicant is not responsible for this delay. See Home Office (2022).
79 Cf. Crown Prosecution Service (2021)—discussing the evidential role of money found in the possession of the defendant, with particular reference to the crime of possession with intent to supply. Notice that the generalisation mentioned in the hypothetical is, in fact, the combination of two generalisations, one referring to drug dealers, the other referring to non-drug dealers.
80 See Waite (2017).
81 In formulating this example, I am well aware that in recent years there has been a noticeable cultural change in legal practice for what concerns the testimony of those affected by a disability, partly stimulated by the Youth Justice and Criminal Evidence Act 1999, which provides a series of special measures designed to support the witness and to alleviate the risk of misjudging their testimony. In fact, this risk is probably greater in the case of the disabled defendant, who does not enjoy the same extent of support, in particular, for what concerns the resort to live links and to intermediaries. See Fairclough (2018) and Taggart (2022).
82 See Tanovich (2017, p. 78-81)—discussing, with references to Canadian case law, how stereotypes about disability can affect credibility judgments. For a general discussion of the epistemic injustice affecting the disabled, see Tremain (2017).