The present article focuses on the problem of accountability in criminal adjudication. First, I discuss the seemingly compelling reasons for giving an account of one’s decision and show how this requirement can be easily tackled through legal realism. Justifying a criminal verdict will not automatically lead to more accurate (truth-conducive) decisions. Second, I illuminate the epistemological framework underlying the criminal process, i.e. the Common Sense philosophy. I suggest that common sense cannot deliver what it promises: valid inferential relations between the evidence and the verdict, let alone a stable structure of justification. Utilizing inner sensations (e.g. “feeling sure” or “beyond reasonable doubt”), understandable only by the person feeling them, is an open invitation to self-deception, for it lacks criteria of correctness. Third, I bring the private-language-argument into play and show that the idea of an inner sensation as a standard of proof in criminal adjudication is incoherent, indeed unintelligible. Finally, I make the case for a new turn to epistemology, which can provide us with a new, sustainable framework, i.e. a structure of justification, which can be applied, communicated and scrutinized. Such a turn will not interfere with the validity of the currently valid system of criminal adjudication. Giving reasons for decisions is one thing; having reasons for a decision is quite another.
Keywords: criminal adjudication, common sense, accountability, black-box, reasons for decisions, private language
1. The Problem of Accountability – Accountability as a Problem
1.1 Trial by Jury
Throughout its “great and strange career” the trial by jury has been regarded as a cornerstone of – what we, simply at the cost of oversimplification, would call – the criminal adjudication of common law countries and systems of evidence. Notwithstanding the fact that only a small percentage of criminal cases (1% in England and Wales) culminate in trial by jury, the system in which fact-finders (jurors or professional judges) determine the guilt of the defendant, arrive at holistic judgments and deliver unreasoned verdicts, has quite successfully and despite all its institutional adaptations immunized itself against criticism as a “perfectly legitimate model of fact-finding”. Criminal adjudication finds itself on the horns of a structural dilemma, to wit, rules of admissibility (gate-keeping decisions on behalf of criminal judges) or rules of epistemic evaluation of the fact-finder’s decision. The Anglo-American criminal justice system has traditionally chosen the former. However, our traditional focus on exclusionary rules comes at the cost of neglecting central aspects of epistemic evaluation of the decision-making process. As a result, the fact-finder is not required, indeed not permitted to give an account of the respective verdict, i.e. they do not have to explain the underlying reasoning process. This institutional fact may suffice to explain why the recent decision of the Grand Chamber of the European Court of Human Rights, which recognized that the practice of providing no reasons for decisions does not infringe fundamental rights of the accused, has in general been welcomed with some sort of relief. For common sense justice and its main feature, the lack of accountability, i.e. the lack of any legal obligation to provide reasons for decisions, have traditionally been subjected to severe criticism. Now, the question is: What has the practice of providing unreasoned verdicts been criticized for? I will organize the plethora of partially overlapping and competing rationales into three groups according to their structural characteristics.
1.1.1 Political Morality
Accountability is all the rage in modern political and legal philosophy. Accountability can be regarded as a basic tenet, indeed as a bedrock political principle in liberal societies since Plato, whose concept thereof (“λόγον διδόναι”) meant that a public decision maker has the duty to give a systematic account, which is explanatory and open to scrutiny. Although the debate is old, the subject is still fresh – fresher and more relevant now than it was a few decades ago. Echoing through the centuries the concept of accountability has found its place at the heart of modern political rationale. JF Kennedy stressed in his address before the American Newspaper Publishers Association that “the very word ‘secrecy’ is repugnant in a free and open society”. Openness enables us seeing which values and which style of argumentation fact-finders bring into their decision making. With a conclusion that stands alone, we are simply resting our case on a brute assertion, whereas providing reasons for decisions fosters public legitimation of legal decisions. Conversely, lack of accountability can have deleterious consequences for the fabric of society, as it removes justifications and reasoning processes entirely from the public arena.
Reasons enable us to examine whether the fact-finders manage to perform their task and evince respect even for the (defeated) parties. In fact, accountability produces an effect both vertically (to those higher up in the hierarchy but also to the defendant) and horizontally by securing uniformity of similar decisions across different jurisdictions. Providing reasons is the best-known strategy to dissipate conflicts over facts, as it makes the outcome (more) bearable. An oracular, “sphinx-like verdict” stated in one or two words (guilty or not guilty) is no comfort for the convicted, since it is impossible to know which reasoning patterns the fact-finders have used to arrive at their decision. Epistemic justification can instill confidence in the accuracy of the decision-making process and thus prevent a disagreement over facts from mutating into recriminatory sources of further conflict. It secures public confidence in the decision. Vital as the parental reason of the last resort – just think of the phrase: ‘because I say so!’ – may be in certain circumstances, it is anything but helpful in institutional settings of liberal democratic societies. Publicity – and that includes epistemic justification, too – is the “very soul of justice”, as Bentham pithily illustrated it. Secrecy and democracy seem to have an antithetical nature. Therefore, we can brand decision-making environments devoid of accountability as the hallmark of an authoritarian regime, indeed as undemocratic. It is no compliment, Schauer remarks, to characterize a conclusion as an ipse dixit – a bare assertion unsupported by reasons. The present analysis explains why especially criminal adjudication should be treated as an applied branch of political morality or as a barometer of current constitutional principles. Giving an explanatory account of one’s decisions is a paramount duty of the fact-finder in liberal societies.
The second category of criticism derives from the belief that first, rationality has to be regarded as a cardinal epistemic attainment and second, justified verdicts are substantially better verdicts. Accountability triggers cognitive processes which, in turn, lead to increased effort for impartiality and rigorous assessment of the arguments in play. Specifically, it has been argued that awareness of the cognitive processes induced by accountability can reduce cognitive biases. This will, in turn, ameliorate the routines for testing and validating knowledge claims in criminal adjudication. Rationality resembles thus a sword of Damocles hanging over the heads of fact-finders. Criminal verdicts qua justified belief is a function of a rational process of evaluating reasons. To give reasons, Jackson points out, “concentrates the mind with the result that a reasoned decision is more likely to be soundly based on the evidence than one which is not reasoned”. The chain of reasoning enables us to ascertain that the criminal verdict is via justification-transmitting inferences connected with the evidence admitted at trial. Otherwise good decisions would be indiscernible from bad ones. As Lord Woolf has put it: “it is often difficult to know if the jury has made a mistake or not”. In other words, since a rational resolution of disagreement over facts requires a criterion of correctness, the requirement of giving reasons for decisions results to a greater incentive for decision makers to carefully assess the evidence and base their decisions on acknowledged facts, i.e. reveal and communicate their criterion.
The argument, according to which lack of accountability is an incentive for arbitrariness may adjoin to speculation yet is perfectly plausible. If, as George Herbert Mead emphatically put it, “the very process of thinking is, of course, simply an inner conversation that goes on”, then it is of great importance to disclose the transcript of that monologue. By entrusting ourselves with a judgment/verdict without opening up the decision-making process to transparency and scrutiny, we are actually begging the question before even the judgment has begun. If it is true that “[p]rocedure deals with the machinery by which controversies are settled”, it is imperative to understand how the engine works.
So far, we have taken a closer look at arguments which stretch from the political (accountability) to the decision-theoretical (rationality). One could demur that –at least from a legal point of view– there is insofar nothing to criticize as long as a decision is correct. After all, it is a common belief that fact-finders do “get it right”. However, a criminal verdict should do more than this. We need to identify which reasons vindicate the criminal verdict for a very important legal reason, too. By giving reasons for a decision, the fact-finder lays down inferential rules. And rules generate (by definition) normativity by guiding other fact-finders in similar cases in the future and by securing a level of uniformity of decision making. The same rules enable others to treat like cases alike and different cases differently, thus facilitating a coherent management of the “great multitude of decisions, emerging day by day” as well as a smooth application of the respective standard of proof by holding that such and such evidence (does not) meet its requirements.
Establishing consistency promotes justice and equality. Let me be more specific on that issue. We saw above that each verdict generates a horizontal effect. By being general, verdicts raise a claim to be applied not only to the individual case at hand but to a class of similar cases. The adhocness of a decision is incompatible with the very concept of law. Hart stresses that the law must refer to “classes of person, and to classes of acts”. Its successful operation hinges on the ability to see unique historic facts “as instances of the general classification which the law makes”. The insight that reasons for decisions are nothing but “results taken to a greater level of generality” enables us to understand how the verdict strips the evidence of its uniqueness and embeds it within a more abstract normative framework (legal rule). We operate thus on a two-way street, whereby the standard of proof is applied (top-down) but at the same time the fact-finder lays down a rule which communicates the inferential process which led to the decision and encompasses thus other cases (bottom-up), so that everyone may profit from his or her experience by following the (new) rule. The more importance we attach to consistency as a cornerstone of legal systems, the bigger the criticism against the practice of delivering naked verdicts becomes.
1.2 Legal Realism Strikes Back
Let us take stock. We have seen so far that the reasons for introducing accountability in criminal procedure are seemingly compelling especially in liberal societies. Accountability gains its character as an ineradicable element of (rational) decision making processes. So, how could the trial by jury and its main characteristic, to wit, the suspension of epistemic duties for fact-finders, defend its legitimacy? The answer seems to be rather simple: the debate on accountability, so the argument goes, contains a certain truth. But it still misses its target, since from a legal point of view it is indiscernible whether the written reasons (context of justification), which support and warrant a criminal verdict, bear any resemblance with the actual thought process, which led to the respective conclusion (context of discovery). This very fact suggests that we need to nuance the legal argument and distinguish between the reasons for decisions on the one hand and the arguments chosen to advertise and justify them on the other hand.
At this very moment the legal realist shows his teeth. Legal realism does not, of course, put the epistemic goal of issuing an accurate and fair verdict into doubt. The argument is a rather different one. Accountability-demands, so the legal realist, simply are not premised on the structure of the actual decision-making process. Therefore, we should, the legal realist contends, concentrate our enquiry on, and predict “what the courts will do in fact”.
Undeniably, the oft-repeated aphorism according to which criminal verdicts are nothing but an ex post facto rationalization of decisions made by ideology, came at a time when similar hypotheses could not find empirical confirmation. Times move on. It is no more mere speculation that extraneous determinants elicit the judicial decision. Nowadays, we should have, psychologists claim persistently, sufficient empirical evidence that even the caricatures of legal realism, e.g. the hypothesis that justice is a function of ‘what judges ate for breakfast’, are truth conducive. According to a recent empirical study the likelihood of a favorable ruling for male felons with rehabilitation program covaries not with the inferential force of the evidence but, surprisingly, with judges’ hunger and the time they had a break. Similar hypotheses could be formulated for matters of guilt and innocence.
By using terminology, which was unavailable at the time when Holmes, the father of American Realism, shook the rationality-driven legal community and their belief in a reliable faculty of legal reasoning, we could hold that even a (legal) obligation of accountability cannot inoculate professional judges, let alone jurors, from their cognitive biases. For the latter are robust mental phenomena. Fact-finders might, and probably will, first decide by hunch or by ideology and then deploy pre-existing reasoning patterns and suitable arguments, in order to accomplish their reasoning task. Empirical studies do not deny that reasoning is governed by accountability processes. The expectation that the fact-finder will be called to justify a conclusion puts significant pressure on him or her. In fact, no one wants to appear like a fool in front of the public or his or her peers. But this pressure will not make fact-finders work harder to weigh the evidence and discover the truth. In other words imposing accountability on decision-makers and fact-finders would not promote only explanatory thought (pre-decisional accountability) as a condition of good judgment but would amplify confirmation thought (post-decisional accountability) for an already taken decision too. And confirmatory thought is, as expression of confirmation bias, an “one-sided attempt to rationalize a particular point of view”. Legal realism disconnects epistemic justification from the rectitude of a verdict by disrupting the inferential link(s) between the evidence and the constituent elements of the offence.
To sum up, the epistemic virtue of accountability per se cannot bring us further. Fact-finders can simply choose inference patterns out of a catalogue and bypass any legal obligation to give an account of one’s epistemic activity. Providing reasons for decisions will create according to legal realists only the illusion of a rational decision-making process; the requirement of epistemic justification cannot therefore by itself pre-empt the risk of arbitrariness, but rather conceal it. Legal realism makes the debate on accountability seemingly shrink into nothingness. In order to unravel the real problem of accountability, we have to dig deeper.
2. What is Inside the Black Box?
2.1 Fact-finders in the Black Box
The expectation of fact-finders to give reasons for decisions does not necessarily equate better decisions. We should rather focus on a better grasp of the dynamics of evidence evaluation and patterns of inferential reasoning for the concretization of the respective standard of proof. However, the road that leads there is blocked. As Bentham famously put it: “in the map of science, the department of judicial evidence remains to this hour a perfect blank”. This “state of wilderness” has been guarded and preserved for good reasons. In almost every common law jurisdiction disputed allegations are not meant to be resolved by professional fact-finders applying juridical reasoning procedures. Fact-finders arrive at verdicts on the basis of unarticulated common-sense reasoning to the evidence in the respective case. This has been the influence of an Enlightenment value, i.e. the belief that people are hard-wired with efficient cognitive mechanisms which enable them to draw accurate conclusions. In the words of one of Enlightenment’s originators: Required is only “plain and ordinary good sense”, which is regarded as “a less fallacious guide than the knowledge of a judge accustomed to find guilty, and to reduce all things to an artificial system, borrowed from his studies”. Beccaria’s aphorism echoes through the court rooms to this day. “Happy the nation”, he states, “where the knowledge of the law is not a science!”. As Lord Justice Rose put it
“[j]urors evaluate evidence and reach a conclusion not by means of a formula, mathematical or otherwise, but by the joint application of their individual common sense and knowledge of the world to the evidence before them”.
It seems that the fact-finders’ inferential process is based on common sense and not on analytic inferential relations. The respective reasoning process is a black box, in which we cannot look. That is problematic especially concerning the jury, which has been famously described by Lord Devlin as something much more than “an instrument of justice”. As his Lordship remarked, the jury is “the lamp that shows that freedom lives”. Therefore, I suggest that we move into another gear, and focus on the concept of moral certainty – after all, the formula beyond a reasonable doubt is a development thereof – and its epistemological foundation, the Common Sense philosophy. This should allow us to understand the theoretical blueprint of the criminal adjudication in common law systems, on which the law of evidence and its main feature, i.e. the lack of accountability, is grounded.
2.2 Common sense
Common sense –as an argument or even as an excuse– is ubiquitous. Ranging from the daily over the political to the jurisprudential, common sense is extensively deployed especially when all other arguments have been exhausted. Generally, the term common sense refers to what people would agree upon. Yet the diversity of our multicultural modern world, which makes it impossible to find a lowest common denominator among our beliefs, is the least of our problems. For common sense is much more than a set of undisputed beliefs. It is an epistemology.
Notwithstanding its ancient heritage, common sense was a dominant concept in the philosophy of Scotland and England during the 18th century. Much of the theoretical literature on the philosophical concept of common sense traces back to the Scottish philosopher Thomas Reid, who formed this school of thought and developed an epistemological concept, which took common sense as the ground of justification for daily and philosophical knowledge. Reid argued that common sense and common sense-knowledge does not require proof, since is itself the source of all proof. Common sense was treated as an original source of knowledge inherent to all humans. It is:
“that power of the mind which perceives truth, or commands belief, not by progressive argumentation, but by an instantaneous, instinctive, and irresistible impulse; derived nor from education nor from habit, but from nature”.
According to that view conclusions somehow emerge from perceptual input and become visible/understandable in a mental theatre, but only from the first-person perspective. This feature turns common sense inferential reasoning to a black box, since we lack any intersubjective and shareable information regarding its form and structure. So there are not only legal limitations to our ability to extract justifications and explanations from the jury, but conceptual limitations as well. Knowledge-claims, which are based on common sense, are not, so the argument goes, in need of epistemic justification: common sense is itself the cognitive mechanism providing justification. It is supposed to be infallible, unsophisticated, and with results that are obvious: Common sense justifies itself by its epistemological credentials which make it intrinsic to human nature. This view explains how universality and cognitive inaccessibility have been regarded as perhaps the main features of that concept. For common sense would allow us to arrive at conclusions without needing or being even conscious of any intermediate steps. Rules of reasoning would thus be an unnecessary constraint and limitation on a mighty cognitive tool.
The constant effort of the 17th century philosophic community in England to build an intermediate level of knowledge –lying between absolute (mathematical) knowledge and mere opinion– found in common sense a suitable epistemological mechanism. The new epistemology could, philosophers argued, overcome Hume’s paralyzing skepticism and give to all people access to knowledge. This level of knowledge became widely known as moral certainty and its epistemological foundation was the common-sense philosophy. The underlying cognitive process is supposed to be inaccessible, since nothing but its results enters our awareness. Common sense is by design incompatible with deep reflection or inferential (logical) thinking. As Giambattista Vico remarks “[c]ommon sense is judgment without reflection, shared by an entire class, an entire nation, or the entire human race”. People thus make decisions based on common sense without having to or even being able to reflect upon the decision-making process. All in all, common sense was regarded as an intuitive mechanism, which is fast, effortless, and hardwired in human cognition, thus facilitating an effective decision-making process with minimal cognitive effort.
That made common sense an excellent candidate for a functional standard of proof in (criminal) adjudication. Furthermore, it is of vital importance to mention that English writing on evidence “drew heavily” on contemporary Scottish philosophy and its ouevre: common sense. There is no doubt today that “[t]raditional common law evidentiary principles and doctrines are built on the foundations of common sense inference”.
The American Judge Patricia Wald describes critically common sense as “some kind of amorphous community conscience”, which is supposed to simply jump to the mind of the ordinary man (jury) but cannot be described or explained further. For it is supposed to be a self-authorizing, directly apprehended sensation, “derived from sentiment not reason, and we attain moral knowledge by an immediate feeling and finer internal sense not by a chain of argument and induction”. As Beccaria remarked, “it is much easier to feel this moral certainty, than to define it exactly”. And since moral certainty and its underlying epistemology of common sense can only be felt as an immediate, private sensation, which cannot be taught, communicated or backed with reasons, we cannot or should not expect, let alone ask fact-finders to give an account of their respective decision. The idea that everyone is equipped with common sense facilitates, indeed necessitates, an oracular decision on behalf of the fact-finders, containing no more than two words: ‘Guilty’ or ‘not guilty’. Asking fact-finders, who are “steeped in common sense inferential reasoning”, to provide reasons for decisions could only result from a deep misunderstanding of the epistemological foundation of the common law system of criminal adjudication.
It cannot be stressed enough that common sense is deployed at the cost of defeating other values, especially accountability. The automatic cognitive process supposedly enabled by common sense thus serves as a justification for not imposing other mechanisms of accountability on fact finders. The criminal verdict is not backed by reasons as a result of the theory design of common sense justice, not due to some authoritarian secrecy policy. In that sense, moral certainty (later concretized as proof beyond reasonable doubt) is not a private entity, which is unshared as a matter of fact. It is one which is unshareable as a matter of principle. All it takes is critical self-examination, in order to detect this ‘state of the case’. This is –I think– one of the most crucial theoretical presuppositions of the epistemology underlying the modern criminal procedure, namely the fact that the psychological vocabulary such as “moral certainty”, “feeling sure” etc., stands for phenomena in a mental theatre, which are accessible only to the respective individual/fact-finder. To put with William James, proof is being regarded as a “bare phenomenal fact” of consciousness, accessible to us as “the IMMEDIATELY KNOWN”.
The image of the fact-finder as a black box is not a caricature but a theory-laden concept for the evaluation of legal evidence. According to this view, we can ask for nothing more than “a set of practical rules which experience has shown to be best fitted to elicit the truth”, as Lord Dunedin suggests. According to that view we simply have to accept that the fact-finder will make his or her decision based on an amorphous mass, whose propositional structure cannot be known to others. The secrecy of the fact-finders’ modus operandi conceals no conspiratorial plans or authoritarian political rationale. Unreasoned verdicts are simply the institutional bulwark of the underlying epistemological framework. If we want to criticize the legitimacy of an institution, we’d rather focus on its epistemological basis. For if we endeavor to attack the most basic epistemological assumption of the criminal adjudication, we have to show that common sense is not what it has been proclaimed to be: a bedrock assumption. As Shapiro unequivocally remarks: “Once the epistemological problem is solved, the institutional problem is solved”.
2.3 Is ‘proof’ in the black box?
We saw above that the fact-finder’s decision is based on common sense reasoning, i.e. on by and large subconscious behavior, rather than on analytic thought processes or patterns of inferential reasoning. Contrary to the system of proof of the Romano-canon inquisition process whereby professional judges operated with rigid evidentiary rules (e.g. two good witnesses or a confession amounted to a “full proof”), common-sense laden evidentiary systems in criminal adjudication adopt a laissez-faire approach to the assessment of evidence. The fact-finder is authorized to reach whatever conclusions he or she deems appropriate. Accordingly, fact-finders have to turn their attention inwards and attach the label ‘proof’ to what they find there. This can be done by associating a private sensation (“feeling sure”), which is supposedly known only through introspection and is accessible only to the fact-finder herself, on the one hand, and a public locution (criminal verdict: ‘Guilty’ or ‘Not Guilty’), which is known and accessible to everyone, on the other hand. However, the image, according to which proof has to be seen as a natural expression of an inner sensation, accessible only to the fact-finder, is intractable, indeed it is conceptually confused, for three reasons.
2.3.1 Lack of Normativity
It has already been pointed out in literature that extreme subjectivity is mainly a problem of measurement; for each fact-finder will deploy his or her own understanding by reference to her own inner sensations. Given that for measurement we need something to serve as a standard (of proof) –in order to determine by reference to it the unit of measurement qua canonical sample– the rule of law demands that a replicable quantum of proof is to be applied at every instantiation of the standard of proof qua general rule. It is deeply undesirable and deleterious for the coherence of a legal system to tolerate a practice whereby different fact-finders understand proof thresholds (standards of proof) in radically different ways. By doing that we do not apply the standard of proof in order to measure our subjective beliefs. On the contrary, we use the respective inner sensation in order to articulate the respective content of the standard of proof. With each measurement we alter the measuring instrument. Whereas we should let the standard of proof determine the measuring instrument, the opposite is the case. The standard of proof looks rather like play dough rather than a canonical sample. Yet, common sense philosophy manages to neutralize this argument by stressing the adhocness and the effectiveness of common sense cognition. Or maybe not?
2.3.2 Private standard of proof?
Now, the question is, whether the general discussion about private sensations such as ‘pain’ or ‘feeling sure’ is intelligible or, as I will try to show, convoluted and indeed self-contradictory. Does it make sense to utter inner sensations (e.g. ‘I am the only one who feels my pain’) by using a private language, i.e. a language whose terms are understandable only to the person using it? Wittgenstein unleashed his unparalleled attack against the private language argument by showing that our (psychological) vocabulary is not private and that this very notion is incoherent. He used the famous Beetle-example, whereby everyone has something in a box, which alas cannot be seen by others.
“– Suppose everyone had a box with something in it: we call it a ‘beetle’. No one can look into anyone else’s box, and everyone says he knows what a beetle is only by looking at his beetle.”
The main crux of Wittgenstein’s remarks on private language is to show that the meaningfulness of our expressions, including our allegedly private psychological expressions, is a function of public correctness criteria salient in the respective language game. Without any criteria of correctness there is also no way to draw a boundary between right and wrong application of a concept, i.e. between sense and nonsense. Admittedly, pointing out that ‘pain’ or ‘feeling sure’ is something that is inside us, i.e. inside a private quasi-spatial realm of sensations which is inaccessible to others, is true but misses its target. The phenomenon of pain or feeling sure is inside us, but the concept which articulates the respective sensation qua phenomenal appearance and enables us to perceive it as a distinct experience is not; concepts are public and thus open to scrutiny. As McGinn remarks, when one uses S whilst we do not have a linguistic technique of using S, then there is “nothing that can count as a correct use of S in the future.
The argument outlined above is tailor-made for our present discussion. No fact-finder is supposedly able to look into the respective black box of the other fact-finders; they can only speculate about the content of “feeling sure” or “reasonable doubt” by looking at their own beetle. i.e. the feeling of being sure inside their own box. The situation, where each one looks into their own box and utters ‘proof’ raises the question of how the different utterances, if at all, connect and relate to each other. As far as the meaning of the word proof or reasonable doubt is concerned, our current practice “cancels out”, whatever its propositional content is. However, Wittgenstein argues, this is not how we learn and understand language in general or any word in particular, express ourselves and communicate. If we cannot learn anything about supposedly private sensations, then we cannot distinguish between right and wrong uses thereof. At some point the question (e.g. as part of a discussion with oneself) will perforce arise, whether the subject has identified the correct “beetle”, the correct “proof” as “IT”.
The very fact that private sensations are not shareable means that the rules for their detection and ascertainment cannot be applied, or even taught, communicated and understood – not only from others, but most importantly from the fact-finder him or herself. Lacking criteria of correctness for our own allegedly private sensations means also that there is no not-self-referring sample to justify the use of that sensation by linking its use with public criteria of application. In other words, common sense cognition and the claim that someone has proved something to his or her own satisfaction is an invitation to self-deception. The moment we understand language and meaning as a rule-governed activity rather than an anything-goes one, we have to concede that
“[t]he thing in the box has no place in the language-game at all; not even as a something: for the box might even be empty.”
It is unintelligible, Wittgenstein contends, that the words of a “private language” can only be known to the person speaking. Actually these words cannot be known at all. They are nonsensical, since they lack grammatical rules which would fix and stabilize their meaning. Each time a fact-finder utters “proof” or “guilty” as a verbal expression of his or her private sensation, there is nothing that can serve as a sample –what criteria or values is that justification or explanation to be measured against?– for future or present use, not because private sensations are incommensurable, but because they are meaningless. As McGinn observes, “our ordinary language-game can be taught, learnt and participated in whatever this object is”.
We should not confuse the uncertainty about somebody’s inner sensation or psychological state on the one hand and the respective meaning of these concepts on the other hand. In other words, the uncertainty that surrounds our ascription of a sensation to others (e.g. how can we exclude, that A was only pretending being in pain or that A was not really sure about D’s guilt) doesn’t map on the criteria for their correct application, i.e. on the logico-grammatical features of a concept, dictating its right or wrong use.
2.3.3 Is BRD a ‘beetle’?
Pointing from a philosophical point of view at the instability of the semantics for terms such as “feeling sure” or “proof beyond a reasonable doubt” comes as no surprise to the legal community. A rapidly growing body of literature has already shown that the rate of comprehension for basic evidentiary concepts among –more or less educated– jurors is “relatively low”. In fact, empirical studies testing aspects of the ability of fact-finders to assess evidence, confirm that our attempt to provide them with instructions does not promote understanding of evidentiary concepts. Let me make clear that I do not attempt to adduce empirical evidence in order to corroborate a conceptual approach. Yet, the insight that the semantics of “proof beyond a reasonable doubt” qua standard of proof look like an amorphous mass rather than a set of (shareable) rules, which would dictate its right and wrong application, can explain why instructions given by judges and judicial summing-up of trial evidence – crucial as they may be – fail to improve comprehension of fundamental procedural devices. The lacking of rules for the correct use of these procedural devices leads to instructions that are, ultimately, gibberish. The accuracy of criminal verdicts, i.e. their truth-conducive character, is regarding the standard of proof a function of sheer luck.
The utterance “feeling sure” or “proof beyond a reasonable doubt” is due to complete lack of explicit rules, which could guide its proper use, similar to the utterance “I feel in my hand that the water is three feet under the ground”. Undeniably, it is possible that we could find water, but it’s not the hand that led us there. Similarly, we do not know how the grammar of “feeling sure” or “proof beyond a reasonable doubt” relates to the grammar of “proof”, since no one has explained, what sure means in this particular context. For we lack “a great deal of stage- setting in the language” which must be presupposed, if the simple speech-act of uttering “guilty”, “proof” or “sure” is to make sense as a criterion of rectitude which is not self-referent, to wit, which is a result of inferential reasoning based on a detailed structure of justification. In the absence of an infallible epistemic mechanism or due to lacking a hotline to a supernatural omniscient creature informing us about the correspondence of the criminal verdict to the real facts, our confidence in the institution responsible for determining issues of guilt and innocence is a function of the epistemic performance of the respective fact-finder / decision maker. Truth is a matter of justification and accountability – not the other way around. It is thus from a legal point of view imperative that we articulate the content of the respective standard of proof. In order to fix the semantics of the latter we need more than mere linguistic refinements or “changes in the wording and presentation of judicial instructions”. We need a microanalysis of inferential patterns, which enable fact-finders to reach warranted and truth-conducive verdicts. Otherwise jury instructions are “like foreign movies without subtitles”. Judicial instructions and especially standards of proof applied by fact-finders are a core element of criminal justice. By failing to provide fact-finders with the propositional content of proof we voluntarily cause fact-finders to display poor levels of comprehension of these instructions.
3. Moving forward
3.1 Turn to Epistemology
The epistemological scaffold on which the current system of criminal adjudication in common law countries relies, is both from an epistemological and a practical point of view unsustainable. Fact-finders have no non-circular criterion of correctness for their verdict. Unless we accept that something can be epistemically privileged and justified by itself –only divine entities could ever be regarded as self-referring and self-explanatory– we are in desperate need of public standards for decisions, which are open to scrutiny. The legitimacy of the current system is questionable, since its theoretical blueprint contradicts one of its central tenets: its Rationalist Tradition and the requirement that all decisions, which affect the interests of individuals by resolving disputed questions of fact, are justifiable. The crisis of the system of criminal justice as a common-sense laden system of evidence evaluation can be best described through the Humpty-Dumpty effect: Its pieces can never be rejoined.
The impossibility to explicate the standards of proof hinges on the fact that every effort to explain it is another step in “muddy waters”. Jury instructions only paraphrase words, albeit they do not lay down rules for their use. They provide instructions on what is to be done but not on how. No one has yet managed to instruct fact-finders upon the propositional content of the beyond reasonable doubt standard, not because the concept is a deep, impenetrable mystery or the task intractable, but simply because it lacks grammar, i.e. rules governing its use. The challenge we nowadays face is to find a way and articulate this set of rules without falling back to legal formalism/linguistic determinism. Otherwise fact-finders will continue to fly blind by trying to detect an inner sensation that is supposed to be self-explanatory. Evidence does not speak up and say “I am adequate proof of guilt”. Unless we provide fact-finders with logico-grammatical rules, governing the use of “feeling sure” or its synonym “proof beyond a reasonable doubt”, to wit, unless we fix the semantics of these words by providing a structure of justification, certainty will remain an elusive phenomenon which obeys no criteria of correctness. Our grammatical investigation exposed the emptiness of our picture of “certainty” and “proof” as common-sense phenomena. And this emptiness poses a serious threat for the legitimacy of criminal adjudication which by design has to be truth conducive.
The system in its present form has reached its limits in an increasingly complex world. No matter how we approach and answer the Shakespearean question of whether ‘to define or not to define the standard of proof’, it still remains a fact that for the last two hundred years we have been fighting against philosophical shadows, i.e. the assumption that common sense would enable us to reach automatic conclusions allegedly jumping into our consciousness. After all, the Strasbourg Court regarded the ability to understand the verdict as “a vital safeguard against arbitrariness”. True, the Grand Chamber made clear that lay juries have no epistemic duties, i.e. no obligation to give reasons for criminal verdicts in order to comply with Art 6 of the Convention. But to reiterate this, misses its target. Giving reasons for decisions is one thing. Having reasons for a decision is quite another. Common sense justice fails, I think, to meet e.g. the Convention’s test of “fair trial” (Art 6(1)). For Article 6 requires an assessment of whether sufficient safeguards have been in place to avoid any risk of arbitrariness and to enable the accused to understand the reasons for his or her conviction. In other words, failing to give reasons will not render the accompanying verdict vulnerable to a quashing appeal or even unlawful. On the contrary, however, failing to have reasons for the respective decision maps on the criterion of arbitrariness set by the Strasbourg Court. Cryptic decisions based on inarticulate common-sense inferences or appearances into one’s consciousness are a paradigmatic case of arbitrary decision-making.
All this makes a clear case for a paradigm shift. But we still have no model for the transition. On the one hand, rigid rules lead either to paralysis or to the practical necessity of extracting confessions. Probabilistic models, on the other hand, give expression to a very old dream of humanity: to eliminate the need for a decision. It is an unreal dream. We simply cannot treat the fact-finder as la bouche de l’algorithm. We need a fresh start: to launch a debate on the structure of justification behind the evidence. We have to turn once again to epistemology which might yield –as it has done in the past– a firm theoretical foundation (structure of justification) on which we can build a resilient evidence assessment.
3.2 Epistemology. Which epistemology?
The aspiration to an epistemological turn is insofar an empty shell, as it is not clearly stated what epistemology is on the first place, and what implications such a move would have for our law of evidence. The term ‘epistemology’ has nowadays come to denote a bewildering number of (very) different approaches. Clearly, due to text-consistency and page-limit constraints such questions go beyond the scope of this article. The discussion about what should come next as heir to the throne can only be deferred to a distinct paper facilitating analysis of the micro-structure of the inferential steps toward ascribing criminal liability. What I would I like to do in this final part, is briefly spell out some desiderata, which express not only personal opinion and preferences but constraints, brought forward by eminent Evidence scholars.
Structure of Law
In order to replace an epistemology, we need more than a promising theoretical framework. The reconceptualization of basic evidentiary concepts of Evidence & Proof, justification and forensic knowledge has to take place during business-as-usual operation. As Stephen Toulmin has pointed out the very fact that some routines for validating knowledge claims “have established themselves in practice may have to be enough for us”. In other words, what is needed, is epistemological explication, not epistemological theory. For the latter comprises necessarily normative claims, which would interfere with the internal operations of the already established routines mentioned above and usurps the autonomy of a (structurally closed) system of criminal adjudication. Legal epistemologists who unequivocally state that “[e]pistemology, like philosophy generally, is essentially universal”, imply that epistemological principles are a priori. However, the abovementioned philosophers need to realize that they are not bringing forward a claim about a global epistemology versus a local law of evidence. The disagreement, if any, is not between disciplines –this would be terribly over-simplistic– but rather between two distinct forms of thought, i.e. between universal and pragmatist epistemology.
The epistemological model needs to do justice to a simple constraint: the evaluation of criminal evidence is a decision-making process under uncertainty. In criminal adjudication, fact-finders are not engaging in an academic exercise, reflecting on the abstract meaning of knowledge or truth. Nor is there any realistic way to assess whether a justified belief corresponds with an objective reality.
Factual rectitude is not the only goal
The evaluation of evidence does not take place in a truth-searching procedural vacuum. The epistemological model must be able to integrate other procedural objectives as well. The search for truth is simply not “the main business” of the courts, and sometimes the “actual truth” is “simply irrelevant”. The respect for the confidentiality of certain communication or materials and the exclusion of probative evidence obtained by illegal means are, as Stein remarks, “just two representative examples of evidentiary rules thwarting fact-finding for the sake of other objectives and values”.
 JB Thayer, A preliminary treatise on evidence at the common law (Boston: Little, Brown, 1898), 85, pp. 47–182.
 See Lord Justice Auld’s Review: 2001, London, H.M.S.O. The report will hereafter be referred to as “the Review"; J Horder, Ashworth’s Principles of Criminal Law, 8th ed. (Oxford: Oxford Univ. Press, 2016), p. 10, remarks that this “simplistic picture can lead to a distorted perception of the way in which the criminal justice system works in practice”.
 P Roberts, Does Article 6 of the European Convention on Human Rights Require Reasoned Verdicts in Criminal Trials?, in: 21 International Review of Victimology (2015), pp. 139–160 (229); W.R. Cornish, The Jury (Penguin Books, 1971), p. 10.
 According to B Shapiro, ‘Beyond Reasonable Doubt’ and ‘Probable Cause’ (Univ. of Cal. Press, 1991) p. 2, these are the two kinds of rules of evidence in common-law jurisdictions; cf. Thayer, Treatise on Evidence (n. 1), p. 83.
 Cf. s8 Contempt of Court Act 1981.
 On January 13, 2009, the ECtHR ruled (in: Taxquet v. Belgium Taxquet v. Belgium, App. No. 926/05, (Eur. Ct. H.R., Jan. 13, 2009) that the conviction of the applicant from a Belgian criminal court violated the defendant’s rights (Art 6(1) ECHR) because the jury did not give reasons for its verdict. On November 16, 2010, the Grand Chamber of the ECtHR (en banc court) affirmed the decision but did not put the legitimacy of jury trials into doubt. Taxquet v. Belgium (GC), App. No. 926/05, (Eur. Ct. H.R., Nov. 16, 2010), para 90 (hereinafter: Taxquet (GC)).
 See Roberts, Article 6 (n. 3), p. 217.
 JD Jackson, Making Juries Accountable, in: 50 The American Journal of Comparative Law (2002), pp. 477–530 (479), talks of “increasing signs that this lack of accountability on the part of juries is coming under strain”.
 See N. Bamforth and P. Leyland (eds.), Accountability in the Contemporary Constitution (Oxford: Oxford Univ. Press, 2013); see also JS Lerner and PE Tetlock, Accounting for the Effects of Accountability, in: 125 Psychological Bulletin (1999), pp. 255–275.
 Different societies across the political spectrum, across history and space incorporate the element of accountability in their political grammar; see e.g. JS Lerner and PE Tetlock, ‘Bridging individual, interpersonal, and institutional approaches to judgment and choice, in: S. Schneider and J. Shanteau (eds.), Emerging perspectives in judgment (Cambridge: Cambridge University Press, 2002), pp. 431–457 (431–432).
 Plato, Republic 534b-c.
 In Aischylus’ trilogy of tragedies ‘Oresteia’ (see The Oresteia. A Trilogy Consisting of the tragedies: Agamemnon, The Libation Bearers, The Eumenides, by Aeschylus; transl. by Ian Johnston), when Erinyes close in on Orestis by smelling the blood of his slain mother, the Goddess Athena intervenes and brings in twelve Athenians to join her in forming a jury to judge her supplicant. Justice is thus decided by a jury representing the citizen body and its values; see Eumenides, Episode 4, para 10–11 “You citizens of Athens, you judges at the first trial ever held for murder, hear what I decree. Now and forever this court of judges will convene here to serve Aegeus’ people […] So here I now establish this tribunal, incorruptible, magnificent, swift in punishment — it stands above you, your country’s guardian as you sleep.”
 For the full transcript of his speech see https://www.jfklibrary.org/Research/Research-Aids/JFK-Speeches/American-Newspaper-Publishers-Association_19610427.aspx.
 The term accountability is being treated, as far as I can see, as synonymous and partially co-extensive with the terms ‘responsibility’, ‘answerability’ and ‘responsiveness’; for more discussion see D Oliver, Government in the United Kingdom – The search for accountability (Open Univ. Press 1991); Jackson, Making Juries Accountable (n. 8), p. 483, gives the following definition: “Accountability may be defined as the duty of a public decision maker to explain, legitimate and justify a decision and to make amends where a decision causes injustice and harm”.
 DM Kahan/DA Hoffman/D Braman, Whose Eyes Are You Going to Believe? Scott v. Harris and the Perils of Cognitive Illiberalism, in: 122 Harvard Law Review (2009), pp. 838–906 (902), for a similar argument.
 Jackson, Making Juries Accountable (n. 8), p. 488.
 J Bentham, The Works of Jeremy Bentham vol. 4, 1843, p. 316.
 It can only be seen as a political scandal of our time that TTIP (Transatlantic Trade and Investment Partnership) negotiations between the EU and US authorities remain secret until after the negotiations are completed, although the affected areas include subjects of vital importance such as health and food safety.
 See S Doran, Trial by Jury, in: Mike McConville and Geoffrey Wilson (eds.), The handbook of the criminal justice process (Oxford, Oxford University Press, 2002), p. 380.
 F Schauer, Giving Reasons (1995), in: 47 Stanford Law Review, pp. 633–659 (634).
 P Roberts, ‘Introduction’, in: Roberts (ed.), Theoretical Foundations of Criminal Trial Procedure, 2014, p. xxv.
 C Roxin, B Schünemann, Strafverfahrensrecht, 27. ed., Munich 2012, § 2, para 1.
 See A Goldman, The Sciences and Epistemology, in: PK Moser (ed.), Oxford Handbook of Epistemology, Published to Oxford Scholarship Online: November 2003, 147; Thayer, Treatise on Evidence (n. 1), 270, calls our Law of Evidence “emphatically, a rational system”. T Anderson, D Schum, W Twining, Analysis of Evidence, 2nd ed. (Cambridge: Cambridge University Press, 2005), 78, remind us that the assumption “upon which evidence scholars have based their work for two centuries is ‘the Rationalist Tradition”.
 See Lerner/Tetlock, Accountability (n. 9), pp. 262–63 for more discussion.
 Jackson, Making Juries Accountable (n. 8), p. 484.
 J Jackson, Unbecoming Jurors and Unreasoned Verdicts: Realising Integrity in the Jury Room, in: J. Hunter/ P. Roberts/P. Young/N.M. Simon/D. Dixon (eds.), The Integrity of Criminal Process: From Theory to Practice (Hart Publishing, 2016), pp. 281–307: “There is a difference between an accused understanding what evidence has been accepted in order to lead to his or her conviction, why that evidence has been accepted and how the decision was actually reached.”
 Lord Woolf, Lord Chief Justice of England, The Times, January 30, 2001.
 M Williams, Scepticism without Theory, in: 41 The Review of Metaphysics (1988), pp. 547–588 (570).
 Lerner/Tetlock, Accountability (n. 9), p. 257.
 GH Mead, The Self and the Organism, Section 18, in: CW Morris (ed.), in Mind Self and Society from the Standpoint of a Social Behaviorist (Chicago: University of Chicago, 1934), pp. 135–144 (141).
 This argument goes back to Pyrrhonism and the Hellenistic Epistemology, see Sextus Empiricus, Outlines of Pyrrhonism 1.90.
 Thayer, Treatise on Evidence (n. 1), p. 198.
 The Review (n. 2), para 1.
 Jackson, Making Juries Accountable (n. 8), p. 486.
 Thayer, Treatise on Evidence (n. 1), p. 269.
 H.L.A. Hart, The Concept of Law, 2nd ed., 1994, Clarendon Press: Oxford, p. 121.
 Schauer, Giving Reasons (n. 20), p. 648.
 It is possible and highly likely that different fact-finders are acting in accordance with a rule (standard of proof). However, it cannot be meaningfully asserted that these fact-finders are following a rule, unless each and every one of them makes their reasons public.
 Note that we should not confuse simplicity with superficiality.
 I will treat here – again, at the cost of oversimplification – the main proponents of legal realism, ie the American and the Scandinavian school of legal realism as a unity. See e.g. JC. Hutcheson, Jr., The Judgment Intuitive: The Function of the ‘Hunch’ in Judicial Decision, 14 Cornell Law Quarterly (1929), p. 274, 285. For an introduction to legal realism see R Wacks, Understanding Jurisprudence, 4th ed. (Oxford, 2015), ch. 6, pp. 162–189; See also W Twining, K Llewellyn and the Realist Movement, (London: Weidenfeld & Nicholson, 1973).
 According to K Llewellyn, A Realistic jurisprudence-The Next Step, 30 Columbia Law Review (1930), p. 431, we should focus on “the area of contact, of interaction, between official regulatory behavior and the behavior of those affecting or affected by official regulatory behaviour”; as Wacks, Understanding Jurisprudence (n. 40), p. 173, remarks, “though they [Legal Realists] accept, along with the positivists, the need for a scientific analysis of law, the realists reject the single avenue of logic and seek to apply the numerous avenues of scientific enquiry, including sociology and psychology”.
 OW Holmes, Jr., The Path of the Law, 10 Harvard Law Review (1897), pp. 460-61; see also J Frank, Law and the Modern Mind (New York: Brentano's, 1930); KN Llewellyn, The Bramble Bush (New York: Oceana, 1930).
 See Hutcheson, The Judgment Intuitive (n. 40), pp. 274, 285.
 S Danzigera et al, Extraneous factors in judicial decisions, 108(17) Proceedings of the National Academy of Science of the USA, pp. 6889–6892; the researchers found that the percentage of favourable rulings drops gradually from ca. 65% to almost zero within each decision session and returns rapidly to ca. 65% after a break.
 In fact, Judge PM Wald, Guilty Beyond a Reasonable Doubt: A Norm Gives Way to the Numbers, University of Chicago Legal Forum (1993), pp. 111–126 (111), reports that “few judges […] believe that every jury plausibly and precisely follow the beyond a reasonable doubt standard in deciding guilt or innocence”.
 Hart, The Concept of Law (n. 36), pp. 140–41, makes a similar pressure point. It is possible, he says, “that, in a given society judges might always first reach their decisions intuitively or by ‘hunch’ and then merely choose from a catalogue of legal rules one which, they pretended, resembled the case in hand; they might then claim that this was the rule which they regarded as requiring their decision, although nothing else in their actions or words suggested that they regarded it as a rule binding on them.”
 Lerner/Tetlock, Accountability (n. 9), p. 263.
 Lerner/Tetlock, Accountability (n. 9), pp. 440–42.
 Lerner/Tetlock, Accountability (n. 9), pp. 440–42.
 Jackson, Making Juries Accountable (n. 8), p. 481 remarks that “[t]he difficulty with all these measures is that they cannot guarantee that a jury has not acted with bias during the course of its deliberations”.
 J Bentham, The Works of Jeremy Bentham, Part XI, 1843, p. 209.
 C Beccaria, An Essay on Crime and Punishment, ch. XIV, p. 22; online available at: www.thefederalistpapers.org.
 R v Adams  2Cr App R 467, 481.
 See also JO Newman, Madison Lecture: Beyond Reasonable Doubt, 68 New York University Law Review (1993), p. 979 (1002), for a similar argument.
 Sir Patrick Devlin, Trial by Jury (London: Stevens & Sons Ltd, 1966), p. 164–65. My emphasis.
 Shapiro, Beyond Reasonable Doubt (n. 4), p. 21, states unequivocally that “term ‘moral certainty’ was taken to mean proof beyond reasonable doubt.”
 Thayer, Treatise on Evidence (n. 1), p. 4, emphasizes that the rules of Evidence “originate in the instinctive suggestions of good sense, legal experience, and a sound practical understanding”. He then (p. 281) states that “it is common sense, not Common Law, that presently functions as its [fact-finding] principal guide”; similarly, among others Paul Roberts and Colin Aitken, The Logic of Forensic Proof: Inferential Reasoning in Criminal Evidence and Forensic Science Guidance for Judges, Lawyers, Forensic Scientists and Expert Witnesses, Royal Statistical Society, p. 15.
 R. J. Allen, Common Sense, Rationality, and the Legal Process, 22 Cardozo Law Review (2000-2001), pp. 1417–31, remarks that the phrases “common sense”, “commonsensical”, and “sensible” used as an argument appear “upwards of 70,000 times in Westlaw”, which makes ‘common sense’ them the most cited authority.
 See the relevant entry at the New World Encyclopaedia, online available at: http://www.newworldencyclopedia.org/entry/Philosophy_of_Common_Sense.
 It is well documented that the concept became immediately through doctrinal migration very famous in France and the rest of continental Europe; see Shapiro, Beyond Reasonable Doubt (n. 4), p. 27, stresses that the Scottish common-sense school “would dominate Anglo-American epistemological discussion and moral philosophy in the late eighteenth and early nineteenth centuries. It also had a major impact on evidence scholars: Judges had little choice but to borrow from the epistemology that could be drawn from religious doctrine and philosophy”.
 See the relevant entry at the Stanford Encyclopedia of Philosophy, online available at: http://plato.stanford.edu/entries/reid/. See also Shapiro, Beyond Reasonable Doubt (n. 4), p. 27-–28 for more discussion.
 Thomas Reid, Works, ed. 1863, p. 922.
 James Beattie, An Essay on the Nature and Immutability of Truth, 1771, p. 40.
 See Paul Roberts and Colin Aitken, The Logic of Forensic Proof: Inferential Reasoning in Criminal Evidence and Forensic Science, Royal Statistical Society, p. 25–26.
 Robert Nehring, Kritik des Common Sense, Berlin 2010, pp. 22–25.
 See only Shapiro, Beyond Reasonable Doubt (n. 4), p. 7.
 V. Giambattista, The New Science of Giambattista Vico, 1st Book, para 142, (transl. by T. G. Bergin / M. H. Fisch, Cornell Univ. Press 1948).
 Shapiro, Beyond Reasonable Doubt (n. 4), p 124.
 Roberts and Aitken, The Logic of Forensic Proof (n. 66), p. 15.
 Wald, Reasonable Doubt (n. 45), p. 110.
 David Hume, An enquiry concerning the principle of morals, 1777, p. 2.
 Beccaria, Essay (n. 53), Chapter XIV p. 22.
 Roberts and Aitken, The Logic of Forensic Proof (n. 66), p. 24.
 Similarly: Lord Devlin, Trial by Jury (n. 57), 13.
 W. James The Principles of Psychology, 1890, 182.
 Thompson v R  AC 221, 226.
 See Robert C. Power, Reasonable and Other Doubts: The Problem of Jury Instructions, 67 Tennessee Law Review (1999), pp. 45–123 (122) et passim for more discussion.
 Shapiro, Beyond Reasonable Doubt (n. 4), p. 46.
 Shapiro, Beyond Reasonable Doubt (n. 4), p. 3, describes the judge in criminal cases as an “accountant, who totalled the proof fractions”.
 See also L. Laudan, Truth, error, and criminal law: an essay in legal epistemology, 2006, p. 5; Shapiro, Beyond Reasonable Doubt (n. 4), p. 28 remarks that “the Common Sense school was especially anxious to counter the scepticism of Hume, which insisted that no amount of evidence could remove doubt.” In fact, they went to the other extreme, since they made it too easy to raise a knowledge claim.
 ‘Proof’ is a verbal description of an inner state / feeling.
 See e.g. The Crown Court Compendium. Part I: Jury and Trial Management and Summing Up”, May 2016, Ch. 5: “If reference has been made to “beyond reasonable doubt” by any advocate, the following may be added: You have heard reference to the phrase ‘beyond reasonable doubt’. This means the same as being sure”.
 See G. P. Baker and P.M.S. Hacker, Wittgenstein: Understanding and Meaning: Volume 1 of an Analytical Commentary on the Philosophical Investigations, Part I – Essays, p. 190–192: “The measure has a normative rule. It has a uniform use as a standard of comparison for other objects. It is always what measures, never what is measured.”
 Remember that we have already encountered the argument from generality, according to which a private standard of proof lacks normativity because it cannot be applied to other cases. Reasoning has to do with classes of pieces of evidence and classes of witnesses, generalizing the unique character of each individual case.
 L. Wittgenstein, Philosophical Investigations, tr. by G. E. M. Anscombe, 1958, § 293.
 Marie McGinn, Routledge philosophy guidebook to Wittgenstein and the Philosophical investigations, London: Routledge (1997), pp. 155–56.
 McGinn, Wittgenstein (n. 88), p. 158.
 I do not wish to suggest that fact-finders understand necessarily something radically different under ‘feeling sure’ or ‘BRD’. The problem is that we cannot rule out the possibility that the fact-finder arbitrarily associates public concepts with private sensations, because she does not understand the meaning of the standard of proof.
 Wittgenstein, Philosophical Investigation (n. 87), § 293.
 See McGinn, Wittgenstein (n. 88), p. 187.
 McGinn, Wittgenstein (n. 88), p. 155.
 Wittgenstein, Philosophical Investigations (n. 87), § 243.
 See McGinn, Wittgenstein (n. 88), p. 149.
 McGinn, Wittgenstein (n. 88), p. 162.
 McGinn, Wittgenstein (n. 88), p. 147.
 G. P. Kramer and D. M. Koenig, Do Jurors Understand Criminal Jury Instructions? Analyzing the Results of the Michigan Juror Comprehension Project, in: 23 Univ. of Michigan Journal of Law Reform (1990) pp. 401–37 (429); for a brief history of jury research see DJ Devine, Jury Decision Making. The State of the Science, 2012, pp. 14–20; see also F. Picinali, The threshold lies in the method: Instructing jurors about reasoning beyond reasonable doubt, in: 19 International Journal of Evidence & Proof (2015), pp. 139–53 (140).
 See Mitchell J. Frank and Dawn Broschard, The silent criminal defendant and the presumption of innocence, in: 10 Lewis & Clark Law Review (2006), p. 283, with more literature; see also Newman, Beyond Reasonable Doubt (n. 56), p. 984: “it rather unsettling that we are using a formulation that we believe will become less clear the more we explain it.”
 Kramer/Koenig, Criminal Jury Instructions (n. 98), p. 415.
 L Wittgenstein, The Blue and Brown Books: Preliminary Studies for the 'Philosophical Investigations' (Wiley-Blackwell; 2 ed., 1991), p. 9.
 Frank/ Broschard, Silent Criminal Defendant (n. 99), p. 283, make a similar pressure point: “Jurors cannot, except by accident, follow their instructions unless they first understand them.”
 Wittgenstein, Philosophical Investigations (n. 87), § 257.
 Similarly: Power, Reasonable and Other Doubts (n. 79), p. 113.
 This suggestion has been made eg from Kramer/Koenig, Criminal Jury Instructions (n. 98), pp. 401–37 (402).
 Quoted from Judge Wald, Reasonable Doubt (n. 45), p. 111.
 Devine, Jury Decision Making (n. 98), 55, describes this poor level of comprehension as “one of the strongest and most robust findings in the empirical literature on juries, with study after study yielding the same troubling conclusion across a variety of methodologies”.
 R. Allen, Common Sense, Rationality and the Legal Process, in M. McCrimmon/P. Tillers, The Dynamics of Judicial Proof, (New York: Physica-Verl., 2002), pp. 43–54 (45): “A funny thing happened to common sense on the way to modernity [...] it was waylaid by the Enlightenment. To the thinkers of the Enlightenment, the common understanding of the mass of humanity was not so much the cure as the problem, filled as it was with prejudice, ignorance, and beliefs inculcated by the powerful for their own benefit.”
 The Rationalist Tradition was first described in W. Twining, The Rationalist Tradition of Legal Scholarship, in: E. Campbell and L. Waller (eds.), Well and Truly Tried: Essays on Evidence in Honour of Sir Richard Eggleston (Law Books: Sydney, 1982) pp. 211–49.
 Cf. Frank/Broschard, Silent criminal defendant (n. 99), p. 282.
 United States v. Glass, 846 F.2d 386, 387 (7th Cir. 1988).
 As a result of that we outsource the analysis of the SoP from the realm of law to that of cognitive psychology. Quite naturally, our main focus shifted to the actual psychological mechanisms that govern the evaluation of evidence.
 The same development took place when the world consisted no more of “tiny, intensely interdependent agricultural communities”, where “jurors were drawn from the neighbourhood of the contested events”. As a result, the medieval system of self-informing juries declined; see J. Langbein, The Historical Foundations of the Law of Evidence: A View from the Ryder Sources, in 96 Columbia Law Review (1996), pp. 1168–1202 (1170).
 See Thomas V. Mulrine, Reasonable Doubt: How in the World Is It Defined?, in: 12 American University International Law Review (1997), pp. 195–225 (198): “In spite of the international judicial attention, however, the ‘reasonable doubt’ concept appears no less elusive today than it was in 1880”.
 The Grand Chamber (Taxquet (GC) para 90); Power, Reasonable and Other Doubts (n. 79), p. 115 makes a similar pressure point: “There is a severe flaw in the black box approach, at least in criminal cases. If defendants only were entitled to a jury trial, then perhaps this would be acceptable [...] However, criminal defendants also have rights to the reasonable doubt standard”.
 Taxquet (GC), para 90: “[t]he Convention does not require jurors to give reasons for their decision”.
 See only J. H. Langbein, Torture and Plea Bargaining, 46 The University of Chicago Law Review (1978), pp. 3–22.
 For a detailed discussion see K.N. Kotsoglou, Forensische Erkenntnistheorie (Berlin: Duncker & Humblot, 2015).
 See K.N. Kotsoglou, Proof Beyond a Context-Relevant Doubt. A Structural Analysis of the Standard of Proof in Criminal Adjudication. In: M. Di Bello / B. Verheij (eds.), Artificial Intelligence & Law, Special Issue: Evidence and Decision in Law (in press).
 Stephen E. Toulmin, The Uses of Argument, Cambridge University Press, 1958, p. 237.
 Susan Haack, Epistemology Legalized: Or, Truth, Justice, and the American Way, in: 49 American Journal of Jurisprudence (2004), pp. 43-61 (49).
 JB Thayer, A preliminary treatise on evidence at the common law (Boston: Little, Brown, 1898), p. 271; Richard A. Posner, Are There Right Answers to Legal Questions, in: Posner (ed.), The Problems of Jurisprudence (Harvard University Press, 1993), p. 205.
 Alex Stein, The Refoundation of Evidence Law, in: 9 Canadian Journal of Law and Jurisprudence (1996), p. 282