Quaestio facti. Revista Internacional sobre Razonamiento Probatorio
Quaestio facti. International Journal on Evidential Legal Reasoning
Sección: Conjeturas y refutaciones
2023 | 5 pp. 189-199
Madrid, 2023
DOI: 10.33115/udg_bib/qf.i5.22844
Marcial Pons Ediciones Jurídicas y Sociales
© Thomas Weigend
ISSN: 2604-6202
Recibido: 28/10/2022 | Aceptado: 21/04/2023 | Publicado online: 16/06/2023
Editado bajo licencia Reconocimiento 4.0 Internacional de Creative Commons

Accuracy and fairness—Siamese twins? A Comment on Sarah Summers «Epistemic Ambitions of the Criminal Trial: Truth, Proof, and Rights»

Thomas Weigend

Retired Professor of Criminal Law and Criminal Procedure
University of Cologne (Germany)
thomas.weigend@uni-koeln.de

ABSTRACT: The author agrees with Sarah Summers’ view that a criminal judgment in a system of fair justice should be substantively accurate and at the same time should be the result of a fair process. The author argues, however, for keeping these two requirements apart because they are linked to different goals and may in individual cases conflict with each other.

KEYWORDS: accuracy, criminal process, fair trial, presumption of innocence, truth.

SUMMARY: 1. SARAH SUMMERS’ «THICK» CONCEPT OF TRUTH IN CRIMINAL ADJUDICATION.— 2. ACCURACY.— 3. FAIRNESS.— 4. ARGUING FOR A THINNER CONCEPT OF TRUTH.— BIBLIOGRAPHY.

1. Sarah Summers’ «thick» concept of truth in criminal adjudication

In her excellent essay, Sarah Summers (2023) presents a novel, comprehensive concept of «truth» in the criminal process. Toward the end of the article, she summarizes her view in these words:

The aim of criminal adjudication in the rule of law should be understood in terms of enabling the establishment of a distinct kind of truth, which is to be determined in line with the substantive and procedural requirements governing justified punishment and fair process. […] The fairness of the process is of intrinsic, non–instrumental value and cannot properly be characterised as a side constraint or as serving only to ensure accurate application of the rules but must instead be seen as integral to criminal adjudication (p. 267).

Consequently, «true belief in legal adjudication implies […] commitment to substantive as well as procedural guarantees and in particular to a normative standard of proof» (p. 264). Summers therefore rejects the widespread view that the rules of evidence in the criminal process are mere instruments for producing «substantive» truth—which in this comment I will refer to as the «accuracy» of the court’s finding of facts. Adherence to the rules of a fair process, in Summers’ view, is part and parcel of achieving a correct outcome of the process; there can be no true result of a flawed process.

As I will explain in the last part of this comment, I agree with Summers in the result that a criminal judgment can be deemed «right» only if the findings of the court on guilt or innocence are accurate (reflecting the facts of the case) and have been generated in fair proceedings. I also agree with her point that evidence law is not merely instrumental toward finding the «truth» but is rooted in specific value judgments on how a fair process is to be conducted in a State recognizing human rights. But I would prefer to keep apart the two prongs of Summers’ «thick» concept of truth. I do so not only for the sake of clarity but also because I think that there exists a certain tension between the search for an accurate result and procedural fairness—a tension which would be obscured by treating all preconditions of a just judgment as aspects of one comprehensive concept of legal truth.

Before expanding on this point, I will briefly comment on some other issues that Sarah Summers has addressed in her tour d’horizon of the epistemic features of the criminal process.

2. Accuracy

Summers’ first claim is that «truth–finding» in a traditional sense—«accuracy» in the terminology used here—is an important function of the criminal process. She correctly points out that

in the «post–metaphysical» age, the legitimacy of law rests on notions of rationality and reason which are secured principally through the manner in which legal decisions are reached. Reason in legal adjudication is guaranteed in large part by the manner in which information is gathered and processed rather than through any sort of commitment to the subjective belief of the factfinder (p. 254-255).

I fully agree with this statement. State punishment should not be left to the intuition of some wise old judge but needs to be based on a rational process of collecting and evaluating relevant evidence. As is well known, there are competing ways of implementing pertinent rules in the adjudicative process: The law can closely regulate the input for the decisionmaking and then rely on the factfinders’ ability to draw reasonable conclusions from the filtered information that they were provided with; or the law can permit a liberal inflow of information but make strict demands on the rational and systematic processing of this information in the deliberations of the court as documented in the eventual judgment. The common law jury trial is an example of the former approach, the continental bench trial is typical of the latter, output–control approach. In that sense, the rules on admission of evidence in the Anglo–American trial and on reasoned judgments in the German system 1 represent these differing approaches toward promoting rationality 2.

Sarah Summers also mentions that a criminal judgment must be «correctly inferred from its premises» (p. 260). But she seems to require more than a proper correspondence between the trial evidence and the contents of the judgment when she speaks of the need that «the conclusion of the factfinder constitutes warranted true belief in the rule of law» (p. 260). It is not quite clear to me what she would require for a «warranted true belief», but it seems that she regards such a belief as «warranted» only if the previous procedure complied with certain rules designed to guarantee fairness. But how is that requirement related to the «true» element of the factfinder’s belief? Surely truth in any everyday sense cannot depend on the fairness of the process by which this belief has been formed. Even facts found on the basis of a trial conducted, for example, in violation of the defendant’s right to a defense lawyer can be «true» facts in that they reflect what «really» happened. But Summers would probably respond that her idea of procedural truth differs from St. Thomas Aquinas’ concept of truth as adaequatio intellectus et rei. In Summers’ view, the defendant’s guilt can be found to be true in the criminal process only under specific procedural conditions 3. Then the critical question becomes what these conditions may be.

In Summers’ view, one of the foremost prerequisites of true, warranted findings of guilt is the adjudicator’s respect for the presumption of innocence. She sees the presumption of innocence as linked with the prohibition of punishment without law in a common idea of «lawfulness»: «The presumption of innocence and the prohibition on punishment without law are linked by the common commitment to “lawfulness” and the idea of law as a constraint on State power» (p. 263).

Whereas the «roots» of the presumption of innocence are difficult to find 4, I am not convinced that there is a meaningful substantive connection between that presumption and the so–called principle of legality (as expressed by the maxim nulla poena sine lege). Summers quotes, in this respect, the German Constitutional Court to the effect that the presumption of innocence is a special feature of the principle of Rechtsstaatlichkeit (a State based on the rule of law) 5. This is an accurate statement; but Rechtsstaatlichkeit in German doctrine is not limited to «legality» but is a broad umbrella term, similar to the due process clause in U. S. constitutional law, and covers substantive as well as procedural principles.

A more important question concerns the meaning and scope of the presumption of innocence. Sarah Summers follows the dominant view that the maxim in dubio pro reo is an element, even a core element of the presumption of innocence 6. She acknowledges that the wording of Art. 6 (2) ECHR does not bear out that interpretation, but quotes Stefan Trechsel’s account of the legislative history of Article 6 ECHR, according to which the drafters thought that the addition of the words «beyond reasonable doubt» was unnecessary because the phrase «proved guilty according to law» would be understood as conveying that standard of persuasion (Summers, 2023, p. 262).

For the sake of clarity, it would however be preferable to keep apart the different rules on a) presuming the innocence of a suspect until a court has found him guilty and b) the conditions under which a court may arrive at a finding of guilt 7. Both rules—the presumption of innocence as well as the maxim in dubio pro reo—are indispensable elements of a fair criminal process, but they reflect different concerns. The presumption of innocence is to protect the integrity and open–endedness of the trial as well as the reputation of the defendant. The heightened threshold for a conviction, by contrast, reflects the serious consequences of a criminal punishment—especially an undeserved one—for the person convicted and therefore demands that the court, if in (rational) doubt, should acquit. The in dubio principle at the same time demonstrates that the State, in situations of doubt, values the defendant’s liberty interest higher than the community and victim interests in closing the case by a conviction.

Sarah Summers, however, proclaims a different rationale of the in dubio pro reo maxim:

The State is under an obligation from fairness and equality to ensure that it does not impose on any individual a greater risk of harm (such as the harm of being wrongfully convicted or punished) than it imposes on other individuals. A standard of proof which requires a high standard of belief, such as that «beyond a reasonable doubt», has more potential to meet the requirement of standardised or equal application across all cases than the fuzzier «more likely than not» type standards (p. 265) 8.

I must confess that I fail to comprehend the connection between the principle of equality and the standard of proof in criminal cases. It is true, of course, that a decision rule that leaves less to the court’s discretion can create a greater expectation that different judges will decide the same case in the same way. But that is not the issue in criminal justice. Since different cases coming before the criminal courts never are alike, the «risk of harm» of being wrongfully convicted cannot sensibly be compared across cases. If A is acquitted of the charges against him because there remained a reasonable doubt about his guilt, B cannot therefore claim to be likewise acquitted in her (quite different) case. And even if a standard of mere probability of guilt were introduced for criminal conviction, that would still leave each defendant with an equal chance of being acquitted.

Sarah Summers also proposes a new «translation» of the in dubio pro reo maxim to the required state of mind of the finder of fact. She claims that

the State will only be justified in convicting the accused and imposing punishment if it knows that the defendant is guilty. […] The attribution of liability and imposition of punishment inherent in the criminal verdict will only be fair if imposed for the right reasons. This notion of process implies commitment not just to procedural but also to certain substantive guarantees, including a commitment to a robust standard of proof (p. 266).

For me, it is a bit difficult to follow the train of thought here. If I understand Summers correctly, she links the requirement of «the right reasons» for attributing criminal liability to «certain substantive guarantees». And «a robust standard of proof» (i. e., proof beyond reasonable doubt) then is deemed to be such a «substantive guarantee». But isn’t it much rather a procedural requirement that the court is convinced of the defendant’s guilt before it can make a finding of guilt? More importantly, the «knowledge» standard that Summers proposes 9 is difficult to apply to a complex subject matter such as a person’s being guilty of a crime. Can a judge ever «know» (i. e., be completely certain) that defendant D at a certain time and place hit and injured V, and was not justified by self–defense when he did so? If we require for conviction the factfinder’s «knowledge» of all circumstances relevant to guilt, self–conscious judges or jurors may not ever be able to render a guilty verdict, because they may (rightfully) think that even highly persuasive evidence presented in court cannot rule out the possibility that a set of past events did not occur quite like it appeared from the evidence. It might therefore be recommendable to demand no more than a high probability of guilt and an absence of rational arguments that might raise a «reasonable» doubt about the defendant’s guilt.

As an example for the superiority of the proposed «knowledge» standard for conviction, Sarah Summers refers to the U.S. Supreme Court case Shinn v. Ramirez 10, arguing that in this case «the reasons for the imposition of the punishment were flawed in that the factfinder’s belief patently did not amount to knowledge that the accused committed the crime» (p. 268). The Supreme Court’s decision in Ramirez does not, however, deal with the factfinder’s «knowledge» vel non that the petitioners were guilty of murder. The petitioners’ claim was that their lawyers in state court had been incompetent and therefore had failed to raise pertinent factual claims in time. The Supreme Court limited its discussion to certain procedural issues concerning the relationship between federal and state law in habeas corpus proceedings. This decision therefore does not provide the reader with any indication of whether more competent lawyers would have presented evidence that might have swayed the jurors’ opinion on the petitioners’ guilt.

As a further guarantee of the accuracy of a criminal court’s judgment, Summers cites the court’s obligation (in many legal systems) to provide written reasons both on its findings of fact and its legal conclusions. She correctly points out that this obligation not only allows the parties «to determine whether or not an exercise of authority is in fact justified» but also has the purpose «to allow some sort of control of the “reasonableness” of the decision» (p. 255). The obligation to give reasons also provides an element of judicial self–control: a judge who knows that she must write reasons that will hold up on appeal will be careful not to just arbitrarily follow her «gut feeling» in convicting (or acquitting) a defendant.

Sarah Summers thinks that the obligation to provide a reasoned judgment includes «a substantive element and emphasises the connection to lawfulness or legality» (p. 255). She therefore welcomes the tendency of recent judgments of the European Court of Human Rights to not only check whether a court provided any reason but to examine whether the reasons given were sufficient to make sure that the parties and even the public can understand the verdict (p. 255). But the ECtHR itself has repeatedly stated that even a procedural system such as the Belgian, which leaves the decision on guilt or innocence to lay jurors who do not give any reasons for their verdict, does not violate the ECHR as long as the defendant can conclude from the totality of the proceedings (including the indictment) what the jurors’ motivation may have been 11. It thus remains doubtful whether the procedural rights that Articles 5 and 6 ECHR provide can be violated by a judgment of a domestic court that the ECtHR deems to be self–contradictory or implausible 12.

3. Fairness

In Sarah Summers’ concept, the second precondition for arriving at «truth» in the criminal process is the fairness of the proceedings: «Of central importance in this regard is the link between respect for individuals and fair procedures. The rights–based regulation of process in Article 6 ECHR suggests that procedures will be fair if they lead to fair treatment» (p. 254).

The last sentence by itself is a rather circular statement. But the author links «fairness» to «respect for individuals» and thus adds some substance to the bland «fairness» standard. One may wonder, however, whether respect for the individuals participating in the process really is the core concern of fair criminal proceedings or whether respect is only a side constraint. Most observers would agree that a process is fair if each party has an equal chance to participate («equality of arms») and to bring their interests to bear on the resolution of the case, and if the proceedings are conducted in a way that avoids violating the dignity of the participants. Summers’ position may not be far removed from this statement since she defines «adjudication in the rule of law» by citing «the “particular form of participation that it accords to the affected party”, namely an “institutionally protected opportunity to present proofs and arguments for a decision in his favor”» (p. 253) 13.

Summers regards the defendant’s right to be heard as the central feature of procedural fairness, relating all individual rights guaranteed in Art. 6 ECHR back to the right to be heard (p.261). Although I agree that the rights listed in Art. 6 (3) ECHR are necessary for the defendant to have his case presented effectively 14, I would not follow Summers in including the right to remain silent (which is commonly regarded as a critical element of the right to a fair hearing under Art. 6 (1) ECHR) under that same heading (p. 261). She points out that there exists, in the jurisprudence of the ECtHR, a link between the right to remain silent and the presumption of innocence; but the ECtHR (correctly) sees the rationale of the right to silence not in a defendant’s right to be heard but in the need to protect suspects from «improper compulsion by the authorities thereby contributing to the avoidance of miscarriages of justice» 15.

Fair proceedings, we can conclude, demand more than just respect for the trial participants’ dignity and their right to be heard. Fairness in fact is a complex concept that requires the criminal justice system, inter alia, to provide impartial and unbiased judges, to refrain from exerting undue pressure on suspects, defendants, and witnesses, to give the defendant sufficient access to information relevant for the defense and to grant him the freedom to conduct his defense in the way he deems best (short of committing criminal offenses). Only if a procedural system provides this broad array of rights can it be said to guarantee a fair process.

4. Arguing for a thinner concept of truth

This takes me to the final part of my comment. Let us recall Sarah Summers’ understanding of «legal truth» in criminal justice:

Legal truth as embodied in the verdict is a normative concept and should be conceptualised as indivisible from the process of adjudication. The distinction between outcome and process is untenable in the context of fair adjudication in the rule of law (p. 256).

I quite agree that «fair adjudication» requires both «outcome and process», that is, a verdict based on accurate facts found in fair proceedings. But the words «legal truth» convey the idea of a special epistemology exclusive to the criminal process. The term suggests that «truth» in the criminal process is not a statement of facts with the greatest possible approximation to reality, but that «truth» here exists only if the process was properly conducted in accordance with applicable rules. Or, as Summers puts it:

Just as legal evidence clearly differs from «the ordinary concept of evidence» and from notions of evidence discussed in the philosophical literature, so too might the belief of the fact finder in legal proceedings be said to differ from that established in other contexts (p. 257).

As we have seen above, the requirement of proof beyond a reasonable doubt establishes a special rule concerning the degree of certainty with which the factfinder in a criminal process must believe that the defendant is guilty. But that rule cannot be said to create a special form of «belief» as to the existence of relevant facts. All the maxim in dubio pro reo does is to prohibit the court from finding the defendant guilty unless the court is—within the limits of human reason—convinced of (or, as Sarah Summers would say, «knows») his guilt. In other words, truth is truth, regardless of the context.

Theorists are of course free to use terms in any way they like as long as they properly define them. But it seems to me that the amalgamation of epistemological and procedural elements in a single term («legal truth») blurs important differences and tensions. Although criminal procedure law provides for a special process in which the truth about an offense is to be determined for the purpose of imposing punishment on the offender, it would not be correct to say that «truth» about a crime can emerge exclusively from this process. For example, if the survivors of a murder victim sue the perpetrator for damages under tort law, the «truth» about the killing can be determined in civil court in accordance with rules that differ distinctly from those of criminal procedure law. The special rules and safeguards of criminal procedure law have not been devised to produce a special kind of truth but to afford special protection to the defendant in light of his weak position vis–à-vis the powerful state and the severe consequences of a criminal conviction.

Mixing up substantive and procedural elements in the definition of «legal truth» has, moreover, the unwelcome effect of papering over the inherent tension between these two types of elements. Summers correctly remarks that «it is important that verdicts are not wrong too often. […] The law should “care about truth or the avoidance of error”» (p. 259). She also points out that «the State assumes responsibility on behalf of the victim and society more broadly for holding offenders accountable and imposing punishment» (p. 264) 16.

These are strong policy arguments for ascertaining the «true» truth in criminal matters, and they form the basis of the «inquisitorial» ethos of the Continental criminal process. The German Federal Constitutional Court consistently emphasizes this ethos, declaring that it is one aspect of Rechtsstaatlichkeit (the rule of law) that the state provides a functional system of criminal justice, without which justice cannot materialize. A state built on the rule of law can be turned into reality, the Constitutional Court proclaims, only if sufficient measures are taken to prosecute, adjudicate and justly punish criminal offenders 17. The inadmissibility of relevant evidence, which impedes the criminal courts’ task to determine guilt and innocence based on the true facts, must therefore be regarded as an exception 18.

Many procedural «fairness» rules functionally limit or even eliminate the trial court’s ability to base its judgment on the «true» facts of the case. This pertains to rules excluding evidence that was obtained illegally, but also, for example, to testimonial privileges (including the privilege against self–incrimination) that preclude the court from requiring witnesses to testify as well as to rules protecting privacy (concerning, e. g., the contents of private computers, conversations between spouses, or images from private homes). To the extent that procedural fairness and respect for personal privacy limit the evidence available to the finder of fact, these concerns prevent the court from completely elucidating the substantive truth. In adversarial trial systems, a further restriction inhibits fact–finding: parties are free to withhold relevant evidence in their possession from the triers of fact, thus further limiting their ability to base the verdict on all potentially available evidence. Such tensions should not be hidden under a comprehensive concept of «legal truth».

If procedural fairness and the search for the truth point in different, even opposite directions, the question arises what their relationship may be. A full answer cannot be given here 19. In my view, the purpose of the criminal trial is the resolution of the social conflict engendered by the suspicion that a crime was committed. This goal can best be reached by a judgment that has the potential of convincing the parties and the public based on the judges’ ostensible good–faith effort to find out about the relevant facts and to apply the law correctly to these facts. «Truth» is hence not the ultimate goal of the criminal process 20, but the court’s search for it is an indispensable element of justice.

Summers correctly remarks that «legal proceedings must […] be capable of engendering public acceptance in both the process and the verdict» (p. 254). The question remains, however, which aspects of the proceedings play a greater role in the appreciation of «the public». Is it the outcome and the ability of the court to explain its verdict based on the evidence that was available? Or is it the respect for the rights of the defendant and other participants in the proceedings? Ideally, both factors come together to convince the public that justice has been done. But what if one of these elements is missing? Take a trial in which the defendant lacked the assistance of competent counsel and was convicted on the basis of persuasive evidence including his own admission of guilt—and a judgment that acquitted the defendant although he had made a detailed confession that was however found to be inadmissible for procedural reasons. I would guess that «the public» would regard the first judgment as more acceptable than the second. Which certainly does not mean that the correct procedure is irrelevant but that the notion of acceptance by the public is not a «safe» argument in difficult cases.

Summers is right in criticizing concepts such as «integrity» on which some authors try to build an integrated theory of criminal evidence and procedure (p. 267). But the problem is not the emptiness of the term «integrity» but the attempt to assemble substantive and procedural requirements under a common conceptual roof—which is what Summers herself proposes to do, just using the term «rule of law» rather than «integrity» (p. 267).

Sarah Summers plausibly asserts that

a conviction will be wrongful whenever the factfinder’s belief in the guilt of the accused is not formed in the correct way. There will be a miscarriage of justice whenever a conviction is imposed in proceedings which do not meet the standards for the establishment of warranted (legal) belief in the guilt of the accused. Seen in this light there is no contradiction between the right not to be wrongfully convicted and the right to fair procedures which run the risk of a conviction (p. 263).

There can be no doubt that this statement as such is correct. But this statement does not prove that a procedurally faulty conviction leads to an «untrue» finding of guilt, but on the contrary demonstrates the independence of the necessary inquiry into the fairness of the proceedings: even a «truthful» finding of guilt can be a wrongful conviction.

In conclusion, Sarah Summers has made many valuable points about the importance of both accuracy of judgments and their procedural requirements. To satisfy the demands of justice, judgments must have a high probability of accuracy and must emerge from an ostensibly fair proceeding. I think, however, that both requirements are best examined independently of each other. In conducting a differentiated analysis, we should be aware of the tensions that can exist between accuracy of fact–finding and fairness of the criminal process.

BIBLIOGRAPHY

Duff, A., Farmer, L., Marshall, S. and Tadros, V. (2007). The Trial on Trial (Volume 3: Towards a Normative Theory of the Trial). Hart Publishing.

Fuller, L. L. (1960). Adjudication in the Rule of Law. Proceedings of the American Society of International Law, 54, 1-8.

Hruschka, J. (2000). Die Unschuldsvermutung in der Rechtsphilosophie der Aufklärung. Zeitschrift für die gesamte Strafrechtswissenschaft, 112, 285-300.

Lazarus, L. (2012). Positive Obligations and Criminal Justice: Duties To Protect Or Coerce. In L. Zedner and J. V. Roberts (Eds), Principles and Values in Criminal Law and Criminal Justice: Essays in Honour of Andrew Ashworth. Oxford University Press.

Picinali, F. (2013). Two Meanings of Reasonableness: Dispelling the «Floating» Reasonable Doubt. Modern Law Review, 76, 845-875.

Stuckenberg, C. F. (1998). Untersuchungen zur Unschuldsvermutung. De Gruyter.

Summers, S. J. (2023). The Epistemic Ambitions of the Criminal Trial: Truth, Proof, and Rights. Quaestio facti: revista internacional sobre razonamiento probatorio, 4, 249-272.

Weigend, Th. (1989). Deliktsopfer und Strafverfahren. Duncker & Humblot.

Weigend, Th. (2014). Assuming that the Defendant is not Guilty: The Presumption of Innocence in the German System of Criminal Justice. Criminal Law and Philosophy, 8, 285-299.

Weigend, Th. (2018). Zur Reichweite der Unschuldsvermutung. In U. Stein et al. (Eds), Systematik in Strafrechtswissenschaft und Gesetzgebung. Festschrift für Klaus Rogall zum 70. Geburtstag (pp. 739-755). Duncker & Humblot.

---