Karaktärsbevisning: - studier av relevansproblematik i personrelaterad bevisning

Research output: ThesisDoctoral Thesis (compilation)


This dissertation is a compilation thesis on character evidence in the Swedish legal system. It consists of five articles published in scientific journals and a comprehensive introductory chapter (kappa) on the key concepts, methodology, and procedural law with regard to character evidence in the Swedish system. The dissertation investigates how character evidence is assessed in Swedish courts, and combines doctrinal legal method and conceptual analysis with empirical studies.
The dissertation shows that character evidence is a special kind of person-related evidence that is predictive in nature. Character evidence uses a personal circumstance about a person, X, in support of an evidentiary theme (hypothesis) about X’s propensity to behave in a certain way. For example, the personal circumstance that X has previously been convicted of perjury provides support for the evidentiary theme that X has a propensity to lie and is therefore not credible as a witness in the case at hand. Character evidence relies on a two-step inference. X’s conviction for perjury supports the evidentiary theme that X has a propensity to lie, which in turn supports the evidentiary theme that X’s testimony in the present case is not credible. Since the second step in this inference is a prediction, character evidence is a special kind of predictive evidence.
There are different kinds of character evidence, and the dissertation divides the domain of character evidence into four subcategories: 1) bodily functions, 2) learning, 3) personality traits, and 4) personal interests. For example, X’s credibility or lack of credibility as an eyewitness can be supported with the following arguments: “X was drunk and cannot be trusted to have observed the situation correctly” (bodily function); “as a police officer, X is trained to observe events correctly in stressful situations” (learning); “X has lied in court before and is ready to lie again” (personality trait); “X is the defendant’s mother, and a mother would say anything to protect her son” (personal interest).
Character evidence often relates to credibility, and the dissertation provides a tool for credibility assessments that models credibility as a conceptual space with three dimensions: 1) perception, 2) memory, and 3) report. In this model, credibility is a function of the ability to perceive a situation correctly, remember this perception, and report the memory truthfully. This model is developed in one of the articles in the dissertation (“Trovärdighetens dimensioner”). The article also investigates how the three dimensions are reflected in the reasoning of legal fact-finders.
One of the articles in the dissertation investigates empirically how Swedish fact-finders assess character evidence on the basis of criminal records (“Att bedöma personer med kriminell belastning”). The study is based on a survey of professional Swedish judges. The survey investigated: 1) whether the assessment of guilt is affected by prior conviction for the same offence and 2) whether a criminal record affects the assessment of a person’s credibility. 62 % of participants responded that prior conviction for the same offense increases the probability that the defendant is guilty in the present case. With regard to credibility, 61 % of participants responded that a criminal conviction for forgery has a negative effect on a person’s credibility, but only 8 % of participants responded that a conviction for assault would have an effect on credibility. The difference between the effects of convictions for forgery and assault confirms the often-heard claim that crimen falsi (crimes involving deception) have a greater impact on credibility.
Character evidence typically involves complex issues about the legal relevance of certain circumstances. The concept of legal relevance and the assessment of legal relevance are therefore central themes throughout the dissertation. The dissertation shows how character evidence can be directly relevant, indirectly relevant, or irrelevant to the legal issue at hand. Character evidence is directly relevant when it supports an evidentiary theme that satisfies a condition in a legal rule. When there is a chain of evidentiary inferences connecting the character evidence to the legal rule, the character evidence is indirectly relevant. Such chains contain several evidentiary themes, and each evidentiary theme in the sequence gives evidential support to the next. If no such chain can be constructed, the evidence is legally irrelevant.
Two of the articles in the dissertation investigate empirically whether legal fact-finders are influenced by legally irrelevant character evidence (“Är det sant att lekmannadomare påverkas av juridiskt irrelevanta omständigheter, men inte juristdomare?” and “Are Judges Influenced by Legally Irrelevant Circumstances?”). The articles present the results of an experiment in which professional judges and lay judges in the Swedish legal system participated in a mock study of a libel case. The test group received the same information as the control group, except for the additional information that the victim of libel in the case at hand had committed libel himself in a previous, unrelated case. The compensation awarded to the victim by professional judges who had received the additional but legally irrelevant information was 35 % lower (mean value) than the compensation awarded by professional judges who had not received this information. Among the lay judges the compensation awarded by the respondents in the test group was 53 % lower than the control group. The study confirms suspicions that legal fact-finders can be influenced by legally irrelevant character evidence.
The opposite problem occurs if legal fact-finders overlook character evidence that actually is relevant. The dissertation investigates this problem in an article on conflicts of interest in environmental law (“Investigator Self-Interest in the Environmental Process”). The study shows that objections about investigator self-interest are not addressed by the courts and do not seem to affect the courts’ decision. The article also analyzes the concept of conflict of interest and proposes a new conceptual category: arguments about behavioral impact (arguments to the effect that the conflict of interest is visible in the investigator’s behavior).
The dissertation provides a conceptual analysis of character evidence that contributes to the theory of legal evidence and empirical studies that contribute to our knowledge about the actual effects of character evidence on legal fact-finders. The investigation of the legal framework within which the empirical studies are conducted contributes to the knowledge of the context in which the evaluation of character evidence takes place.
Original languageSwedish
Awarding Institution
  • Department of Law
  • Dahlman, Christian, Supervisor
  • Wahlberg, Lena, Assistant supervisor
  • Westberg, Peter, Assistant supervisor
Thesis sponsors
Award date2021 May 24
ISBN (Print)9789178958054
ISBN (electronic) 9789178958061
Publication statusPublished - 2021 Apr 21

Bibliographical note

Defence details
Date: 2021-05-24
Time: 10:15
Place: Pufendorfsalen, Juridiska fakulteten, Lilla Gråbrödersgatan 3, Lund
External reviewer(s)
Name: Gräns, Minna
Title: Professor
Affiliation: Uppsala University

Subject classification (UKÄ)

  • Law

Free keywords

  • Procedural law
  • Character evidence
  • Law of evidence
  • Evaluation of evidence
  • Empirical studies
  • Multi-methodological approach
  • Judicial decision-making

Cite this