Studying Religious Symbols and Bias in Court Proceedings

Authors

Nicholas A.R. Fraser, Toronto Metropolitan UniversityFollow
Colton Fehr, College of Law, University of Saskatchewan

Document Type

Article

Abstract

MOCK JURY STUDIES INVESTIGATING the impact of religious symbols on jurors have produced conflicting findings, raising questions about how religious bias should be studied in judicial settings. Our study adds to the literature in three ways. First, we develop a clear theoretical framework which identifies dispositional and situational triggers allowing researchers to more precisely define and measure religiously motivated bias. Second, we show that mock jury experiments with video-based treatments simulating audio-visual cues are more likely to produce reliable findings relative to text-based studies. Third, we present the first empirical evidence of how mock jurors react to courtroom oaths. Our analysis suggests jurors may be prone to trust witnesses who give religious oaths, and this could vary by degree of religiosity and religious group affiliation. These findings shed light on debates concerning whether witnesses should be permitted to swear religious oaths, wear religious garb, or testify about their religious beliefs.

Creative Commons License

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Citation Information

Fraser, Nicholas A.R. and Fehr, Colton. "Studying Religious Symbols and Bias in Court Proceedings." Osgoode Hall Law Journal 61.3 (2025) :
DOI: https://doi.org/10.60082/2817-5069.4067
https://digitalcommons.osgoode.yorku.ca/ohlj/vol61/iss3/3

References

1 Nicholas AR Fraser is a Senior Research Associate at Toronto Metropolitan University’s Canada Excellence Research Chair (CERC) in Migration and Integration. He may be contacted by email at the following address: narfraser@torontomu.ca. Colton Fehr is an Assistant Professor, University of Saskatchewan, College of Law. The authors would like to express their gratitude to the following people who provided valuable feedback at various stages in the research and writing process: Irene Bloemraad, Ming H Chen, Michael J Donnelly, Eric Guntermann, Faisal Kamal, Gabriel Lenz, Emmett Mcfarlane, Kate Puddister, and the Toronto Political Behaviour Group.

2 See Colton Fehr, “Re-thinking the Process for Administering Oaths and Affirmations” (2020) 43 Dal LJ 637 at 640, 650. As Fehr observes, it is not categorically true that those who affirm are atheists. Some religious groups—Quakers and Mennonites—believe swearing to God is a sin. It is doubtful, however, that average jurors would be aware of this nuanced theological fact.

3 Ibid at 642-47. The author cites various sources in support of this proposition.

4 These studies will be reviewed in more detail in Part I, below.

5 See e.g., Canadian Superior Courts Judges Association, “The Judge’s Role,” online: [perma.cc/7B4R-VTV5].

6 See e.g., Evelyn Maeder, Julie Dempsey & Joanna Pozzulo, “Behind the Veil of Juror Decision Making: Testing the Effects of Muslim Veils and Defendant Race in the Courtroom” (2012) 39 Crim Just & Behav 666, DOI: https://doi.org/10.1177/0093854812436478.

7 See e.g., Weyam Fahmy et al, “Unveiling the Truth: The Effect of Muslim Garments and Face Covering on the Perceived Credibility of a Victim’s Court Testimony” (2019) 51 Can J Behavioural Science 53 at 55, DOI: https://doi.org/10.1037/cbs0000116.

8 See Criminal Code of Canada, RSC 1985, c C-46, s 649 [Criminal Code] (criminally prohibiting jurors from speaking about their experience in the juror room); R v Pan; R v Sawyer, 2001 SCC 42 [Pan; Sawyer] (upholding the constitutionality of s 649 of the Criminal Code).

9 See Fehr, supra note 2 at 642-47.

10 Ibid.

11 Ibid at 644.

12 Supra note 8.

13 This provision was found to be consistent with the Canadian Charter of Rights and Freedoms. See Pan; Sawyer, supra note 8. For a more detailed discussion of the research challenges this provision poses, see Paul Quinlan, “Secrecy of Jury Deliberations – Is the Cost Too High?” (1993) 22 CR (4th) 127.

14 See e.g., Valerie Hans & Anthony Doob, “Section 12 of the Canada Evidence Act and the Deliberations of Simulated Juries” (1976) 18 Crim LQ 235.

15 See e.g., Logan Ewanation & Evelyn Maeder, “The Influence of Witness Intoxication, Witness Race, and Defendant Race on Mock Jury Decision Making” (2018) 60 Can J Corr 505, DOI: https://doi.org/10.3138/cjccj.2017-0047.r2; Evelyn M Maeder & Joel Burdett, “The Combined Effect of Defendant and Alleged Gang Affiliation on Mock Jury Decision-Making” (2013) 20 PPL 188, DOI: https://doi.org/10.1080/13218719.2011.633330; Joanna D Pozzulo et al, “The Effects of Victim Gender, Defendant Gender, and Defendant Age on Juror Decision Making” (2010) 37 Crim Just & Behav 47, DOI: https://doi.org/10.1177/0093854809344173; Susan Yamamato & Evelyn M Maeder, “Defendant and Juror Race in a Necessity Case: An Ultimate Attribution Error” (2017) 15 J Ethnicity in Crim Justice 270, DOI: https://doi.org/10.1080/15377938.2017.1347542.

16 See 2012 SCC 72.

17 Fahmy et al, supra note 7.

18 Statistics Canada, “Two-thirds of the Population Declare Christian as their Religion” (19 February 2016), online: [perma.cc/MC42-CPNY]. See also Fehr, supra note 2.

19 See e.g., Angus Reid Institute, “Freedom of Religion Valued Widely in Canada, but the Role of Faith in Society Still a Source of Debate” (4 December 2019), online: [perma.cc/NQ9F-ECEQ].

20 See Marc Helbling & Richard Traunmüller, “What is Islamophobia? Disentangling Citizens’ Feelings Toward Ethnicity, Religion and Religiosity Using a Survey Experiment” (2020) 50 Brit J Pol Sci 811, DOI: https://doi.org/10.1017/S0007123418000054; Lisa Hoffman et al, “Universal Love or One True Religion? Experimental Evidence of the Ambivalent Effect of Religious Ideas on Altruism and Discrimination” (2020) 64 AMJ PS 603, DOI: https://doi.org/10.1111/ajps.12479.

21 We recognize the possibility and likelihood of individuals’ religious beliefs and practices intersecting with other aspects of their identity.

22 For examples, see Hans & Doob, supra note 14; Ewanation & Maeder, supra note 15; Maeder & Burdett, supra note 15; Pozzulo et al, supra note 15; Yamamato & Maeder, supra note 15.

23 A seminal study in the Canadian mock jury literature illustrates this point by using video treatments to show that mock jurors are less likely to trust witnesses who do not make eye contact. See Gordon D Hemsley & Anthony N Doob, “The Effect of Looking Behavior on Perceptions of a Communicator’s Credibility” (1978) 8 J Applied Soc Psychology 136, DOI: https://doi.org/10.1111/j.1559-1816.1978.tb00772.x.

24 See Richard L Wiener, Daniel A Krauss & Joel D Lieberman, “Mock Jury Research: Where Do We Go From Here?” (2011) 29 Behav Sci & L 467 at 468-70, DOI: https://doi.org/10.1002/bsl.989.

25 The smaller the sample size, the harder it is for researchers to distinguish clear patterns in mock jurors’ behaviour from random chance.

26 First, mock jurors were asked to rate the victim’s and the defendant’s testimony on several dimensions (i.e., reliability, accuracy, credibility, truthfulness, and believability) using a 10-point scale (1 = not, 10 = absolutely). Second, mock jurors were asked to render a dichotomous verdict choice (not guilty versus guilty) and then rate their confidence in that choice on a 10-point scale (1 = not confident, 10 = absolutely confident). Third, mock jurors who rendered a guilty dichotomous verdict were asked to rate the victim’s and defendant’s responsibility, desire, and causality for the sexual assault on a 10-point scale (1 = not, 10 = absolutely). Finally, respondents who gave a guilty verdict were asked to provide a sentence to the defendant, rated on a scale from 0 (0-2 months) to 10 (8-10 years). See Maeder, Dempsey & Pozzulo, supra note 6 at 671.

27 See Hemsley & Doob, supra note 23 at 143.

28 See Amna M Qureshi, “Relying on Demeanour Evidence to Assess Credibility During Trial: A Critical Examination” (2014) 61 Crim LQ 235 at 255-61, DOI: https://doi.org/10.2139/ssrn.2384966.

29 See Fahmy et al, supra note 7 at 55. See also William Crain, Theories of Development: Concepts and Applications, 6th ed (Routledge, 2016), DOI: https://doi.org/10.4324/9781315662473.

30 See Fahmy et al, supra note 7 at 55. See also Jackie Gnepp, “Children’s Social Sensitivity: Inferring Emotions from Conflicting Cues” (1983) 19 Developmental Psychology 805, DOI: https://doi.org/10.1037/0012-1649.19.6.805; Linda A Camras & Kevin Allison, “Children’s Understanding of Emotional Facial Expressions and Verbal Labels” (1985) 9 J Nonverbal Behavior 84, DOI: https://doi.org/10.1007/BF00987140.

31 See Fahmy et al, supra note 7 at 55. See also Leslie A Zebrowitz, Luminita Voinescu & Mary Ann Collins, “‘Wide-Eyed’ and ‘Crooked-Faced’: Determinants of Perceived and Real Honesty Across the Life Span” (1996) 22 Personality & Soc Psychology Bulletin 1258, DOI: https://doi.org/10.1177/01461672962212006.

32 See Fahmy et al, supra note 7 at 55. See also Lucy Akehurst et al, “Lay Persons’ and Police Officers’ Beliefs Regarding Deceptive Behaviour” (1996) 10 Applied Cognitive Psychology 461, DOI: https://doi.org/10.1002/(SICI)1099-0720(199612)10:6 461::AID-ACP413>3.0.CO;2-2.

33 See Fehr, supra note 2 at 644-45. See also David Tanovich, “J.(T.R.): Time to Remove Religion from the Oath” (2014) 6 CR (7th) 211.

34 See Tanovich, supra note 33; Fehr, supra note 2 at 645, citing R v AHK, 2011 ONSC 5510 at paras 27-28; R v Bell, 2011 ONSC 1218 at para 57; R v Daud, 2007 BCPC 68 at para 12.

35 See Fehr, supra note 2 at 645, citing R v Ali, 2012 SKPC 62 at para 8.

36 We asked two multiple-choice questions. First, we asked where the fictional crime occurred. Second, we asked what offence the defendant was accused of committing. We only included survey responses from those who correctly answered both questions—identifying that the fictional crime had taken place in a house and that the defendant was accused of theft. The Maeder and Fahmy studies use similar manipulation checks. See Maeder, Dempsey & Polluzo, supra note 6; Fahmy et al, supra note 7.

37 We used the moot courtroom at the University of Alberta, Faculty of Law, which simulates an actual courtroom.

38 As we explained earlier, our fact pattern is designed such that the witness disputes the testimony of another witness supporting the defendant. In this sense, we designed the fact pattern such that the defendant’s credibility is fundamentally tied to their supporting witness who presents an alibi. Hence, by asking about the defendant’s credibility, we encourage participants to also consider that of the witness depicted in our video treatments.

39 The precise wording of the question was as follows: “[p]lease rate the degree to which you believe the defendant’s testimony.” Respondents were given a multiple-choice answer: 1 = “I am confident that the defendant is telling the truth”; 2 = “I am somewhat confident that the defendant is telling the truth”; 3 = “I am uncertain”; 4 = “I am somewhat skeptical of their testimony”; 5 = “I am confident that they are lying.”

40 The precise wording of the question was as follows: “[I]s there any specific reason to believe the witness? Please explain in ten words or less.”

41 See Statistics Canada, supra note 18.

42 We recognize that some religious sects—such as Quakers and Mennonites—refuse to swear an oath. This nuanced theological fact is unlikely to be known to respondents. See Fehr, supra note 2.

43 We recognize there is a clear methodological trade-off here. On the one hand, (mock) jurors commonly expect any witness to give some type of oath or affirmation before giving testimony as this is common practice. As we discuss when presenting the analysis of their written explanations of their credibility ratings, our mock jurors were no different in that those who received the “no oath” video treatment often presumed an oath or affirmation had been given. They did not, however, know which one had been given. On the other hand, the inclusion of a “no oath” control condition allows us to establish a clear and justifiable baseline for comparing the effects that religious oaths and affirmations have on mock jurors’ rating of witness credibility, making it easier to interpret our findings.

44 Notably, we avoided using the accused as the research subject because it may have resulted in skewed results due to general prejudice against those accused of committing a crime.

45 Our fact pattern depicts this person solely in terms of their relationship to the defendant and their alibi. Mock jurors are not given any information about the details of their friendship, how they met, or what they have in common.

46 For a general review, see generally Jeffrey J Rachlinski et al, “Does Unconscious Racial Bias Affect Trial Judges?” (2009) 84 Notre Dame L Rev 1195. The authors explore whether implicit racial biases believed to be present with jurors are also present with judges.

47 Fahmy and colleagues concede this point. See Fahmy et al, supra note 7 at 58. For further discussion of this point, see Katherine H Bullock & Gul Joya Jafri, “Media (Mis)Representations: Muslim Women in the Canadian Nation” (2000) 20 Can Woman Studies 35.

48 See Amy Farrell, Liana Pennington & Shea Cronin, “Juror Perceptions of the Legitimacy of Legal Authorities and Decision Making in Criminal Cases” (2013) 38 Law & Soc Inquiry 773 at 784-92, DOI: https://doi.org/10.1111/j.1747-4469.2012.01323.x.

49 See Statistics Canada, supra note 18.

50 See Angus Reid Institute, supra note 19.

51 It is notable that juries are often criticized for being made up of predominantly white individuals. Recent reforms have nevertheless sought to alleviate this problem. See An Act to amend the Criminal Code, the Youth Criminal Justice Act and other Acts and to make consequential amendments to other Acts, SC 2019, c 25. Kent Roach provides critical commentary on these reforms and how they can be implemented to ensure more representative juries. See “Juries, Miscarriages of Justice and the Bill C-75 Reforms” (2020) 98 Can Bar Rev 315.

52 See e.g., Brian H Bornstein, “The Ecological Validity of Jury Simulations: Is the Jury Still Out?” (1999) 23 Law & Hum Behav 75, DOI: https://doi.org/10.1023/A:1022326807441; Brian H Bornstein et al, “Mock Juror Sampling Issues in Jury Simulation Research: A Meta-Analysis” (2017) 41 Law & Hum Behav 13, DOI: https://doi.org/10.1037/lhb0000223; V Gordon Rose & James RP Ogloff, “Evaluating the Comprehensibility of Jury Instructions: A Method and an Example” (2001) 25 Law & Hum Behav 409, DOI: https://doi.org/10.1023/A:1010659703309.

53 See Angus Reid Institute, supra note 19.

54 It is important to note that the Maeder and Fahmy studies make similar assumptions concerning the external validity of their findings as both rely on student samples. See Maeder, Dempsey & Polluzo, supra note 6; Fahmy et al, supra note 7.

55 This variable was measured using a question with an ordinal scale comprised of three possible answers: 1 = I would only feel bound with a religious oath; 2 = I would feel bound with a religious or secular oath; 3 = I would feel bound without taking any kind of oath. We settled on this measure for several reasons. First, as far as we are aware, there is no standardized way of measuring this in a direct survey question, though we recognize our observational analysis may have yielded different results had we used a different measurement. Second, constructing any direct question about the importance of giving a religious oath in court may cause mock jurors to imagine different types of court cases, including, but not limited to, high- and low-stakes crimes that would require a longer battery of more detailed questions to effectively capture. Third, there was limited space available in our survey. Hence, because this is an exploratory study, we opted for a single question that would reliably capture mock jurors’ broad trust in religious and non-religious oaths.

56 We measured religiosity and trust in different religious groups using Likert scale questions. The question in the survey used to gauge participants’ religiosity asked them to rank themselves on a scale from 0 (“not religious at all”) to 10 (“very religious”). The question about trust in religious minorities asked them to rate their agreement with the following statements: “Devout Christians do more harm than good,”; “devout Muslims do more harm than good,”; “atheists do more harm than good.” They were asked to rate each statement from 0 (“completely agree”) to 10 (“completely disagree”). The phrasing of these questions pertaining to trust in religious groups is based on a similar battery of questions used in a 2019 Angus Reid poll asking Canadians to rate different religious groups’ impact on Canadian society. See Angus Reid Institute, supra note 19.

57 These variables were also measured using Likert scale questions.

58 Figure 1 depicts the predicted probabilities for religiosity in Model 2.

59 The reverse-coded witness credibility rating is as follows: 1-2 = lower trust, 3 = unsure, 4-5 = higher trust.

60 The dots represent mean credibility scores, while the bars display the standard deviation.

61 See Thomas J Leeper, Sara B Hobolt & James Tilley, “Measuring Subgroup Preferences in Conjoint Experiments” (2019) 28 Political Analysis 207, DOI: https://doi.org/10.1017/pan.2019.30.

62 Model 3 shows a statistically significant relationship between all oath types relative to the no-oath condition. By itself, this suggests that mock jurors were agnostic about the type of oath given provided they observed the witness give an oath. Yet, this pattern disappears when we include respondent attributes (Model 4), calling into question the reliability of this particular interpretation of our experimental findings.

63 Technically, we observe a negative correlation between witness credibility and the type of oath the witness takes. Yet, this can be explained by the fact that 1 equals the highest witness credibility rating, whereas 5 equals the lowest. Hence, the negative correlation reflects the way in which we operationalized our witness credibility scale but not the true nature of the relationship between trust in witness testimony and the type of oath mock jurors viewed the witness take.

64 It seems that all participants who mentioned oaths presumed that the witness had taken one, even when they did not directly see it occur (as in the No Oath video). Further, no one criticized the witness for taking a Quran oath or giving a non-religious affirmation. It is notable, though, that more participants mentioned the taking of an oath when they viewed the Bible Oath video. However, due to the small number of participants who fall into this category, we cannot make any (reliable) inferences based on this information.

65 See generally Fehr, supra note 2 at 647.

66 See Thomas White, “Oaths in Judicial Proceedings and Their Effect upon the Competency of Witnesses” (1903) 51 Am L Register 373 at 444, DOI: https://doi.org/10.2307/3306188; Peter Nasmith, “High Time for One ‘Secular Oath’” (1990) 24 L Soc’y Gaz 230; Tanovich, supra note 33 at 213; Jakob de Villiers, “Oath or Affirmation? Or Neither?” (2009) 67 Advocate 199; Sidney N Lederman et al, The Law of Evidence in Canada, 5th ed (Lexis Nexis Canada, 2018) at 936; Ted Matlow, “Let’s Swear off the Oath in Court,” The Globe and Mail (14 March 2000), online: [perma.cc/Y9WW-TUKL].

67 See Fehr, supra note 2 at 648-49, citing White, supra note 66 at 431; White, supra note 66 at 431, citing English Common Law Commission, 2nd Report (1853) at 14; Law Reform Commission of Canada, Report on Evidence, Catalogue No J31-15/1975E-PDF (1975) at 87; Eugene R Milhizer, “So Help Me Allah: An Historical and Prudential Analysis of Oaths as Applied to the Current Controversy of the Bible and Quran in Oath Practices in America” (2009) 70 Ohio St LJ 1.

68 See Fehr, supra note 2 at 650-55. It is notable that Fehr provides a broader defence of this proposal than the one reviewed here.

69 Supra note 16 at para 1.

70 Ibid at para 2.

71 Ibid at paras 16-27. As the Court concludes at para 27:

On the record before us, I conclude that there is a strong connection between the ability to see the face of a witness and a fair trial. Being able to see the face of a witness is not the only — or indeed perhaps the most important — factor in cross-examination or accurate credibility assessment. But its importance is too deeply rooted in our criminal justice system to be set aside absent compelling evidence.

72 Ibid at paras 22, 25.

73 Ibid at para 46.

74 Ibid at para 36.

75 Ibid at para 37. As one important example, the Court observed that requiring a witness to remove her niqab may deter sexual assault complainants from reporting sexual abuse.

76 Ibid at paras 28-44.

77 We recognize that this question requires further empirical study. However, the combination of our study, the Fahmy study, and our criticisms of the Maeder study suggest that there is a reasonable basis for drawing this inference. See Maeder, Dempsey & Polluzo, supra note 6; Fahmy et al, supra note 7.