Fodor’s RHBS Model of Resurrection Evidence: Part II
Individual hallucinations in non-clinical populations
Part I assessed Fodor’s argument for reburial (R). I argued his proposal lacks plausibility in a first century Jewish context.
In Part II, I will consider the next element of the RHBS model — individual hallucinations (H).
Individual hallucinations: a possibility
Surveys show hallucinations can occur in non-clinical populations. There is also a well established literature on hallucinations in bereaved populations. So on this point, Fodor and I agree.
How common are individual hallucinations in non-clinical populations?
I’m less persuaded by his discussion of how common these experiences are. One reason is a recent systematic review (published a few years after his book). Staines et al. (2023) offer a comprehensive summary of these surveys.
A further benefit of Staines et al.’s work, is that they focus on incidence (the number of new cases in a population) of hallucinations in the general population. I think this is more likely to reflect the nature of the disciples’ experience.
Staines and colleagues identified 17 studies, which included over 56,000 people. They focused on studies that used validated measures of hallucinations to minimize bias. They found an incidence of 2% in the general population per year experienced a hallucination or delusion. Since our focus is on hallucinations, this is likely an overestimate. But nonetheless, a helpful reference point.
How does Fodor’s estimates compare?
Fodor (2018) cites far higher estimates: 10–25% (Morrison 2001), 10–37% (Davies et al 2001). The one survey he focused on, found 37% of the general population experienced some form of hallucination (Ohayon, 2000).
An important difference is how they defined hallucinations. Ohayon (2000) included a much broader range of experiences under the term “hallucination” — including out of body experiences and various other phenomena. Around a third of the reported experiences occurred at the onset of sleep or awakening — e.g. hypnagogic hallucinations. As Fofanni (2024) points out:
The DSM-5 [standard diagnostic criteria of the American Psychiatric Association] specifies that “the term hallucination is not ordinarily applied to the false perceptions that occur during dreaming, while falling asleep (hypnagogic), or upon awakening (hypnopompic).” In other words, to experience genuine hallucinations, one should be fully awake and, reciprocally, hypnagogic/ hypnopompic experiences are not hallucinations. (p2)
Interestingly, daytime rates of hallucination (closer to the DSM-5 standard definition) were much lower. Only 0.6% of their sample reported experiencing an auditory hallucination, 2.6% haptic hallucinations (touch), 2.4% visual hallucinations. These estimates are similar to Staines’ findings. The more rigorously defined experiences considered in Staines’ review are in my view preferable.
Bereavement hallucinations (H)
Fodor rightly points out bereavement hallucinations are common. But there are, in my view, issues with his summary. He cites one study (Grimby, 1993), of 50 people. Approximately one-third reported visual, auditory hallucinations, or verbal hallucinations. In addition, 50% reported a sense of presence of the deceased.
There are two main problems. First, it felt to me Fodor’s summary of the study was selective. He did not report the prevalence (6%) of tactile hallucinations (which were far less common). He also did not discuss the prevalence of multi-modal hallucinations (e.g. visual, auditory, and tactile hallucinations experienced at the same time). These were less common (12% for three different modalities, and 3% for four to five modalities). He may dismiss multimodal experiences (including tactile ones) reported in the Gospels as fictional. But it is important to make the case for this assumption.
The second issue is that Grimby’s estimates are unlikely to be representative of the wider literature. In my summary of 18 studies on bereavement hallucinations (Meader, 2024), median estimates were approximately half those cited by Fodor:
- visual: 14%
- auditory: 15%
- verbal: 12%
Grimby’s estimate of tactile hallucinations was similar to the overall literature (6% compared with a median of 5%). Therefore, in my view, the estimates he cites on the prevalence of hallucinations are far higher than that reflected in the literature as a whole. This has important implications for the validity of his RHBS model.
Hallucinations and insight
The insight many non-clinical populations have about their experience, is a further challenge to the RHBS model. For example, Anna Castelnovo (2015), a psychiatrist and researcher argues:
No matter how vivid such visions maybe, to the extent that some people behave
in response to them, reality testing [they know that these experiences are not
reflective of “reality”] seems preserved, at least in the absence of conditions
such as pathologic grief reactions, sub-threshold or overt psychiatric/neurologic
disorders, sensory/cognitive impairment, or drug/alcohol abuse. (p272)
For instance, Linszen et al. (2022) conducted an online study with 10,000 non-clinical participants. They found that only 10.2% of those with auditory hallucinations and 11.4% with visual hallucinations were fully convinced they were real.
This is consistent with many theories, that hallucinations occur on a spectrum of severity. So it is not unexpected that non-clinical populations are more likely to be able to discern reality from hallucinations. This reduces the probability that Jesus’ disciples would interpret these experiences as evidence he was resurrected.
New religious movements and fantasy prone populations
Of course, the RHBS posits other factors, such as bias and socialisation, that help to bolster this hallucination hypothesis. Another component to this argument, is to posit that Jesus’ disciples were hallucination prone:
There is also substantial evidence that some people are considerably
more likely to experience hallucinations than others. Such people have been described as exhibiting a ‘fantasy-prone personality’, and are more likely to be involved in paranormal or new religious movements, both of which facilitate opportunities for unusual experiences. Many of Jesus’ followers
would have fitted into this category, and therefore would have been
particularly susceptible to hallucinations.
Fodor supports this view by citing a study by Day and Peters (1999). They compared ratings of schizotopy (a risk factor for psychosis e.g. hallucinations) in three groups:
- New religious movements (Hare Krishna and Druids).
- Christians.
- Non-religious people.
Members of Hare Krishna and Druid groups scored higher than Christians and non-religious people for the category, “unusual experiences.” Thought to be associated with susceptibility to hallucinations. My impression is that Fodor extrapolates these risk factors found for Hare Krishna and Druids to Jesus’ disciples.
Methodological critique: potential confounding
Day and Peters (1999) argue their data shows an association between new religious movement (NRM) membership and risk of hallucinations. However, their study did not adjust for a crucial risk of bias — confounding. Epidemiologists distinguish between three factors in a study (see Figure 1):
- Exposure (in this case, being a member of an NRM).
- Outcome (what you are trying to predict — risk of hallucinations).
- Confounders, these are factors (in Figure 1, this is “X”) associated with both the exposure and outcome.
Unadjusted confounders can bias estimates on the association between exposure (NRM) and outcome (risk of hallucination). The best way to avoiding confounding is to conduct a large randomized trial — which is not possible (and unethical!) in this context. But good studies generally use one or more of these strategies to reduce, or at least assess, the risk:
- Identify a range of potential confounding factors and ensure you use well-validated measures.
- Where possible, minimize differences between groups on key confounding factors (e.g. matching).
- Conduct statistical adjustments for the impact of confounders.
Day and Peters (1999) provide information on age and gender. However, there are other potential factors neither measured nor accounted for in their analyses:
- Research on NRMs, has suggested long-term members, “tend to have experienced some form of psychological distress and alienation before joining.” (Urban, 2015, p308) If so, confounding is likely to impact the association between NRM membership and risk of hallucinations.
- Control groups (non-religious and Christians) — many were acquaintances of the researchers (42/73). This population may differ from Druid or Hare Krishna groups for various reasons. For example, academics’ acquaintances may be of higher socio-economic position than NRM members. In turn, risk of hallucinations and socio-economic position are associated. Which would bias results.
Methodological critique — Christianity or NRM?
A further challenge for Fodor’s argument is that one of the control groups (i.e. non-NRM) were Christians. He suggests that, since members of NRMs in our day have greater risk of hallucinations, Christians in the first century were also prone to this risk. But application of these data to first century Palestine is far from obvious.
Could we not claim that Day and Peters (1999) show Christians are at low risk of hallucinations, therefore so were Jesus’ disciples? Of course, I’m not arguing that. There are vast cultural differences between first century and twentieth century Christians.
But are cultural differences between first century Christians and twentieth century NRMs less extensive? That is something Fodor has to justify.
Sociological critique
As Hugh Urban (2015), professor of religious studies, points out:
Much of the best scholarship on new religious movements has been generated by sociologists, who hope to understand the broader social and cultural contexts within which new religions arise and the reasons why these become attractive alternatives for many individuals. (p300)
The term NRM, as the academic literature uses it, refers to a particular social and cultural context — primarily the Western world since the mid-1960s. Or at latest, the 19th century. This research has shown NRMs in the Western world:
offer powerfully attractive alternatives to a confusing modern world that can often feel like a “competitive, claustrophobic rat-race” devoid of community or meaning. For many who find contemporary society alienating and isolating, new religions can provide both a new kind of family and a spiritual framework that gives life an ultimate value and purpose. (Urban, 2015, p.301)
There is likely to be some overlap in experience with other times or places. But we cannot assume mechanisms that impact on experiences of NRM members in the West generalise to first century Palestine.
Summary and conclusions
We have found aspects of agreement with the RHBS model. I agree hallucinations occur in the general population. Yet I think Fodor overestimates the probability of hallucinations in these populations.
It is a possibility Jesus’ disciples could have experienced a bereavement hallucination. But I think the bereavement literature, as a whole, suggests lower estimates on how common they are.
Paul was not a disciple of Jesus — so the bereavement hallucination literature is unlikely to be applicable to him. Incidence rates in the broader population are much lower. A recent systematic review suggests 2% of the general population per year, Fodor’s estimates are up to 20 times higher and focus on prevalence.
He also sought to argue Jesus’ disciples were hallucination prone, because they belonged to a new religious movement. Yet, I have pointed out several limitations which challenge this conclusion.
The next instalment of this series will consider the Biases (B) component of the RHBS model.
Bibliography
Castelnovo, Anna et al. Post bereavement hallucinatory experiences: A critical overview of population and clinical studies. Journal of Affective Disorders 186 (2015) 266–274.
Day, Susan and Peters, Emmanuelle. The incidence of schizotypy in new religious movements. Personality and Individual Differences 27 (1999), 55–67.
Foffani, Guglielmo. To be or not to be hallucinating: Implications of hypnagogic/hypnopompic experiences and lucid dreaming for brain disorders. PNAS Nexus 3 (2024) 1–11.
Fodor, James. Unreasonable Faith. Hypatia, 2018.
Grimby, A. Bereavement among elderly people: grief reactions, post-bereavement hallucinations and quality of life. Acta Psychiatrica Scandinavica 87 (1993) 72–80.
Linzsen, Mascha. Occurrence and phenomenology of hallucinations in the general population: A large online survey. Schizophrenia 8 (2022) 41.
Meader, Nick. Resurrection. Wipf&Stock, 2024.
Morrison, Anthony P. The Interpretation of Intrusions in
Psychosis: An Integrative Cognitive Approach to Hallucinations
and Delusions. Behavioural and Cognitive Psychotherapy 29 (2001) 257–276.
Ohayon, Maurice M. Prevalence of Hallucinations and Their Pathological Associations in the General Population. Psychiatry Research 97 (2000) 153–64.
Staines, Lorna et al. Incidence and Persistence of Psychotic Experiences in the General Population: Systematic Review and Meta-Analysis. Schizophr Bull 49 (2023)1007–1021.
Urban, Hugh. New Age, Neopagan, and New Religious Movements. University of California Press, 2015.
*** Disclaimer: I receive royalties for my book linked above
