A practical guide for volunteer peer facilitators of Authentic Relating “connection games”, written as a social-services thesis with the Finnish association Autenttiset ry.
Safety, facilitation and harm
What makes a circle safe, and what goes wrong when it isn’t?
Intense relational practices can unsettle as well as connect. No one has yet studied adverse experiences in Circling, or what its facilitators do to prevent them, systematically.
Builds on 14 works
Why it matters
Credibility depends as much on knowing the risks as on showing the benefits. Practices that ask for disclosure and intense attention will sometimes overwhelm people, and facilitation ranges from trained professionals to peers.
What exists
Silvennoinen (2025), a guide for volunteer facilitators, covers handling difficult situations and intense emotions from the organisers’ side. Nothing yet asks Circling participants systematically about difficult experiences.
The closest precedent is the Stanford study of 1960s encounter groups (Yalom & Lieberman (1971)). It counted lasting casualties, linked them to the leader’s style, and found that the leader was a poor judge of who had been harmed; the other members were better ones.
Later work tests and extends it. In T-groups with no trainer, about one member in ten was named as harmed, but most of them were named as helped too (Bramlette & Tucker (1981)). A study with a placebo group, a control group and a six-month follow-up found signs of harm no more often among encounter-group members than among the others (Posthuma & Posthuma (1973)). Lieberman himself later found no psychiatric casualties among more than 300 people who took a large group awareness training (Lieberman (1987)).
In clinical group treatment, patients who started more distressed, felt a weak alliance with the group or were strongly stirred in sessions reported more negative effects, and those effects went with worse outcomes rather than better ones (Pourová et al. (2023)). In a psychiatric hospital, low moods after single sessions did not predict how patients fared, while unwanted reactions to treatment did (Schneibel et al. (2017)). Roback (2000) gathers the risks on the leader’s side, the group’s and the member’s, and argues that harm comes from how they combine. On prevention, a randomised trial found that feeding back how members rated their relationships in a therapy group turned deteriorating relationships around (Burlingame et al. (2018)).
A possible design
Survey current and former participants about difficult or harmful experiences, adapting instruments from research on adverse effects of meditation. Follow up with interviews of facilitators about how they notice and respond to distress, and, as the Stanford study did, ask the other members of a group as well as its leader. Ask about gains as well as harms, follow people up, and compare them with people who did not take part, so that a hard session is not counted as lasting harm.
Pitfalls
People who were harmed are the least likely to still be in the community, so recruit beyond it.
Builds on
- 2025Yhteyspeliohjaajan opas [Connection-game facilitator’s guide]
- 2022Reconceptualising emotional capital and intimacy using a sociological lens: the Authentic Revolution
An ethnography of Authentic Relating and Circling communities, from interviews and participant observation: how they cultivate emotional intimacy and respond to social isolation.
- 1971A Study of Encounter Group Casualties
Cautions: Same study as another entry · Self-selected participants
Of 170 students who finished a 30-hour encounter group, 16 were left with lasting harm. Harm clustered under forceful, charismatic leaders who pressed individuals hard instead of working with the group.
- 1981Encounter Groups: Positive Change or Deterioration? More Data and a Partial Replication
In 39 T-groups with no trainer, 10% of 327 members were named as harmed and 54% as helped. Most of those named as harmed were named as helped too; 3% had only a bad experience.
- 1973Some Observations on Encounter Group Casualties
Encounter-group members differed from a placebo group and a control group just after the groups ended, but by six months the differences had gone, and signs of harm were no more common among members.
- 1987Effects of large group awareness training on participants’ psychiatric status
Cautions: Self-selected participants
Psychiatric interviews with more than 300 people who had taken a large group awareness training found no psychiatric casualties and only a few stress reactions.
- 2023Negative effects during multicomponent group-based treatment: A multisite study
Cautions: Self-report only · No later follow-up
Patients who began group treatment more distressed, rated the group alliance poorly or were strongly stirred in sessions reported more negative effects, and those effects came with worse outcomes, not better ones.
- 2017Adverse events of group psychotherapy in the in-patient setting – results of a naturalistic trial
Most in-patients in group therapy had low moods after sessions or unwanted reactions. Unwanted reactions went with less improvement, but low moods after sessions did not predict how patients fared.
- 2000Adverse Outcomes in Group Psychotherapy: Risk Factors, Prevention, and Research Directions
A review of what makes therapy, growth and self-help groups harmful. It sorts the risks into the leader, the group and the member, and argues that harm comes from how the three combine.
- 2018The effects of relationship and progress feedback in group psychotherapy using the Group Questionnaire and Outcome Questionnaire–45: A randomized clinical trial
Cautions: Studied by its developers
Adding feedback on members’ relationships in the group to feedback on their progress turned deteriorating relationships around and cut how often they failed. Effects on symptoms were mixed.
- 2016Negative Effects of Psychological Treatments: An Exploratory Factor Analysis of the Negative Effects Questionnaire for Monitoring and Reporting Adverse and Unwanted Events
Cautions: Self-report only
A questionnaire study organised reported negative experiences of psychological treatment into six dimensions, offering a tool for monitoring possible harm.
- 1989Psychological effects of participation in a large group awareness training.
Cautions: Self-selected participants · Self-report only · Commercial interest
A non-randomised study of the Forum found a short-term gain in perceived control, but its main outcome models found no lasting effects.
- 2020Adverse events in meditation practices and meditation‐based therapies: a systematic review
A systematic review found reports of adverse experiences during or after meditation, with prevalence estimates varying greatly by study design and monitoring method.
- 1972The Impact of Encounter Groups on Participants: Some Preliminary Findings
Cautions: Same study as another entry · Self-selected participants
A preliminary evaluation of encounter groups reported both benefits and harms and examined how outcomes related to leaders’ behaviour.