Mediation begins. The parties take their seats at the same table. The mediator explains the process, the parties speak briefly. Then the doors close. Everyone is now in a separate room. The mediator shuttles back and forth: from one room to the other, offer to offer, position to position.
In Turkey, this scene is familiar. It is rarely questioned.
The caucus — as private sessions are known in international practice — refers to the format in which the mediator separates the parties and works with each side individually. It is a powerful tool. Used well, it accelerates the process, breaks deadlock, and allows parties to speak with candour. Used out of habit, or applied without deliberate intent, it quietly dismantles the most valuable element of mediation: the direct connection between the parties.
The Legal Framework: Caucus in Turkish Law
Article 10 of the Regulation on Mediation in Civil Disputes (Law No. 6325) provides clear authority: “The mediator may meet with the parties separately or jointly.” The provision leaves the use of caucus entirely to the mediator’s discretion. There are no restrictions on how it is deployed, when it begins, or when it should end.
The same principle applies internationally. Article 8 of the UNCITRAL Model Law on International Commercial Mediation explicitly confirms that mediators may meet with parties jointly or separately.
The legal foundation is solid. The real debate concerns what happens in practice.
When the Caucus Works
There are circumstances in which the caucus is unambiguously effective.
In cases of high emotional tension, parties may find it difficult to speak openly in each other’s presence. In a long-running partnership dispute or an intergenerational conflict within a family business, direct confrontation can inhibit rather than facilitate dialogue. The caucus functions as a pressure valve in these situations, allowing each party to share genuine concerns and the interests underlying their position with greater candour.
When there is a significant power imbalance, the caucus also serves a protective function. If one party is dominant and the other is becoming withdrawn, a joint session does not offer level ground. Private sessions allow each party’s voice to reach the mediator on equal terms.
Caucus is also indispensable for reality testing. A mediator needs to be able to address the weaknesses in a party’s position — and to do so privately, not in front of the other side. This is one of the most consequential interventions a mediator can make.
When the Caucus Quietly Works Against the Process
Here is a reality that much of the literature overlooks.
Every message is filtered. However skilled the mediator, what travels from one room to the other is not the message itself but the mediator’s interpreted version of it. Tone, hesitation, remorse, sincerity — these cannot be transmitted. They are felt, heard, read on a face. They do not travel down a corridor.
In certain disputes, this limitation produces a meaningful loss. In partnership breakdowns, family business conflicts, and long-standing commercial relationships, what parties often need is not a number. They need to see that the other side genuinely understands something — that a decision had real consequences, that a relationship carried real weight. For that, being in the same room matters.
Apologies illustrate this most clearly. In mediation, parties frequently seek some form of acknowledgement, even when they do not articulate it explicitly. When a mediator carries the words “I understand what this caused” or “I could have handled that differently” from one room to the next, those words often land as a tactical manoeuvre. Delivered face to face, the same words can change the entire dynamic of the negotiation.
Extended caucusing also generates suspicion. Parties begin to wonder what is being said in the other room, what the mediator is emphasising, what is being softened. Even when those concerns are unfounded, the perception alone erodes trust over time. The process becomes transactional: numbers move, but nothing of substance is resolved
Where Global Practice Is Heading
The balance between caucus and joint session has become a substantive debate in the international mediation community.
In the United States, joint sessions — the format in which all parties remain together — declined steadily over three decades as caucus-dominant practice took hold. Since 2021, however, a significant counter-movement has emerged. Galton, Love and Weiss, in their article “The Decline of Dialogue,” characterise the disappearance of joint sessions as a loss of mediation’s dialogue function, and argue that prevailing practice warrants fundamental reconsideration.
In England and Wales, the standard approach is to begin with a joint session and move to private caucus when necessary. In a system where commercial mediation has deep institutional roots, the caucus is positioned as a tool, not a default.
In Australia, the facilitative model predominates. Joint sessions are placed at the centre of the process; caucus is reserved as an intervention when an impasse requires it.
In Singapore, cultural dynamics make caucus more prevalent. In cross-border commercial disputes involving Asian parties in particular, direct confrontation across the table is frequently avoided by preference.
Beneath all these differences lies a consistent finding. A 2022 study involving more than 1,000 mediators across eight US states found that the majority apply the same structure in nearly every case — regardless of the specific dynamics of the dispute. The decision to caucus is driven more by the mediator’s habitual practice than by any assessment of what the case requires. Researchers describe this as the “cookie cutter” problem: one template, applied uniformly.
A Note on Turkish Practice
Mediation practice in Turkey has been built around mandatory mediation. In labour and commercial disputes, many cases conclude within two hours. In this speed-driven environment, caucus has shifted from being a deliberate technical choice to a default shortcut.
As commercial mediation grows, and as institutional and cross-border disputes become more common, this question will become more consequential. In partnership disputes, family business conflicts, and long-term commercial relationships, speed is not the only measure. The durability of the settlement, the future of the relationship, and the quality of the parties’ experience of the process matter at least as much.
The maturation of Turkish mediation practice will, in part, be marked by a shift in the central question: from “how do we resolve this faster?” to “which approach is right for this dispute?”
Tool or Habit?
The caucus is one of mediation’s most effective instruments. Applied at the right moment, for the right reasons, it moves a process forward, breaks deadlock, and gives parties the safety they need to speak honestly.
Applied automatically, as a template independent of the case at hand, it disables the most valuable dimension of mediation: the capacity for parties to see, hear, and understand each other directly.
The right question is straightforward: “Why am I moving to caucus in this case?”
If that question does not have a clear answer, the caucus is functioning as a habit, not a tool. And habits, in mediation, rarely produce the best outcome.
References
- Regulation on Mediation in Civil Disputes (Law No. 6325), Article 10, mevzuat.gov.tr
- UNCITRAL Model Law on International Commercial Mediation, Article 8, uncitral.org
- Wissler, R. & Hinshaw, A. (2022). Joint Session or Caucus? Factors Related to How the Initial Mediation Session Begins. Ohio State Journal on Dispute Resolution, 37(4)
- Wissler, R. & Hinshaw, A. (2024). Comparing Joint Session and Caucus Outcomes. SSRN
- Galton, E., Love, L. & Weiss, J. (2021). The Decline of Dialogue: The Rise of Caucus-Only Mediation and the Disappearance of the Joint Session. Alternatives to the High Cost of Litigation, 39(6)
- Hisey, J. (2026). Caucus: A Blessing and a Curse in Mediation. Mediate.com
- Alexander, N. (2024). Comparative Mediation Law Mapping Mediation Law with the Mediation Matrix. Comparative Procedural Law and Justice, CPLJ XV 4
