Escalation Moves

Tactics That Intensify Rather Than Resolve

Glasl's Conflict Escalation Model

Friedrich Glasl identified 9 stages of conflict escalation, grouped into three phases. Each stage involves increasingly destructive tactics. The patterns documented here show movement from Stages 4-5 (coalition building, character attacks) to Stages 6-7 (threats, limited destructive blows), with some elements of Stage 8 (attempts to destroy the opponent's ability to function).

For quantitative Glasl analysis across 8 community members: View Comparative Analysis

4.2 Average Glasl Elevation Per Attack
4-5 Primary Operating Range
6-7 Peak Escalation
18 mo Pattern Duration

The Glasl Framework

Stages 1-3: Rational Conflict

StageNameCharacteristics
1 Hardening Positions crystallize, but parties still believe rational discussion can resolve differences
2 Debate Polarization begins; parties try to convince each other through argument
3 Actions, Not Words Talk is abandoned; parties act unilaterally. Empathy declines.

In Phase 1, both parties can still win. Conflicts can be resolved through dialogue or mediation.

Stages 4-6: One Must Lose

StageNameCharacteristics
4 Images & Coalitions Parties recruit allies; stereotypes form. Focus shifts from issues to people.
5 Loss of Face Public attacks on character; attempts to expose opponent. Moral positioning.
6 Strategies of Threats Ultimatums; demands. Stress and time pressure. Threat/counter-threat cycles.

Stages 4-5 are the primary operating range in the documented patterns. Stage 6 appears in the ATL incident.

Stages 7-9: Into the Abyss

StageNameCharacteristics
7 Limited Destructive Blows Harm becomes the goal. Opponents seen as objects, not people.
8 Fragmentation Systematic destruction of opponent's support systems and ability to function
9 Together Into the Abyss Mutual destruction accepted. No price too high.

The ATL abuse during mediation shows Stage 7 characteristics — punitive harm became an explicit goal.

Documented Escalation Tactics

Identity Reframing +

Stage 5

Transform technical disagreements into identity attacks

How It Works

A disagreement about materials, scheduling, or process is reframed as sexism, racism, or other identity-based harm. This shifts the conflict from resolvable (we can discuss materials) to existential (you're a sexist).

The Carbon Fiber Example

A technical dispute about cutting carbon fiber materials became:

"TL:DR: If not sexism, what? How are we all now in Bravespace? It started I expressed a specific concern about safety... they were also treating the only 2 women in the conversation like we were idiots." — Elle, July 14, 2024

→ See full context: The Bravespace Incident

Why It Escalates

  • The accused must now defend their character, not their position
  • Allies can join without understanding the original issue
  • Resolution becomes nearly impossible — you can't compromise on identity

Cloud's Recognition

"I don't want my name used to attack another person. Elle uses my gender to attack other people she does not like and ropes me into discussions I did not consent to be a part of." — Cloud

Triangulation +

Stage 4

Recruit third parties to fight by proxy

How It Works

Rather than address the person directly, the triangulator recruits allies who then attack on their behalf. This creates confusion about who the conflict is actually with.

The Nicole DM

"I don't think you have any idea how exhausted Elle is from dealing with how easily you and others dismissed her concerns with this fiberglass guitar project." — twiceasnice (private DM to Wyatt, "on Elle's behalf")

Wyatt's Response

"I'm assuming this is somewhat directed at me because @twiceasnice you also went out of your way to privately dm me on Elle's behalf that she is exhausted with having to deal with this situation." — Wyatt, July 14, 2024

The LX Mobilization

After WE/Z asked Elle to take a break from meetings:

"Elle just told me someone ATLd her... this is honestly a NB sabotaging move of all time" — LX (who wasn't present), DM to nthmost

WE/Z's actual words: "Take a break from meetings for a month" and "You're not in trouble." Elle "gave me a hug."

Disengagement as Weapon +

Stage 6

Use boundary demands to control, then violate them yourself

The Escalating Demands

DateDemand
Dec 21"48 hours disengagement"
Dec 22"an additional week... Reconciliation will require mediation"
Dec 23"Disengage for a week, starting yesterday"
Dec 24"Disengage means leave me alone"

But Boundaries Don't Apply to Elle

"@ellewrite thank you for the offer but I would not feel comfortable with that based on my past interactions with you. The ask to disengage has never ended since July 2024. I feel strongly about maintaining it based on how often that disengagement has been breached online and in-person by you." — Cloud, April 17, 2025

The Mediator's Observation

"I'd also like to remind you that both you and Cloud had agreed to stay of Bravespace (along with other stuff). Cloud has upheld her end of the deal, but you have not. Again, I worry you are taking advantage of Cloud by making her agree and then not really holding up your end of the bargain." — zoda

→ See full context: Mediator Burnout

Process Abuse +

Stage 7

Weaponize community safety tools for punishment

The ATL During Mediation

Two days into active mediation, Elle issued an Ask To Leave against Cloud.

"She said she felt unsafe and would like peace for a while. Today she mentioned it was about consequences. 'There had to be consequences. Otherwise, she may have done the same thing, again...'" — zoda (mediator)

Mediator's Assessment

"I've informed her that's not what Ask To Leave is for and have said this feels like an abuse of process. I'm not really sure what more I can do." — zoda

What ATL Is Actually For

"Ask to leave is also only supposed to be used when mediation and conflict resolution is not being handled and de-escalation is needed." — zoda

Why This Is Stage 7

At Stage 7, harm to the opponent becomes a goal in itself. Using a safety tool for "consequences" during active mediation — when Cloud was cooperating — shows the goal had shifted from resolution to punishment.

→ See full context: Mediator Burnout (ATL Abuse)

Rhetorical Conscription +

Stage 4

Use someone as a prop in arguments without consent

At the Tuesday Meeting

"Elle: Agree with you but there is a high need for obvious safety efforts - one of the reason there is such a high need for that - (...) people with a low acceptance of chaotic situations (two women in the room)." — Meeting notes, July 29, 2025

Cloud's annotation: "i feel uncomfortable she is mentioning me"

Cloud's Objection

"Roping me into that meeting discussion is not okay. Speaking for me, putting words in my mouth, twisting my words around, using subtly and implications that I intend something towards a person is not okay with me." — Cloud, July 31, 2025

The Pattern

When building a case, Elle conscripts others as evidence for her position — claiming to speak for women in general while the actual women present object to being used this way.

Recharacterization +

Stage 5

Describe events to allies in ways that maximize sympathy

What WE/Z Actually Said

"I'm the one that asked Elle to not participate in the Tuesday meeting for a month... I literally just asked her to take a break from the meeting. That's it." — WE/Z (first-hand)
"I told her she's a very valuable member of the community and she's 'not in trouble'" — WE/Z
"Elle responded very well and even gave me a hug" — WE/Z

What Elle Told LX

"Elle just told me someone ATLd her... this is honestly a NB sabotaging move of all time to push away people actually doing things" — LX (after talking to Elle)

The Gap

WE/Z's accountLX's understanding
"Take a break from meetings""someone ATLd her"
"You're not in trouble""sabotaging move"
Elle "gave me a hug"(not mentioned)

Victimhood Inversion +

Stage 5

Frame the other party's protective behavior as aggression

Cloud Documents Boundary Violations

"I feel strongly about maintaining it based on how often that disengagement has been breached online and in-person by you." — Cloud

Elle's Reframe

"Also, I find Cloud's obsession with me disturbing and her mistaken belief that I am communicating with her when a) I am not or b) I reflexively say hello to a group that she happens to be in, distressful." — Elle, July 30, 2025
"Learning that she is going behind my back to gossip and complain makes me feel a) hurt by her slander, b) that backbiting and dragging 3rd parties into a negative whisper campaign is detrimental" — Elle, August 1, 2025

The Inversion

  • Cloud maintains a boundary → "obsession"
  • Cloud discusses the conflict → "whisper campaign"
  • Cloud documents violations → "slander"

The Cumulative Effect

Impact on Targets

The escalation tactics documented here have measurable effects:

"I feel like I'm being bullied at this point into accepting and validating yalls experience at the expense of my own. I don't want to come to noisebridge this week or like teach the class I've been working on because of how uncomfortable this is." — Wyatt, July 15, 2024
"I really am not sure what level of involvement I want to continue to have here though if people are going to lightly throw around such strong accusations." — Wyatt, July 15, 2024

Justin disengaged from fundraising-wg.
Wyatt reduced involvement.
Cloud was ATL'd during mediation while cooperating.
A newcomer's motor project was interrupted; she left shortly after.

→ See full list: Contributors Lost

Recognition Patterns

Questions to Ask

SignalQuestion to Ask
Technical → Identity shift "Wait, we were discussing [materials/scheduling]. How did this become about [sexism/racism]?"
Ally recruitment "Is this person speaking for themselves, or on someone else's behalf?"
Disengagement demands "Is this a genuine boundary or a tactic to control communication channels?"
Process invocation "Is this safety tool being used for safety, or for something else?"
Contradictory accounts "Have I heard directly from the people involved, or only interpretations?"

De-escalation Attempts (That Were Rejected)

Throughout the documentation, targets attempted de-escalation:

For full chronological timeline: View Case History