Conflict and repair·Say It Better reference library

Conflict Resolution Techniques

Separating people from problems, negotiating interests rather than positions, and choosing a conflict mode deliberately.

Originator: Roger Fisher and William Ury; Kenneth W. Thomas and Ralph H. Kilmann, 1974 and 1981

An overview of established models. Say It Better documents them; we did not develop them.

01

What it is

This page covers a family of approaches rather than a single method, because conflict resolution does not have one canonical framework. Two models do most of the work, and they answer different questions.

Principled negotiation, from Fisher and Ury, answers what to do once you are in the conversation: separate the people from the problem, focus on interests rather than positions, invent options before deciding, and use objective criteria so the outcome is not just a measure of who pushed harder.

The Thomas-Kilmann conflict modes answer a question that comes earlier: how hard should you push at all? It describes five modes — competing, collaborating, compromising, avoiding and accommodating — and the point is that none is correct. Each fits a different combination of how much the issue matters and how much the relationship matters.

Put together, they cover the two things people get wrong: entering every conflict the same way, and then handling the conversation badly once they are in it.

02

Where it comes from

Roger Fisher and William Ury developed principled negotiation at the Harvard Negotiation Project, part of Harvard Law School. Getting to Yes: Negotiating Agreement Without Giving In was published in 1981 and has since become the standard reference in negotiation teaching worldwide. The Harvard Program on Negotiation continues to publish and teach in this tradition.

Kenneth W. Thomas and Ralph H. Kilmann published the Thomas-Kilmann Conflict Mode Instrument in 1974. It built on the earlier Blake-Mouton managerial grid, which had used the same two dimensions — concern for production and concern for people — to describe management styles. The TKI reframed those dimensions as assertiveness and cooperativeness and produced five conflict-handling modes. It remains one of the most widely used instruments in management training.

A third strand runs alongside both: John Gottman's observational research on couples, which is the most cited work on how conflict escalates in intimate relationships. It is narrower in scope than the other two — it is about romantic partnerships — but it is where most of the practical guidance about de-escalation comes from.

Say It Better uses all three in our conflict guidance. None of them is ours.

03

What problem it addresses

Two problems, and they reinforce each other.

The first is positional bargaining. Each person stakes out a demand and the conversation becomes a contest of will. It is slow, it produces outcomes neither party is happy with, and it damages the relationship — because the only way to move is to push harder, which means the other person has to be pushed.

Fisher and Ury's observation was that positions are almost always hiding interests. I want the report by Friday is a position. I need to brief the board on Monday is an interest. Interests can usually be met in more than one way; positions cannot, which is why arguing about positions is so unproductive.

The second problem is the choice problem. Most people default to one conflict style — usually avoiding or competing — regardless of the situation. They let the trivial go by without comment and then treat something important the same way, or they fight every point with equal force until people stop raising things with them. The Thomas-Kilmann modes make the choice explicit, which means it can be made deliberately.

Underneath both is a third problem: escalation. Once a conversation becomes a fight, neither person can negotiate well, because the physiological state that comes with conflict makes listening and problem-solving markedly harder. That is not stubbornness. It is the reason de-escalation comes before negotiation rather than after.

04

When to use it

  • A disagreement that keeps resurfacing without resolving.
  • Negotiating anything where the relationship matters beyond this one issue.
  • Deciding whether to raise something at all, and how hard to push if you do.
  • Mediating between two other people — the interests and criteria questions work on both sides.
  • Any conflict that has become about winning rather than about the problem.
  • Before a difficult conversation, to work out what you actually want out of it.
05

When not to use it

  • Where there is coercion, abuse or a serious power imbalance. Negotiation frameworks assume both parties have some ability to walk away, and that assumption does not always hold.
  • When the issue is a safety matter, where the answer is a boundary or an exit rather than a negotiation.
  • In an acute emotional moment. De-escalation comes first; trying to negotiate while both people are flooded does not work.
  • When there is genuinely nothing to negotiate — where the answer is a limit you hold on your own, or a decision that has already been made.
  • As a way of managing someone into agreement. The frameworks are for reaching an outcome, not for engineering one.
06

Step by step

  1. 1

    Separate the people from the problem

    Treat the disagreement as something you are both facing, rather than something between you. This is the single move that makes the rest possible — it stops the conversation being about who is right.

  2. 2

    Identify interests, not positions

    Ask what each side actually needs, underneath what they are asking for. I need to brief the board on Monday can be met by Thursday's draft as easily as by Friday's final report.

  3. 3

    Invent options before evaluating them

    Brainstorm without committing, and keep the generation and the judgement separate. Evaluating while generating shuts down the range of options, which is the point of the step.

  4. 4

    Use objective criteria

    Market rates, precedent, policy, expert opinion, an independent assessment. Criteria matter because they let the outcome be decided on something other than who pushed harder.

  5. 5

    Choose a mode deliberately (Thomas-Kilmann)

    Competing when the issue is important and the relationship is not. Collaborating when both matter. Compromising when time is short. Avoiding when the issue is trivial, or when you need to cool down first. Accommodating when it matters more to them than to you.

  6. 6

    De-escalate first if the temperature is up

    Lower the heat before trying to solve anything. A timed break with a stated return is a de-escalation move, not an exit — and it is the condition under which the conversation can work.

  7. 7

    Agree what happens next

    A decision, a date, or an agreed pause with a time attached. A conflict conversation that ends without one of those three has not ended.

07

Real-world examples

Two teams are deadlocked over a deadline.

Instead of arguing dates, ask what each needs: one needs the code freeze to hold, the other needs to hit a launch window. Options appear once the interests are named — a partial freeze, a staged release, a scope cut.

Siblings disagreeing over care for an ageing parent.

Positions: she should move closer. Interests: safety, cost, being able to visit weekly, not being the one who carries it. Objective criteria: the doctor's assessment, the budget, the waiting list. The criteria let the decision stop being about who cares most.

A colleague raises the same complaint every week and nothing changes.

This is a mode problem. Accommodating has been the default and it is not working. Choosing to compete on one specific point — or to avoid the conversation entirely and address it through a process — is a deliberate choice rather than a drift.

An argument that has gone past the point of usefulness.

I am going to stop for a second, because I do not think either of us is hearing the other right now. Can we take twenty minutes and come back to it? De-escalation first; the negotiation can happen afterwards, and it will go better.

08

Example scripts

Before we argue about the answer, can we talk about what each of us needs?

I am not attached to my version — I am attached to the outcome. What would yours look like?

What would a fair way to decide this look like, independent of us?

I do not think either of us is in a state to settle this. Can we come back at seven?

This matters more to you than it does to me, so I will go with your option — but I want to flag the one thing I need.

Let us get the options on the table before we start ruling them out.

These lines show the shape of the pattern. They are examples to adapt, not scripts to repeat word for word.

09

Common mistakes

Treating collaborating as always correct

Collaborating is the most expensive mode — it takes the most time and effort. Using it on a trivial issue wastes everyone's goodwill, and using it on everything is exhausting.

Confusing interests with demands

You can only negotiate interests. If the interest is stated as a non-negotiable requirement, it is a position again and you are back to positional bargaining.

Inventing and evaluating at the same time

Options offered and immediately criticised stop being offered. The generation step only works if judgement is suspended for it.

Using objective criteria selectively

Citing precedent only when it favours you is transparent, and it destroys the value of criteria for the next disagreement.

Skipping de-escalation

Negotiating while both people are flooded does not work, and the attempt usually makes the position harder. The break is not avoidance; it is a precondition.

Assuming bad faith

People often defend a position they did not choose either — one handed to them by a manager, a policy or a deadline. Treating that as stubbornness closes off the interest conversation.

10

Limitations and considerations

  • Principled negotiation assumes a level playing field and a shared interest in reaching agreement. It has very little to say about bad-faith actors, coercion, or a counterpart who simply refuses to engage.
  • A well-known critique of Getting to Yes argues that focusing openly on your interests can be exploited by a counterpart who does not reciprocate. Being reasonable is not always symmetrical.
  • The Thomas-Kilmann modes describe styles, not outcomes. Knowing which mode you are in does not tell you what to say, and the instrument has been criticised for measuring self-reported preference rather than behaviour.
  • Gottman's research is largely observational and based on studies of couples, which makes it most directly applicable to romantic partnerships and less so to workplace or family conflict.
  • Cultural expectations about direct confrontation vary considerably. Several of these techniques assume a fairly direct communication norm and a relatively low cost to disagreeing openly.
  • None of these frameworks addresses what to do when the conflict is structural — a bad policy, an unreasonable workload, a manager who will not change. Sometimes the problem is not the conversation.
12

Sources and further reading

These are the published works behind this page. Where a framework has a named originator, that person or organisation is credited above.

All communication frameworks