The Intervention Steps Parallel the Diagnostic Steps

The three intervention steps of the mutual learning cycle parallel the three diagnostic steps (Figure 9.1). In short, in the intervention steps, you make public your private reasoning in the diagnostic steps and test whether group members agree with your thinking:

• In step 4, you describe to the group what you privately observed in step 1 that led you to intervene, and you test whether the group agrees with your observations.
• In step 5, you describe the inferences you privately made in step 2 and test whether group members agree with your inferences.
• In step 6, you and the group jointly design the next step.

Assuming group members have decided to change their behavior as a result of your intervention, the cycle begins again, and you observe whether the group members have in fact changed their behavior. You use the mutual learning cycle continually to determine whether there is any group behavior for you to intervene on.

Mutual Learning Behaviors and Mindset Are Embedded in the Intervention Steps

The mutual learning mindset and behaviors are embedded in the intervention steps of the cycle.

• In step 4, as you describe observable behavior, you are using the mutual learning behavior, “use specific examples and agree on what important words mean.”
• In step 5, as you share your inference, you are using the behavior, “test assumptions and inferences.”
• In step 6, when you are jointly designing next steps with group members, you are using that behavior: “jointly design next steps.”
• In each step 4 through 6, you can use the mutual learning behavior, “explain reasoning and intent.” In step 4, you use it to briefly explain why you are intervening. In step 5, you explain how the group member's behavior may be ineffective. In step 6, you explain how the next step you are suggesting can create a more effective process.
• Steps 4 through 6 each have two parts: In the first part, you explain your view, sometimes adding additional reasoning; in the second part, you ask a genuine question to learn what group members think about your view. These two parts make up the mutual learning behavior, “state views and ask genuine questions.”

Because the mutual learning behaviors are a way of operationalizing the mutual learning mindset, when you intervene, you are using the mutual learning values and assumptions:

• You are being transparent about what you are observing, what it means to you, and why you are intervening.
• You are being curious about whether the people you are intervening with see things differently from you. You assume that others have information you might not have and that they may see things that you miss. You assume that differences between your and others' views are opportunities for learning.
• You are enabling group members to make an informed choice about how they proceed and are enabling yourself to do the same.
• You are being accountable to the group by moving forward only after having tested whether your data and inferences are correct. You are asking group members to be accountable by responding to your genuine questions.
• You are intervening with compassion, recognizing that group members have likely created their own unintended consequences. You also assume that you may be contributing to the problem you are observing.

Compassion is a fundamental element in intervening. It's relatively easy for group members to agree to use the behaviors and agree that you can intervene when they act inconsistently with the behaviors. But when you start intervening on their behavior, it can become difficult for group members. They may feel bad about their ineffectiveness or believe they've been put on the spot by your intervention. If you see the group members primarily as having promised to act effectively, then you may view your interventions simply as a way to police their transgressions. If they continue to act ineffectively, you may become frustrated with them for not having the skills that they hired you to help them with. At this point, your interventions shift from being helpful to being hurtful.

Instead, if you see the group members as simply trying to work together without the necessary skill set and mindset, you view your interventions as a way to provide (or teach them) the skill set and mindset they are missing. If they continue to act ineffectively, you can have compassion for them, thinking about the difficulty they are having in learning this approach, just as you learned it.

Begin by Addressing Members by Name

When you begin step 4, you address the group member by name so the group knows to whom you're speaking. It's also helpful to include a phrase such as “I think I heard you say…” or “I think you said…” Using such a phrase sets you up to complete the sentence by quoting what the group member said, which is exactly what you want to do in step 4.

Share Your Observations without Adding Meaning

It's important to share your observations without making an inference or attribution. If you've done step 1 of the cycle well, this won't be a problem. Exhibit 9.2 shows several statements you might make to share your observations with the group and to test whether the group agrees. The easiest way to avoid making an inference or attribution is to repeat the words you heard the group members use. Once you add an inference or attribution, you move beyond observation to interpretation. A video camera simply observes; it doesn't interpret.

It's essential to share your observations without making an inference or attribution, because the observable data are the basis of your intervention. If group members don't see the observable information as you see it, they're unlikely to agree with your inference that is based on the observable information.

Consider Explaining the Intervention You Are About to Make

At any point in steps 4, 5, or 6, you can explain why you're intervening. This is shown in Figure 9.1, with a bracket leading from the three interventions steps pointing to the line, “Explain Reasoning and Intent,” and the opening line, “The reason I'm saying/asking this is…” This is an important part of the mutual learning cycle because it helps group members better understand how you're trying to help them.

At step 4 it's sometimes helpful to briefly explain the intervention you're about to make, before you share the directly observable information. You might begin your intervention by saying, “I want to check whether the conversation is still on track” before describing the observable behavior. This is particularly useful when you're about to describe a complex pattern that requires you to share and test a large number of behaviors. You might say, “I've noticed a pattern of behavior in the group that I think makes it difficult for you to reach consensus. Let me share what I've observed, check it out with you, and see whether you agree.” The statement quickly helps members understand not only why you're intervening but also that you are going to check your understanding with them, which reassures group members. Without this context, you would be describing and testing many behaviors, and group members would be wondering where you're heading.

Be Prepared for Members to Disagree with You

Be prepared for group members to see things differently. The second part of step 4 involves asking the person whether you've accurately captured what the person said. If you ask, “Have I misstated anything you've said?” and the person simply answers yes, then follow up by saying, “What did I miss?” Finding out that you've misheard a group member is not a mistake; on the contrary, it's a success. Using the mutual learning cycle well means learning that others see things differently from the way you do.

After the person tells you what she actually said, you can then decide if it still makes sense to proceed with your intervention. If the corrected version no longer leads you to infer that the person has acted inconsistently with the mutual learning behavior, you stop your intervention. You can say, “I was going to follow up on what I thought you said, but given what was actually said, it's not relevant.”

Sometimes, between the time a person has spoken and the time you start to intervene, you may forget exactly what the person said that led you to intervene. In this situation, you can simply say something like, “Sheryl, would you repeat what you just said? I wanted to follow up on it, but I forget exactly what you said.” Group members don't expect you to be infallible; it's okay to occasionally ask people to repeat themselves.

Be Prepared for Group Members to Disagree with You

Be prepared for group members to see things differently. The second part of step 5 involves asking the person whether your inference is accurate. As with step 4, if you ask, “Is my inference off in any way?” and the person simply answers yes, then follow up by asking what meaning he makes of his statement. Here, too, finding out that a group member's meaning is different from yours is not a mistake; it's a success.

Whether you're testing an inference about content or process affects how you respond if a group member doesn't agree with the inference you have made. If you are testing a content inference—one about the meaning of the person's words—you are likely to accept the person's interpretation if it makes sense to you as a possible explanation. You accept it because it is difficult for someone other than that person to independently know what he or she means.

If, however, you're making a process inference—for example, that a group member is acting inconsistently with a behavior—then you pursue the differing interpretations because as a process expert, you have relevant information and can logically explain the reasoning by which you made your inference. For instance, if in step 5 you share your inference that Amal has stated his view and not asked a genuine question and Amal disagrees with your inference, you might then say, “Amal, would you explain how you see your statement as both making a statement and asking a genuine question so I can understand where we see it differently?” After listening to Amal's explanation, you may explain as a process expert how you think his understanding is incomplete, or you might learn from his explanation that he has, in fact, asked a genuine question and you missed it.

Don't State Your Inference as a Question

Some facilitators make their interventions in the form of a question.⁴ The facilitator might begin an intervention by asking, “Andrew, do you think you're off track?” Author Edgar Schein states that the question format is generally more helpful “because it encourages, even forces, the client to maintain the initiative. If the goal is to help the client to solve his own problem, to own the responsibility, then the question is the best way to communicate that expectation.”⁵

I share Schein's goal of helping the client to own responsibility, and I think his approach can be appropriate for a group in developmental facilitation that is becoming skilled at diagnosing its own behavior. However, I see several potential problems with the question format. First, it requires you to withhold relevant information. If you're thinking, I heard Lois say “X,” which leads me to infer that Lois is not explaining her reasoning, but say to the group “Do you think Lois is explaining her reasoning?” or “What do you think of Lois's comment?” then you're withholding relevant information and easing in. Unless you state the relevant information, you can't check whether the group shares your point of view.

This leads to the second problem: The group can feel set up. If the group answers, “Yes, Lois is sharing her reasoning,” then unless you share your differing view or drop the intervention, you need to continue asking leading questions until the group figures out the supposed answer. Eventually, the group may infer correctly that you are looking for the so-called right answer, but for whatever reason, you're unwilling to share it with them.

Finally, using the question format can send the message that the group needs to take responsibility for its observations but that you don't have to take responsibility for yours. This message is inconsistent with the mutual learning principle that the facilitator is supposed to model effective behavior. Therefore, unless the group understands that your question is designed to help the group learn to diagnose its own behavior, it is more effective to share your inference about the member's behavior and then ask whether the member or others view it differently.

Consider Explaining Why You Are Testing This Inference

As I mentioned in step 4, at any point in steps 4, 5, or 6 you can explain why you are intervening. At step 5, it's sometimes helpful to briefly explain the potential negative consequences that lead you to intervene.

You might say, “The reason I'm testing my inference that group members are raising issues and not following them through to a conclusion is that I think it's a major contributor to the frustration you've said you're feeling about not being able to get consensus.” If a member is making an untested inference or assumption, you might explain by saying, “I'm inferring that you're thinking that the problem needs to be solved using the current staffing available. Is my inference correct? The reason I raise this is because if it turns out that your assumption is incorrect, you may end up designing a solution that makes it more difficult to implement.”

Consider Explaining Your Reason for the Joint Design

In step 6, you might want to explain your reasoning for asking the person to take a next step, including changing his or her behavior to act consistently with the mutual learning behaviors. Generally, by step 6, there may be less need to explain your reasoning, especially if you have done so at steps 4 and/or 5. If you haven't explained in step 5 the potential problem with making untested inferences, you can do so in step 6. You might say, “The reason I'm suggesting this is that by testing your inference, you can be sure you are designing the solution on the basis of accurate information.”

Help People Design Their Next Step If They Need Help

When group members choose to redesign their behavior to take the next step, sometimes they need help. If they've just been introduced to the mutual learning behaviors, they may not know how to test an inference, focus on interests, or combine stating their views with a genuine question. Here, you can help the members redesign their behavior either by coaching them or by modeling the behavior for them. Of course, you can jointly design how you help them by saying something like, “Do you want some help? If so, would you like me to coach you and/or model it for you?”

Even if members redesign their behavior, it may still be inconsistent with the desired behavior. In this situation, you use the mutual learning cycle again, sharing your observation, testing your inference that they are, for instance, still focusing on positions (explaining why you believe that), and asking the members whether they want to try again.

Honor People's Choice to Not Redesign Their Behavior

It's rare that group members choose not to redesign behavior that is inconsistent with the mutual learning behaviors if they've validated your observations and inferences. But what if, for example, a member says no when you ask whether he would be willing to give a specific example? First, you can ask, “Can you say what leads you to say no? I'm asking not because I want to pressure you or put you on the spot, but to find out if there is something that is keeping you from giving an example so people can better understand your concern. The choice is still yours.” If the person is willing to share his concern, you can then explore that concern with the person and the group and ask, “What needs to happen for you to be willing to share an example?” Ultimately, to preserve the group's free choice, you must respect members' decisions about whether they want to change their behavior.

Viewing a reluctant group member as resistant is not a particularly useful frame. It attributes to the person that he is unreasonable—even if you don't understand his reasons—and questions his motives. Instead, if you view a reluctant group member as having unmet needs—needs that you may not fully understand—you can maintain your compassion even if he chooses not to change his behavior.

When to Skip Intervention Steps

So far, I have described how you can use the three intervention steps in their full form. Each of the steps has two parts: In the first part, you state your view, and in the second part, you ask a genuine question. In addition, for any of the steps, you can add an explanation of your view or question.

Although each step of the cycle serves a purpose, sometimes you can skip a step—or part of a step—and still intervene effectively. One reason to skip a step is that using the full mutual learning cycle repeatedly when it's not needed can sound unnatural, awkward, or laborious. However, skipping a step inappropriately can create problems.

Deciding when to skip a step is a judgment call. Steps 4, 5, and 6 are each designed to share your reasoning and check with group members to see if they see the situation differently. Skipping a step creates two risks. First, you increase the chance that group members will not understand part of your reasoning for asking them to change their behavior. Second, if you skip the part of the step that tests for differing views, you end up assuming that members agree with your observation or inference. If your assumption is incorrect, then you move to the next step of the cycle without the members having agreed to a previous step.

There are several conditions under which you sometimes skip a step.

When Not to Skip Intervention Steps

There are also conditions under which it's not helpful to skip an intervention step.

When to Repeat Intervention Steps

I have described the mutual learning cycle as a one-way process in which you move from step 4 to 5 and 6 (sometimes skipping part of a step along the way). However, sometimes you repeat steps. For example, if you're intervening on a pattern of behavior, you may identify several people's observations and test them in step 4. This is essentially using step 4 each time to describe a behavior and test to see if you are seeing it the same way as the group members. After you have identified and tested all the observable information you are using to make your inference, then you move to step 5 and test the several inferences you have made from the observations.

When to Change the Order of Intervention Steps

Sometimes it's more helpful to start at step 5 of the cycle, stating your inference, and then move to step 4, describing the behavior you've observed that led you to make the inference. This is particularly helpful when you have observed a lot of behaviors that together form a pattern that you are intervening on. In this case, you might say, “Marc, you've said a number of things in the last few minutes that led me to think that you're frustrated with the amount of time it's taking the team to make a decision.” At this point, you could skip the testing for agreement and go directly to step 4 and share the behavior you saw that led to this inference. You might say, “You said, ‘This should have been done a month ago,’ ‘Guys, this is a simple decision,’ and ‘I don't have time for this.’” Then you can return to step 5 and ask, “Is my inference correct?” Notice that this combines changing the order of steps and skipping the testing part of step 4. As you start using the mutual learning cycle, you can experiment with skipping steps and changing the order.

If You Launch an Intervention, Land It

Whatever intervention you make, be sure to complete it. I think of an intervention as launching a plane into the air. Until you land it, you haven't completed the trip. Practically, this means several things. First, if you're asking a question or proposing something to the group, make sure you actually ask a question for people to respond to. Second, ensure that you get an answer from all the people you're asking. If you're asking group members whether they agree with your suggestion to move to the next topic, don't take a number of people saying yes as an indication to move ahead. Get a response from everyone.

If, while people are responding to your question, the group starts to discuss a different topic, ask the group to finish responding to your question. You might say, “Drew, I'm not seeing how what you're saying is responding to my question. Did I miss it? If not, can you answer my question about whether you think the group should move to the next topic so we can find out if everyone is ready to move on?”

Sometimes it's difficult to finish an intervention you've begun. Group members start by answering your question, and at some point the conversation shifts without your realizing it. If you're always thinking about whether the group is on the specific topic, including whether it is responding to your specific questions or requests, you increase the chance of landing the interventions you launch and helping the group become more effective.

Avoid Jargon

Some people find the language of mutual learning relatively jargon-free. To others it feels like jargon. For example, I use the term inference throughout the book, but if that sounds like jargon to you or you think it would sound like jargon to the people you are using it with, choose a different term. Instead, of saying, “I am inferring…,” you can say “I'm thinking that…” or “It sounds to me as if…” The purpose is to use words that are clearly understood and that reflect your own voice, not mine.

Use Words with One (Correct) Meaning

To avoid being misinterpreted, use words and phrases that have one meaning—the meaning you want to convey. Certain phrases carry more than one meaning. As part of a union-management cooperative effort, I served as the facilitator in a three-member training team that included union and management representatives. When an employee asked whether a workshop exercise could be approached in several ways, the union representative on our training team said, “I don't care which way you do it.” Later, the employee said he was annoyed that the union representative wasn't interested in answering the question. He misinterpreted the union representative's statement to mean “I'm not interested in your problem.” The representative could have avoided this misinterpretation if he had said, “I don't have a preference for how you conduct the exercise.”

Use Descriptive Words

Use descriptive words if they can be substituted for evaluative words. Words that identify directly observable behavior are more easily validated than judgmental words. The behavior to “use specific examples and agree on what important words mean” is based partly on this principle. It's easier for a group to agree whether a member spoke during the meeting than to agree on whether he was engaged.

Descriptive words are consistent with the facilitator's neutral role. Judgmental words contain some built-in evaluation, implying that the facilitator either approves or disapproves of a behavior or idea. For example, in facilitating a conflict between teachers and administrators, I intervened to summarize the two alternatives posed by members, saying: “You have identified two alternatives. The radical one is X. The other alternative is Y.” As soon as the word radical left my lips, I realized I had made a mistake, even though some group members had used that term. By labeling one alternative with the adjective radical, I loaded it with all the political connotations with which that word has come to be associated. For some group members, radical probably had a positive connotation, while for others it was negative. In any case, by framing the alternative as radical, I added an unnecessary evaluative component. Had I intervened in a purely descriptive way, I would have said, “You described two alternatives. One is X. The other is Y.”

Use Proper Nouns

Use proper nouns or other nouns rather than pronouns. Consider this short exchange between José and a facilitator:

1. JOSé: I talked with Peter about how to handle my conflict with Fred. Peter said he wasn't the kind of person who was particularly good at resolving conflicts and that I should try to solve the problem on my own. Or, I could work something out with Jack, Beth, or Nancy. Frankly, I don't think I can get help from any of them.
2. FACILITATOR: Do you think you can solve the conflict without him?

By using the pronoun “him,” the facilitator fails to clarify whether he is referring to Peter, Fred, or Jack. When pronoun confusion becomes complicated, it begins to sound like the famous Abbott and Costello routine, “Who's on First?” The pronoun that (or this) creates similar confusion. By using individuals' names or their distinct titles, you can avoid confusion.

Use Active Voice

Use active voice unless the identity of the actor isn't clear. Active voice identifies who or what is taking the action; passive voice doesn't. “Sue decided to promote Glen” uses active voice. “It was decided to promote Glen” uses passive voice. When you use active voice, you provide valid information to the group, reduce potential ambiguity, and act consistently with the belief that individuals should be accountable for their actions. As my eighth-grade English teacher repeatedly told our class, “Use active voice. I want to know who is doing what to whom!”

However, it can be helpful to intentionally use passive voice when it's not clear who the person in question is. If two group members are disagreeing about who told the director that the team task couldn't get finished today, you might ask, “When the director was told that the deadline couldn't be met, how did it create a problem?”

Use Words Bestowing Equal Recognition

Use words that give equal recognition to all members and tasks. To be credible with all subgroups, refer to each subgroup in a way that maintains its identity and doesn't subordinate it to other groups. The principle is to refer to each subgroup in terms that reflect its independent identity rather than in terms that use another subgroup as a point of reference. My colleague Peg Carlson and I used to be members of a university department faculty made up mostly of lawyers. At faculty meetings, faculty members would refer to “the lawyer faculty” and “the nonlawyer faculty.” Peg, who like me is an organizational psychologist, commented that until she came to this organization, she had not realized that she was a nonlawyer. She then said that she realized that she was also a nonastronaut, a nonphysicist, and so on.

The point is that when you identify people in terms of a reference group that they don't belong to, you minimize part of their identity. The reference subgroup often consists of those with the highest status or power in the group, even if that reference group doesn't represent a majority. In the example with Peg, if you refer to psychologists as nonlawyers, you are also using the language of the high-power subgroup to define a low-power subgroup. In a subtle way, this can lead others to infer that you are aligning yourself with the high-power subgroup members, which can lead the low-power subgroup members to question your commitment to the full group and, if you're a facilitator, your neutrality.

Use Words That Distinguish Your Facilitative Role

Choose words that distinguish your facilitative role from group members' roles. Refer to the group as you or the group rather than as we or our group. If you're part of the decision-making process (for example, if the group and you are deciding about the meeting process), it's appropriate to use the term we.

Avoid Imperatives (Like This One)

An imperative is a command, such as “make a decision” or, more loosely, “you must make a decision” (“Avoid imperatives” is also an imperative!). One problem with using imperatives is that as a third party, you don't have the authority to give the group commands. As a facilitator, consultant, or coach to the group, you're an advisor, not in a decision-making or commanding role.

Another problem is that, in general, people don't respond well to being told what they must do. They are much more receptive to hearing what the options are and what the consequences are likely to be if they choose each option. This means talking with the group members about cause and effect—what the effect will be if they use or don't use a particular behavior. For example, rather than simply saying, “You need to check out your assumption before acting on it,” you might say something like this: “When you act on your assumption without testing it, you may end up acting on incorrect information, and that can reduce your ability to make a more informed choice. Do you see it differently?” Focusing on cause and effect also helps people see the systemic nature of their actions.

Be Careful with Humor

Avoid humor that puts down or discounts members or that can be misinterpreted. Humor can be a valuable tool for relieving tension in the group, emphasizing a point, and helping members examine their behavior. However, certain types of humor can reduce your effectiveness. Sarcastic humor about a member's ineffective behavior can decrease trust in you because people are likely to interpret the humor as unsupportive. Sarcastic humor can create problems because it requires the listener to interpret your meaning as the reverse of the literal meaning of the words. The sarcastic statement, “I can see there is a high level of trust in this group,” means “there is not a high level of trust in this group.” If members don't detect the sarcasm, they will question your diagnostic skills if they believe there is low trust in the group. Even if they do detect the irony, you haven't modeled transparent communication.