Search Peter’s website:

Empathy in Risk Communication

Everybody agrees that empathy is crucial to risk communication. Vincent Covello, for example, argues that caring/empathy accounts for fifty percent of trust; the other fifty percent, he says, is shared about equally by dedication/commitment, honesty/openness, and competence/expertise. He often quotes an old saying to the effect that people (especially people who are upset) don’t care what you know until they know that you care.

So if you do care, showing you care is obviously crucial.

What isn’t so obvious is how to show you care, how to express your empathy. Advice from a risk communication consultant to “be empathic” or “show you care” is of limited value without specifics about how. Too often it leaves clients thinking they’re supposed to check off some kind of empathy box – say something like “we are all so sorry for your loss” or “we don’t blame you for being upset” and then move on to the topic at hand. Or it just leaves clients rightly wondering what exactly they are supposed to do.

Nor is the consultant’s advice to “just use your intuition” very helpful (or very empathic). Our intuition about empathy is often wrong. We need guidelines.

And the guidelines can be profoundly counterintuitive. When a stakeholder is recounting how upset he or she is about something your company has done or plans to do, it might be tempting to murmur sympathetically, “I know how you feel.” That sounds like it should be an empathic thing to say. Yet every risk communication practitioner soon discovers that it doesn’t feel the least bit empathic to the recipient. The invariable response is an angry, “No you don’t!” Paradoxically, “I can’t imagine how that must feel” makes people feel more understood than “I know how you feel.”

Clearly there must be more to expressing empathy than just telling others you feel their pain. It turns out people are often very proprietary about their pain. They don’t take well to strangers who claim to feel it. It’s theirs!

And yet they don’t take well to strangers who seem oblivious to their pain either, or who deny that they’re feeling it, or who tell them they shouldn’t be feeling it. For example when there’s a terrorist attack, an infectious disease outbreak, or some other crisis, officials sometimes think the empathic thing to do is to reassure the frightened public. “We have the situation under control,” they may announce (whether it’s true or not). “There is no cause for alarm.” But experienced crisis managers know that this sort of empty reassurance isn’t reassuring. Instead of making people feel understood, supported, and cared for, it makes them feel abandoned, alone with their fear. “The situation looks pretty bleak right now” is actually a more empathic piece of crisis communication than “everything is under control.”

Of course you don’t just say the situation looks bleak and stop there. You tell people what you’re doing to protect them, what they can do to protect themselves, how much you hope things will improve, etc. But starting with a worried acknowledgment of how bad the situation looks so far is a huge empathic improvement over starting with a cheery expression of overconfidence and a false claim that there’s nothing to worry about. Even if there really isn’t much to worry about, and your goal is to help people calm down rather than gear up for the emergency, it’s still essential to empathize with their concern, not to trash it. But “I know how unnecessarily upset you are” isn’t the way!

The dilemma of empathic communication, in other words, is finding a middle path between two mistakes: being oblivious to your stakeholders’ feelings and intruding on your stakeholders’ feelings. Much of what has been written about empathy focuses on not being oblivious, so novice risk communicators are likely to think obliviousness is the only problem. They’re likely to suppose that the vague admonition to “Be empathic!” means they should keep pointing out what their stakeholders are feeling (“You’re upset”), that they know what their stakeholders are feeling (“I can tell you’re upset”), and that they appreciate how their stakeholders came to feel that way (“I understand why you’re so upset”). There are worse ways of intruding (“You’re stupid to be upset,” for example) – but this is still intruding … and it thus falls short of the empathic ideal.

So before you go any further in this article, please reconsider your anchoring frame, your mental model of what “being empathic” means. It’s not enough to pay attention to stakeholder feelings. It’s not enough to show we’re paying attention. We have to find ways to do it respectfully and gently.

The Essence of Empathy:
Sort-of Acknowledgment

The essence of empathic risk communication is understanding what your stakeholders are feeling, and then finding a way to sort-of acknowledge what they’re feeling – without trespassing on their emotional property.

I wrote “sort-of acknowledge” instead of just “acknowledge” because you’re dealing with emotional dynamite here. Your acknowledgments have to be gentle, unintrusive. In fact, your acknowledgments may have to be deniable, especially by your stakeholders themselves. People who show you how they feel may really want you to show them that you get it (that you “get them”) – and still need to deny their feelings. For instance, people who are frightened may not want to admit it, and therefore may not want you to notice it too overtly.

It isn’t kind, or wise, or empathic to push your stakeholders to acknowledge the feelings you’re sort-of acknowledging, or even to acknowledge your sort-of acknowledgments.

Empathic risk communication, in other words, aims to get your stakeholders’ feelings “into the room” without making your stakeholders feel exposed or pinned down. You find a way to signal what you think they might be feeling, and to validate that it’s a pretty understandable way to feel. If you’re reading them right, they will feel more understood, better able to bear their feelings, and thus better able to cope with the situation. They may or may not tell you you’re right. They may even tell you you’re wrong, and insist that you back off – which of course you should do. If you were reading them right in the first place, they will nonetheless feel more understood, better able to bear their feelings, and thus better able to cope with the situation.

All this is symmetrical. The relationship itself may not be symmetrical. You may be Goliath to their David, the embodiment of a powerful multinational corporation graciously consenting to listen to their concerns; or you may be David to their Goliath, a lowly corporate supplicant desperately seeking access to their community and a “social license to operate.” (You may be one of these in your mind and the other in theirs.) Either way, empathy is symmetrical. If your stakeholders understand you better, and find ways to sort-of acknowledge what’s going on for you, that too will improve the interaction.

What is it, exactly, that needs to be sort-of acknowledged?

• How your stakeholders feel about the situation – their feelings of anger, fear, unfairness, betrayal, etc. This includes not just the emotions themselves, but also the judgments that intermix with those emotions – their belief that X happened or Y is going to happen, that X was a calamity or Y will be an injustice. It also includes what they wish would happen or fear might happen.
• How your stakeholders feel about you – mistrust, for example, or dislike, or envy. Once again, this includes judgments as well as emotions – their belief that you did X or you’re going to do Y.
• Who your stakeholders imagine you are (both in your role and as a person), and how they imagine you feel about the situation and about them – their sense that you don’t like them or don’t trust them; their sense that you’re a callous corporate lackey, an east coast elitist, or whatever they think you are.

Some of what you’re sort-of acknowledging is objectively true. Maybe the situation really is unfair, or you really have acted in an untrustworthy way, or you really are an east coast elitist. Some of it is partly true. Some of it is neither true nor false; it’s a matter of opinion, more about values than facts.

And some of it is flat-out mistaken. You can’t sort-of acknowledge that you’re an east coast elitist if you’re actually a Midwestern populist; empathy doesn’t mean agreeing with falsehoods. But you can still sort-of acknowledge that it looks that way to them (or if that’s too intrusive, that it might look that way to some people). And you can still sort-of acknowledge the things you have been doing, or have been accused of doing, that might give that impression.

Importantly, some of what you’re sort-of acknowledging is projection. If I’m angry at you, and not comfortable admitting it even to myself, I may project it instead, and thus imagine that you’re angry at me. (And I may try to get you angry at me; projections are often self-fulfilling.) Or if I wish you’d just go away even though I know we have to complete our business together first, I may imagine that you’re trying to get away before our business is done. Or if I’m afraid I’ll say something stupid, I may imagine you’re thinking I’m stupid. I am projecting my feelings, wishes, or fears onto you. So your sort-of acknowledgments may have to be multi-level. You need to sort-of acknowledge the projections, and also what might underlie the projections.

Take the first example. Assume a stakeholder is angry at you and is projecting the anger, imagining that you’re angry at him or her. (Let’s also assume the stakeholder is wrong in this case, although as a rule when stakeholders are angry at you it’s a good bet you’re probably angry back.) To sort-of acknowledge the projection, you might say something like this: “I wonder if I sound angry. I think I may be coming across as angry.” To sort-of acknowledge what underlies the projection, you might say something like this: “A lot of people might get pretty angry about what our company has been doing. I think if I lived in this neighborhood I might feel pretty angry right now.” Here’s what you don’t say: “I have figured you out. You’re angry at me, and that’s making you think I’m angry at you.” You don’t say that partly because you might be wrong, but mostly because even if you’re right it’s far too intrusive a thing to say so overtly.

This all sounds pretty shrinky. It is pretty shrinky. The best guidance I know about how to do empathic risk communication comes from Leston Havens, a Harvard psychiatry professor. His 1986 book Making Contact: Uses of Language in Psychotherapy was written to help psychotherapists communicate empathically with their patients. But it’s as close as you’re going to come to a bible on empathic stakeholder relations. Although I have met Les only a handful of times, my wife and colleague Jody Lanard did a psychiatry residency under his leadership at Cambridge Hospital. What follows is in large measure Sandman’s take on Lanard’s take on Havens’s genius.

Two warnings about this column, before I delve more deeply into empathic risk communication:

• Don’t expect a cookbook. I promise to offer more concrete, practical recommendations than the vague advice to “be empathic” or “show you care.” But I can’t turn empathy into a checklist, and I’m not going to try. In fact, some of what follows is heavy going, especially toward the end of the column. Readers who don’t like abstractions – and shrinky abstractions at that – may want to skim quickly past #7 and #8 below.
• Although much of the thinking behind this column comes from the field of existential and interpersonal psychiatry, I am not implying that risk communicators are in a therapist role vis-à-vis stakeholders-as-patients. Stakeholders don’t come to us for help with their psychological problems; we are neither authorized nor qualified to tinker with their psyches – no more so than vice-versa. But trying to understand other people is fundamental to all relationships, even (maybe especially) relationships that are largely antagonistic. Showing that you’re trying to understand is also fundamental. We can borrow from psychiatry without imagining that we are psychiatrists, and without picturing our stakeholders as patients.

Empathy and the Kinds of
Risk Communication

I routinely distinguish three main kinds of risk communication. I start with the distinction between a risk’s “hazard” (how much harm it’s likely to do) and its “outrage” (how upset it’s likely to make people). Based on this distinction, I categorize risk communication into three tasks:

• When hazard is low and outrage is high, the task is “outrage management” – reassuring excessively upset people about small risks. “Calm down.”
• When hazard is high and outrage is low, the task is “precaution advocacy” – alerting insufficiently upset people about serious risks. “Watch out!”
• When hazard is high and outrage is also high, the task is “crisis communication” – helping appropriately upset people cope with serious risks. “We’ll get through this together.”

For a brief summary of the kinds of risk communication, see “Four Kinds of Risk Communication.” My terminology has changed a little since that 2003 column was written, but the concepts haven’t.

Empathy is important in all three kinds of risk communication, but not necessarily in the same way.

Empathy is obviously crucial to outrage management – when people are upset about a risk even though you’re pretty confident they’re not actually very endangered. Under those circumstances, the toughest part of the risk communication job is to reassure your stakeholders that the technical risk is low. Especially if you’re responsible for causing the risk in the first place and have an economic interest in dissuading them from objecting to it, your stakeholders have good reasons not to believe your reassurances, even when your evidence is strong. In this situation, establishing an emotional connection is incredibly difficult and absolutely essential. Whether you think of it as “showing empathy” or simply as acknowledging people’s grievances and concerns, it’s a prerequisite to progress.

Without an empathic approach, it is virtually impossible to tell people that they are wrong to be upset about some situation or wrong to blame you for it.

The role of empathy in precaution advocacy is quite different. Unlike outrage management, in which people are upset (often about you) and need an emotional bond to help them calm down, precaution advocacy is all about talking to people who are apathetic. Your job is to make the issue interesting. There’s no need to be endlessly expressing your understanding and compassion for people who are barely paying attention.

But you’ll need all the empathy you can muster to figure out how to get them to pay attention. The empathic challenge in precaution advocacy isn’t in how you communicate your fellow-feeling; it’s in how you feel your way into your stakeholders’ state of mind so you can appreciate why your issue isn’t firing them up and what might have the capacity to fire them up. You don’t have to show them you understand their apathy – but you will have to understand it deeply if you’re going to have much of a chance to overcome it.

Sometimes, moreover, what looks like apathy about a risk is really something else entirely. Two possibilities of special relevance to empathic communication:

• Denial can masquerade as apathy. Far from being apathetic, stakeholders may be so upset about a risk that they have tripped a psychological circuit-breaker to protect themselves from unbearable feelings. Terror about cancer, AIDS, or nuclear holocaust, for example, may hide behind denial. When stakeholders are in denial, it is very important not to misperceive them as apathetic.
• Excessive outrage about a precaution can masquerade as insufficient outrage about a risk. Employees who can’t stand the face mask you want them to wear, for example, may claim that they’re not worried about the solvents or the asbestos. This is quite common, and often missed.

It takes empathy to detect these two situations. And it takes empathic communication to address them.

What about crisis communication? In February 2007, a Canadian risk communicator wrote to my website Guestbook, suggesting that in an emergency people have more important priorities than empathy – that they need “leadership, someone in control of the situation telling them what’s going on and what they can do, an army-general-like approach, clear instructions, clear respectful orders.” (See “Are empathy and compassion really what matters in mid-emergency?”) My answer, in part:

There have got to be moments when people are in extremis, know they’re in extremis, have already come to terms with the situation emotionally, and just want practical help – information about what’s going on and instructions or advice about what to do.

When that’s the situation, “empathy” means noticing that that’s the situation and not wasting precious time trying to show you care.

But I think that’s very much the exception. Even in the middle of an emergency, most people need emotional support, not just practical help. And our ability to absorb and respond to practical help often depends on our getting the emotional support up-front. Paramedics and EMTs, for example, are often in situations where seconds count; even so, they find time to murmur words of support to the patient and the patient’s family. Ditto for 9-1-1 operators. Even surgeons are learning to find ways to demonstrate their humanity before the patient goes under.

Narratives by survivors of the 9/11 attack on the World Trade Center frequently emphasize the empathy and compassion shown by firefighters, police, and their fellow survivors as they struggled to find their way to safety. They emphasize the practical help too. It’s not either/or.

The most famous quotation to come out of 9/11 was what New York City Mayor Rudy Giuliani told a reporter who asked how many people had died in the Twin Towers. “The number of casualties,” he said, “will be more than any of us can bear.” Why did his words resonate so deeply? Giuliani was saying, in effect, that the attack was unbearable. He was bearing it, but with difficulty – and he let the difficulty show.

That’s what helped New Yorkers (and the rest of us) bear it too.

Interestingly, Giuliani’s speeches since then have expressed the view that it was his strong, calm leadership that rallied New Yorkers after the attack. I think it was his empathy and compassion. Of course it was important that he didn’t fall apart. But the unexpected blessing was that we could all watch him struggling, successfully, not to fall apart. Watching Giuliani hold it together helped millions of us to hold it together too.

Empathy in a genuine crisis is different from empathy when people are understandably but unnecessarily upset about a small risk. Since the crisis is genuine, presumably you are upset too, not just your stakeholders. So you can go beyond sort-of acknowledging their feelings to showing that you share their feelings – and that you too are struggling to bear it all and carry on. This is what Mayor Giuliani did so magnificently in the crucial period after the 9/11 attacks. In a crisis, empathy should be deployed to help you guide people, to help you help them cope with the crisis – not to reassure them falsely. Crisis managers who imagine that showing empathy means over-reassuring people, “emphasizing the positive” or “calming them down,” are way off the mark. But so are crisis managers who neglect showing empathy in their haste to tell people what to do.

The rest of this column is devoted to specific elements of empathic communication. I’ll start with some that are largely common sense, and proceed to some that are more complicated and more shrinky. I’ll try to include plenty of examples as we go along to balance all the abstract theorizing.

1. Feeling and Attitude: Empathy Isn’t a Strategy

Empathy is first and foremost a feeling. In Salinger’s Catcher in the Rye, Holden Caulfield’s alienation was grounded in his belief that other people “don’t get me.” The Martian verb in Heinlein’s Stranger in a Strange Land was “grok”: “I grok you.”

You actually can feel some of your stakeholders’ pain, and that can help relieve the pain. As Les Havens puts it:

Contagion of every affect [emotion] has been reported by therapists: anxiety, depression, anger, excitement, ecstasy, even affectlessness…. Often if the therapist of a depressed person feels depressed himself, the patient improves. It is as if the therapist has relieved the patient of his despondency. It is a dangerous sign if a worker does not become depressed while caring for a suicidal patient; he may not have come close enough to aid the suffering person.

[Making Contact, pp. 17–18]

Nor is it just therapists who feel spontaneous empathy. Havens references a photograph of people “who twist in unison while observing a pole-vaulter clear the bar.” Similarly, I have noticed that when my risk communication seminars dwell on atomic radiation (and fear of atomic radiation) as an example, many of the men in the audience cross their legs.

Feeling your way into other people’s feelings is a capacity we all have, though of course not all the time. Old couples and young lovers sometimes can complete each other’s sentences; Havens says empathic therapists sometimes can (silently) complete their patients’ sentences.

When I say empathy is a feeling, I don’t mean to imply that it has to be intuitive. Some people really are intuitively empathic; they sense what’s going on for those around them, and they sense how they should respond. Other people’s empathy is more analytic; they assess the cues others are putting out, figure out what those cues are likely to signify, and consciously decide how to respond. Sometimes the intuitively empathic people think theirs is the only way, and resist articles like this one that try to come up with guidelines for being empathic. “If you have to think about it,” some readers may be saying to themselves, “it’s not empathy.”

That may be the case for those of us who are intuitively empathic. They may have little need for guidelines; the guidelines may even get in the way of their intuition. But more often than not my clients have a pretty good understanding of what’s going on for their stakeholders (whether they got there intuitively or analytically) – and still don’t know how to communicate that understanding. I’m writing for them, mostly.

When feeling fails, at minimum empathy is an attitude: You are genuinely trying to get how your stakeholders feel.

Empathy is not a strategy. I am not telling you to fake empathy, and faking empathy is hard anyway. If you think your stakeholders are being foolish or treating you unfairly, it’s probably going to show. If you wish you didn’t have to spend time with them, it’s probably going to show. Above all if you haven’t a clue why they’re doing and saying what they’re doing and saying, and don’t really care to find out, it’s probably going to show.

The recommendations that follow are ways to make sure the empathy shows … if it’s there. You can’t use most of these recommendations when all you’re feeling is hostility toward your stakeholders.

Not that empathic risk communication is reserved for use with stakeholders toward whom you have only the kindest of feelings. You’re usually going to be ambivalent. My clients often say of their stakeholders, “I know just how they feel. I’d feel the same way in their shoes.” Even though they mean it, they still have a tough time hanging onto their empathy when the stakeholders start berating them at a public meeting. Use all the techniques you know to hold onto your empathic attitude. Keep reminding yourself that you’d feel the same way in their shoes. Count to ten. Find a confidant you can rant and rave to about what jerks they are. Write angry letters you never send. And then try to feel your way back into their shoes.

On a bad day when all you’re feeling is hostility toward your stakeholders, don’t imagine that these recommendations are going to help. When you’re feeling partly hostile and partly sympathetic, then some of them ought to help.

So what should you do when all you’re feeling is hostility? Take a break if you can. Get somebody else to take over for a while. Do whatever works for you when you need to recover your equanimity – and your empathy. Then these recommendations will help again.

2. Candor and Humanity: Being Real

When all you’re feeling is hostility, you’re not going to be able to act like an empathic risk communicator. When your feelings are mixed, when empathy and hostility coexist, consider the possibility that you should let some of the negative stuff show.

Psychotherapists don’t share all their feelings with their patients, and risk communicators shouldn’t share all their feelings with their stakeholders. But it’s important to share enough of what you’re feeling to come across as real. It is very, very hard to communicate empathically if you’re not also being real.

Les Havens teaches his psychiatry students that getting angry at patients is occasionally the most empathic thing they can do. Losing control is never the right thing to do, but deciding to let some anger show sometimes is. One time when a patient was making racist and anti-Semitic comments in a group session, Havens leaned toward the patient and said, with deliberate anger in his voice: “That makes me really angry. If you don’t stop that right now I’m going to leave the room.” This helped the patient regain control over his own leaking emotions. And it established that Havens wasn’t an impersonal, impervious therapist-doll, but a real person who was really offended by what the patient was saying.

Similarly, I have occasionally advised clients to go ahead and express some of their anger at their stakeholders. Maybe they’re questioning your competence and your integrity; maybe they’re costing your company or your agency hundreds of hours and millions of dollars. Of course you’re upset. You may very well be as outraged as they are! But while they’re freely expressing their outrage, you’re struggling to suppress yours, to hide it even from yourself. Almost inevitably it leaks out – usually in the form of cold courtesy. You may tell yourself that’s what it means to act professionally, but to your stakeholders you’re probably coming across as uncaring or even passive-aggressive, certainly not as empathic. Letting some of your anger show is very likely an improvement over converting it into frigid, obviously hostile courtesy.

It’s possible to carry “being real” too far – and end up being really unprofessional. But most professionals are overly preoccupied with looking professional, and insufficiently attentive to looking human. It’s rare for the public to end up thinking some risk communicator or crisis manager was too emotional, too involved, too personal. It’s quite common for people to think he or she was too controlled, too calm, too uncaring. So while “too emotional” certainly exists, it’s mostly a theoretical problem. “Too controlled” is by far the likelier problem. And that kind of robotic control is incompatible with empathic risk communication.

Anger may be the hardest emotion to express empathically. That’s why marriage counselors spend so much time working on how to fight fair. Other emotions are less fraught.

But some of my clients have trouble getting themselves to express any emotions. Consider these two stories, both of them about government environmental protection officials, both of them, interestingly, women. In the first example, the official failed to express compassion; in the second, she succeeded in expressing hurt.

• At a public meeting, a mother was testifying about her daughter’s leukemia: how awful it is to watch your child sicken and wonder whether it might be the emissions from the nearby waste treatment plant, and all the experts can say is they don’t know, they’re not sure, there’s no proof. She was weeping, and many in the audience were tearing up … but not the hearing examiner, who remained calm, aloof, unruffled. I’m sure she thought that the leukemia had nothing to do with the waste treatment plant, and I guess she thought that showing compassion for mother and daughter might be tantamount to conceding a connection. So her only reaction came at the end: “Your five minutes are up. Thank you for your input.” You could hear the audience gasp.
• At a different meeting, an agency spokesperson was being given a hard time by angry townspeople. After a few hours of hostile rhetorical questions, she asked for a break. “I need a few minutes alone,” she said. “What’s happening tonight is important and legitimate, and I really want to hear all your criticisms and bring them back to my agency. But it’s hard for me. I don’t want to start crying. That would be terribly unprofessional. I just need some time to pull myself together.” When the hearing resumed after the break, the criticisms continued, but the tone was much more substantive instead of personally hostile.

Do you think the first official was more professional than the second? Maybe so. But the second was a lot more empathic – and the second ran a much better meeting.

On one level, perhaps, being real can be seen as the opposite of being empathic. Havens divides Making Contact into two main sections, “Empathic Language” and “Interpersonal Language.” Empathic statements, he writes, are “credulity operationalized.” The goal is to be “taken in” by the other person, and thereby to understand the other person’s perspective as if from the inside. By contrast, interpersonal statements are “skepticism operationalized: the goal is not to be taken in, not to allow the patient to settle assumptions or projections upon the therapist.” Put differently, empathic statements help the patient – or the stakeholder – feel understood, validated, even held; interpersonal statements help the patient/stakeholder feel like a real person in genuine contact with another real person. They are yin and yang.

But like most people I use the word “empathy” more broadly than Les Havens does. Another book that has had a profound impact on the way Jody and I see risk communication is Martin Buber’s I and Thou. (Havens often cites Buber too.) Buber argues that “I-Thou” relationships are authentic because they treat others as real, whereas “I-It” relationships (which he says are far more common) objectify others for one’s own purposes. Because “I-Thou” relationships are necessarily mutual, they require being real, as well as accepting and validating that the other person is also real.

Much later in this long column I will return to Havens’s useful distinction between empathic statements and interpersonal statements. For now, I want to see them all as part of an empathic, “I-Thou” whole.

You shouldn’t have to choose between candor and empathy, between being yourself and showing you care. Quite the contrary. Even though candid statements about how things seem to you are obviously distinct from empathic statements about how things might seem to your stakeholders, the essence of risk communication is to find ways to combine the two. Your goal as a communicator is very often to change your stakeholders’ minds about something. So empathy alone won’t accomplish your goal; at some point you’re going to need to tell them what you think and why. But candor alone probably won’t accomplish your goal either. If you’re not attuned to their perspectives, they’re profoundly unlikely to accept yours.

A core task for risk communicators is to learn how to disagree empathically.

When the anti-inflammatory drug Vioxx was pulled from the market in October 2004 because of cardiac risk, pharmaceutical company executives immediately started trying to differentiate Vioxx from their other anti-inflammatory products that were still on sale. Here are two quotations, both aimed at protecting other products by disagreeing with the public’s tendency to paint with a broad brush. The second one used empathic language that acknowledged the way many laypeople saw the situation; the first one didn’t. (Which perspective is turning out to be more accurate medically, the public’s or the pharmaceutical industry’s, is a separate issue.)

“It is important to note that the results of clinical studies with one drug in a given class are not necessarily applicable to others in a class.”

– Peter Kim, president of Merck Research Laboratories

“It’s hard to comprehend how two drugs in the same class could be so different. But they are each in a subclass of their own, and those different chemical structures translate into a different safety profile.”

– Mitch Gandelman, Vice President of Pfizer

Here’s another example, showing how an unempathic disagreement can be converted into an empathic disagreement, integrating candor and empathy. In early 2005, when surviving Southeast Asian fishermen started to fish again after the devastating tsunami that had wrecked most of their boats, they had trouble finding customers. Fearing that the fish might have fed on the bodies of people who had drowned, their surviving friends and relatives were reluctant to eat the fish. So officials were at pains to explain that there was no risk of disease from eating the fish. Dr. Lee Jong-wook, then Director-General of the World Health Organization, tried to set a good example. While visiting a fishing village, he told a CNN reporter that the widely held perception that the fish were contaminated by the human corpses washed out to sea was wrong. Dr. Lee said: “Fish is a good source of protein and I am eating fish every day. No problem.”

In a post-tsunami column on “Talking about Dead Bodies,” Jody and I wrote:

It’s that perky “No problem” that sounds so out of touch with people’s normal feelings. How much better if Dr. Lee had said something like this: “I must admit I hesitated before I ate the fish. I couldn’t help wondering what the fish had eaten. But the experts tell me it is safe, and I want to do everything I can to help the local fishing industry recover. So I am eating the fish.”

3. Deflection: You → I → They → Some People → It

One key way to make your acknowledgments less intrusive, to convert them into empathic sort-of acknowledgments, is to deflect what you say away from the stakeholders you’re actually talking to. Deflection is what you’re doing when you say “a lot of people might get angry” instead of “you’re angry.” And deflection is what Jody and I thought Dr. Lee should have tried. What he actually said (“I am eating fish every day. No problem.”) sounded oblivious to people’s fears. “You are afraid to eat the fish” would have felt intrusive. “I must admit I hesitated before I ate the fish” uses deflection to find the empathic middle.

Of course if Dr. Lee felt no hesitation whatsoever about eating the fish, he shouldn’t claim otherwise. In that case, maybe he could have talked about a colleague who hesitated and then overcame the hesitation. Or he could “some people” it: “I sympathize with some people who feel uncomfortable eating the fish. They can’t help wondering if….”

Years ago, Jody taught me what she called the “I – you – it – some people” framework to control the extent of the deflection. Thinking the framework through for this column, I revised it to “you – I – they – some people – it,” in order from most to least intrusive.

Suppose for example you’re in the middle of a controversy over groundwater contamination from your factory. Although the evidence shows a very low probability of any health effects, some of your neighbors insist the contamination is deadly. You suspect that part of what’s going on is an unacknowledged worry about property values, which may be fueling people’s resistance to the reassuring health data. (Of course other things are fueling it too: mistrust of your company, scientific uncertainty, etc.) You want to raise the property values issue without being too intrusive. Here are some of your options:

Undeflected:	You	“You’re not really worried about health! You’re afraid your property values might be affected.”
Deflected:	I	“I was in a situation like this when I lived near an industrial park. What worried me even more than the health effects was the possibility that my property values might be affected.”
More deflected:	They	“One of your neighbors was talking with me last week about this situation, and the thing that worried him the most was the possibility of an effect on his property values.”
Still more deflected:	Some people	“Some people in a situation like this would probably be worried about their property values.”
Most deflected:	It	“It’s possible there could be some concern about property values here.”

Lots of variations and combinations are possible. Consider this example: “I wonder if some people here tonight might also be worried about their property values.” Talking about what “people here tonight” are thinking is pretty intrusive for a public meeting. But making it “some people” reduces the intrusiveness; “I wonder if” and “might also” deflect it further.

Deflection serves two empathic purposes.

First, the more you deflect what you say, the less likely it is to provoke denial. People can more easily accept that it’s possible there could be some concern about property values than that they themselves are actually more worried about property values than health.

Second, deflection makes it easier to keep the issue on the table (or at least in the room) even in the face of denial. If you tell me I’m worried about property values and I tell you I’m not, we’re at loggerheads. It’s hard to go on to talk about property values. But if you say only that there could be some concern about property values, I can say “not for me there isn’t” (or perhaps just say it to myself) and the property values topic is still on the table. It’s best to avoid provoking stakeholders to express their denial. So if you decide to say that some people in a situation like mine would probably be worried about their property values, you should just let the statement hang there. Don’t add “What about you?” – which might easily force me to say, “No, not me.”

Of course deflection isn’t an excuse for dishonesty. If you were never in a situation like this, don’t say you were. If you haven’t talked to a neighbor who was worried about property values, don’t say you have.

Why not just deflect everything you say to the max? Because deflected statements are less clear and less powerful than direct statements. You deflect when you have to, as much as you have to, to avoid intrusiveness. But if a person is already “owning” a viewpoint or a feeling, then intrusiveness is much less of an issue and you can respond or ask questions more directly without much risk of provoking denial. Thus if I have already told you I’m worried about property values, it’s not necessary and not especially empathic to deflect my worry onto “some people.” Better to echo my worry back at me (“I can hear how worried you are about property values”) or ask me to amplify (“Has anything happened yet that made you think property values might be going down?”).

But if you’re sensing that I might be worried about property values even though I’m claiming to be worried about health, then it’s too intrusive to just tell me how you think I feel. If you do, I’m likely to take offense, even (perhaps especially) if you’re right. How dare you accuse me of feeling that way! How arrogant of you to think you know better than I do how I feel! Those reactions will of course distance me all the more from being ready to notice and acknowledge my worries about property values. To avoid pushing me further and further into denial, you need to deflect what you say.

The example I’ve been using, “worried about property values,” is a stand-in for a whole range of content that stakeholders may not want to acknowledge openly – content that empathic risk communicators should therefore sort-of acknowledge in deflected ways. Not surprisingly, most of what needs to be deflected has a strong emotional component.

Anger often needs to be deflected, because people so often deny they are angry. People who are angry at you may angrily (or coldly) deny it if you call them on it too overtly, especially if their anger is masquerading as righteous indignation. “I’m not angry,” they’ll insist, their teeth clenched. “I’m just right!” Some people will deny even their anger at third parties, and excessively intrusive efforts to get them to admit it will only drive it further underground. (Unacknowledged rage at terrorists and at specific ethnic groups, for example, may underlie some unwise counterterrorism policies.)

Again and again in risk controversies, I find that my clients are reluctant to acknowledge their stakeholders’ anger at all. They want to pretend that everything’s fine. On those rare occasions when they do acknowledge the anger, their acknowledgments tend to be too explicit and unempathic, focused on insisting that the anger is misplaced. “There’s no good reason for you to be so angry at our company.” I try to persuade clients to “sort-of acknowledge” the anger, to deflect the anger empathically: “A lot of people have told me this decision makes them feel angry at our company.” But it’s a tough sell. I try to explain that deflection helps people move past their anger – whether they acknowledge it or not – and become more willing to at least consider the company’s position. It’s hard for people to move past their anger while they’re feeling “accused” of being angry. And of course it’s even harder for people to move past their anger when the object of that anger doesn’t even seem to have noticed it yet.

Fear also belongs on the shortlist of emotions that are likely to need deflection. Some people who are afraid want to hide that fact from others; some need to hide it even from themselves. Telling them they’re afraid, and especially telling them they should stop being afraid, will tend to push them more deeply into denial. Once again, deflection can help. “People often tell me they find it frightening to discuss plans for a severe flu pandemic” is likely to feel more empathic than “I know this frightens you.” But despite its lack of deflection, even “I know this frightens you” is more empathic than “Don’t be frightened” or “There’s no reason to be frightened.” From child-rearing to surgery, this truth is firmly established: To help people bear their fear, or to help them begin to let go of their fear (if the fear is excessive and letting go is appropriate), first help them accept that it is okay to be fearful.

When H5N1 bird flu first arrives in a country that hasn’t yet experienced it, poultry sales typically plummet for a while. Most governments have overtly expressed disdain for their citizens’ initial reaction of avoiding chicken. But when Nigeria had its first known human case of avian influenza in early 2007, an official news release expressed respectful empathy for the temporary fears of normal people. It started by gently deflecting those fears onto “other countries”:

[W]e have observed that in other countries experiencing their first human H5N1 cases, there has been widespread fear of poultry and poultry products, with a concomitant drop in consumption and sales. For a short time, that may happen in Nigeria too. It is entirely understandable that the population may be overly worried about all chickens, not just sick chickens.

[Nigeria Avian Influenza Crisis Management Center, Bulletin 29]

Deflection isn’t reserved for just anger and fear. Emotions in the “sadness” family – misery, depression, etc. – are also common responses to risk that people may not be prepared to acknowledge directly. So are emotions in the “ego injury” family – hurt, rejection, resentment, envy. Even positive emotions may need deflection. (Think about what it takes to signal empathically to an emotionally unexpressive friend that you know he or she cares about you.)

And of course we’re not talking about pure emotions here. What needs to be deflected usually has a cognitive component as well as an emotional component. In the example I started with, your stakeholders were perfectly willing to acknowledge that they were worried. What needed deflection was how much of their worry might actually be about property values rather than health.

Deflection is a core skill of empathic risk communication, and it is greatly underutilized. But it’s not for every situation. As I mentioned earlier, when people are well-aware of what they’re feeling, sometimes you can just go ahead and express empathy directly, without deflecting at all. Here is an example.

A few weeks after the attacks on the World Trade Center and the Pentagon, a small number of people started receiving letters laced with weaponized anthrax spores. Several died. People throughout the country became afraid that they too might encounter such a letter. Many were reluctant to open their mail. Although some officials responded with ridicule, most tried to keep their reassurances respectful.

Timothy Paustian, a microbiologist at the University of Wisconsin, quickly updated the anthrax section of his online microbiology textbook. The no-nonsense factual information he provided certainly didn’t ridicule people’s fears about their mail; it simply explained why the fears were probably unfounded. But it fell short of empathizing with fearful readers. Jody revised the new section using risk communication principles, and Paustian posted her revisions. Here is one before-and-after excerpt. Jody’s version addresses “you” the fearful reader without deflection, but nonetheless with empathy:

Before: However, it will be very unlikely that you will receive one of these letters. Think about how many pieces of mail go out and how many people there are. Your chances are very low.

After: You know it’s unlikely that you will receive one of these letters, but you’re still scared. You know how many pieces of mail go out, and how many people there are, but you can’t completely shake that inner worry. You know your chances are very low, but you find yourself reaching cautiously for the envelope, and you feel … just a little nuts. Welcome to the human race.

In each situation you need to figure out which content is on the table already, which content is comfortable enough that you can raise it directly, and which content is too private and unacknowledged, too fraught, to handle so intrusively. For the content that you think is fraught, use deflection to get it into the room.

4. Questioning: “How Does That Make You Feel?”

Questions aren’t always empathic. Sometimes they can feel too intrusive, more like an interrogation than a conversation among equals. The most intrusive questions don’t even feel like questions, but rather like patronizing or disdainful accusations: “Aren’t you really more worried about property values than about health?”

But at a very fundamental level asking people how they feel is more empathic than not caring how they feel or telling them how you think they feel. As a rule, people like to tell their stories. They even like telling their stories to people they don’t like … and sometimes end up liking them more as a result.

I am constantly surprised how reluctant my clients are to ask their stakeholders (especially their unhappy stakeholders) any questions at all. For example, I have sat in on dozens of planning sessions over the years to set the agenda for a public meeting. Invariably I suggest making a handful of phone calls to regular attendees to ask them what they want to talk about this time. My recommendation is seldom followed. Maybe it’s a control thing. A lot of my clients really want to do their own agenda-setting. They don’t mind guessing what their stakeholders want on the agenda – but if they actually ask, they’ll feel bound by the answers they get.

Risk communication practitioners may not like asking questions, fearing that they’ll be stuck with the answers. But if the answers are likely to be honest, getting them onto the table will probably do more good than harm in the long term, even if the answers are sure to be hostile and the interaction is sure to be unpleasant. Asking stakeholders why they mistrust you, for example, or why they oppose your proposal, is a good way to get their objections out into the open. Even if you haven’t got a good response, or they don’t want to hear your response, just ventilating their grievances will reduce their outrage a bit, so the conversation can focus more productively on the disputed extent of the hazard.

The rule of thumb here: If people are upset, you’re usually better off letting them get it off their chest. You’re better off if they’ve had a chance to vent, if you’ve heard what they’ve got to say, and if they know you’ve heard it. Letting people criticize you is itself an empathic thing to do. So find a way to ask the question.

But sometimes your stakeholders aren’t likely to answer your question honestly, especially a question about feelings. They may be denying or projecting some of what they’re feeling. They may be confused about how they feel. Or they may be thoroughly in touch with their feelings but disinclined to share them with you – maybe they think it would be rude or tactically unwise to do so; maybe they think it’s just none of your business. A question that’s likely to get no answer, or an evasive answer, or an inaccurate answer, is probably an unwise question to ask. It’s very likely to force the truth more deeply into hiding, and it’s very unlikely to feel like an empathic interaction.

So think about revising your question to make it more empathic. Here are some guidelines for empathic questioning:

Make sure the question doesn’t presuppose a particular answer. “Aren’t you worried about property values?” presupposes that you are. “Are you worried about property values?” leaves the question open.

Try deflecting the question. “Do you think some people are worried about property values?” is a less intrusive question than “Are you worried about property values?” By asking your stakeholders what they think others might be feeling, rather than what they themselves are feeling, you’re making it safer to give an honest answer. You’re inviting your stakeholders to project their feelings onto others so they can express them without acknowledging them.

Try putting the question in the form of a statement (“I wonder if you’re worried about property values”) or even a deflected statement (“Some people might be worried about property values”). By asking the question indirectly rather than directly, you’re making it both easier to answer (your stakeholders feel less under scrutiny, less on-the-spot) and easier not to answer (your stakeholders can let it slide without giving a false answer or explicitly refusing to answer). Indirect questioning also allows people to use your statement as a jumping-off point for what is really on their minds, if they want to. If you are close to correct, they might start out, “Well, it’s not property values, exactly, but let me tell you about….” A series of indirect questions feels more like a conversation and less like an inquisition than a series of direct questions; through successive approximations it takes you closer and closer to your stakeholders’ actual concerns.

Try to ask open-ended questions, which usually feel less intrusive than specific questions, and therefore tend to elicit more (and more honest) information. “How does that make you feel?” is almost a burlesque of a typical psychotherapist’s question – but it’s actually a pretty good question in some situations, a lot less intrusive than “Are you angry about that?” If your open-ended question might be too vague, you can combine it with deflection: “A lot of people might be angry about that. I wonder how it makes you feel.” (If you stick to the first half, you’re back to putting the question in the form of a statement.)

Think about how much to say about yourself. Revealing some of your own thoughts and feelings is certainly more “I-Thou”; it establishes that you’re a real person too. It can help make questioning feel more empathic – and more an exchange among equals instead of an interrogation. It sets an example and implicitly asks for reciprocity: “I’ll show you mine if you show me yours.” But it might also distort your stakeholders’ responses to your questions. Compliant stakeholders may say they agree with you, even if they don’t. Oppositional stakeholders may say they disagree with you, even if they don’t. Moreover, the focus may shift from your stakeholders’ experiences, opinions, and emotions to yours – and to what they think about yours. It’s a judgment call. Keep rethinking whether there’s too much of you, or too little of you, in the interaction. And bear in mind which error you tend to make more often: inserting yourself too much or keeping yourself (your “self”) too hidden.

Don’t start with the hardest questions. Although telling you how they feel is usually what makes stakeholders feel most understood, most empathized with, questions about feelings are also the most intrusive questions. Fact questions are much safer (unless the facts themselves are embarrassing or controversial). But fact questions that call for a one-word or two-word answer don’t launch much of a conversation. The best questions ask stakeholders to narrate a story: “What did you do when the sirens went off?” “What happened next?” Such questions open the door to an emotional answer without demanding one. Later when the bond between you is stronger, you can consider following up with questions specifically about their feelings.

Give stakeholders permission to be reticent. “If you want to tell me, I’d be grateful to learn more about why you feel that way. But I can understand that you might not want to talk about it right now.” Paradoxically, most people are likelier to talk about loaded topics after you’ve said they might not want to. (This is an example of the seesaw of risk communication.)

Sound like you want to learn. Questioning feels empathic when the questioner gives the impression of honest curiosity – of wanting to know the answer. If your questions feel perfunctory – if you sound like you already know the answer or don’t really care about the answer – your stakeholders won’t experience the questioning as empathic. Less empathic still is if you sound like you’re trying to catch your stakeholders in a contradiction or an admission, if you seem to be collecting ammunition or scoring points rather than trying to understand. Of course the best way to sound like you want to learn from your stakeholders is by actually wanting to learn from your stakeholders. Just as you can’t express real empathy if you don’t feel empathic, you can’t express real curiosity if you’re not curious. Questioning isn’t something you do to check it off on a list. You do it because you want to understand your stakeholders better.

Look for role models, people who know how to ask questions empathically. Jody points out that one of the best exemplars of empathic (and interpersonal) questioning is National Public Radio’s Terry Gross. In the service of her genuine curiosity, and her desire to make interviews come alive for listeners, she does a superb job of asking open-ended questions, converting questions into statements, using deflection, revealing pieces of herself, asking permission before an intrusive question, etc.

5. Listening and Echoing: “I Hear You”

If you’re going to ask questions, you should shut up and listen to the answers. Obvious though that is, it’s not always done. In fact I don’t always do it myself. Far too often I ask a good question, then become impatient with the response I’m getting and interrupt to answer my own question. Pretty dumb. And certainly not empathic.

Listening isn’t the same thing as daydreaming. So practice your active listening skills. To show that you are really listening, nod your head from time to time, keep your face responsive (smile or frown; furrow your brow in effort or raise your eyebrows in surprise), murmur an occasional supportive “uh-huh,” maybe even take a note or two.

As a rule, don’t interrupt. If you don’t understand what the speaker is getting at, it’s important to say so and ask a clarifying question – even if it requires interrupting the flow. But every time you’re tempted to do that, ask yourself first if you’re really trying to “get it” or if you’re actually just trying to prove you get it, or showing off, or reasserting control. Sometimes my clarifying questions genuinely help my clients bring me up to speed. Sometimes they just frustrate my clients and delay the process of getting me briefed.

And when the speaker falls silent, don’t be in too much of a hurry to fill the silence. Allow a few seconds to pass. Ponder what your stakeholder has just told you, and how you might want to respond. See if he or she finds something to add, or if somebody else decides to jump in.

I urge my clients to structure most public meetings around listening to what stakeholders want to say. There are exceptions, meetings where the audience really comes to listen and you’ve got a lot to tell them. But much of the time you wouldn’t be meeting at all if you didn’t have to, either because some law or policy requires it or because some stakeholders are anxious to give you a piece of their mind. People come to such meetings largely in order to vent their thoughts and feelings, and until they have done sufficient venting they are in no mood to listen to you. The underlying purpose of the meeting, then, is to let them vent, let them listen to each other vent, and let them see that you, too, are listening and trying to understand.

The meeting is at least partly a pressure relief valve for stakeholder outrage. Instead of allowing the pressure to build until there’s an explosion, you want to help your stakeholders vent their outrage. That should relieve the pressure a little, for a little while. A good meeting, however, aims at more than just temporary relief. To relieve the pressure for more than a little while, listen to what your stakeholders are saying, learn from it, and figure out what your company or agency needs to do differently.

So an empathic meeting doesn’t start with three hours of official PowerPoint presentations, followed by “public input” from 10 to 11 p.m. just before adjournment. In an empathic meeting structure, you start by asking people why they came. As they voice their questions and concerns, you write them all down on a flipchart. (Better yet, ask somebody else to write them down, so you don’t have to turn your back or dilute your attention.) Then you go back over the list: “This one is on the agenda already. This one we can add to Cindy’s presentation at 7:30. This one I’ll answer right now. This one a lot of people want to respond to, so let’s add an open discussion of that to the agenda. This one we can’t deal with tonight, but if the person who raised it will see me during the break, we can schedule a time to talk about it….”

It’s uncomfortable to adjust your agenda in real time to what the people in the room want to talk about. And you’re entitled to reserve some time for things you want to talk about, things you think people need to know even if they don’t realize it yet. But if a key purpose of the meeting is to listen to people’s outrage, and if no other purpose is likely to get accomplished until you’ve made some progress on that one, then it’s self-defeating to make your stakeholders listen to you before you listen to them.

A strange thing happens when you go into a meeting determined to listen to outraged stakeholders. They start wanting to listen to you. That wasn’t their goal on their way in the door, but after an hour or two of telling you how they feel and what they think, they begin to wonder how you feel and what you think about everything they’ve been telling you. (This is yet another risk communication seesaw. If you weren’t listening so hard, they wouldn’t want you to talk.) “Well!” someone is likely to pronounce. “What’s your reaction to all this?”

The first time you’re asked to talk, it’s probably wise to demur: “There are people who haven’t spoken yet, and I’m learning so much tonight.” (You can’t say this last phrase unless you really mean it; it’ll come out sounding sarcastic and obnoxious.) But eventually there’s a consensus that it’s your turn. Now what do you say?

You echo. As I pointed out at the start of this column, “I know just how you feel!” isn’t an empathic echo. It sounds more like a claim to omniscience than an effort to learn. And it’s intrusive, virtually guaranteeing that your stakeholders will respond by pushing you further away: “You can’t possibly know how we feel!” Even “I hear you” claims too much (and it sounds so Sixties). A far better way to transition from listening to responding is something like this:

Let me see if I’ve heard you right. I’ve been listening hard this evening, and it seems to me that most of the people who have talked are focusing on three main issues. A lot of people really want us to do X, they’re very worried about Y, and they think we were wrong to do Z.

Why is this a better echo than “I know how you feel”? For one thing, it deflects the echo from “you” to “a lot of people.” But even more important than that, it’s tentative, provisional. It tests whether you “get it” instead of claiming you do. That makes it easy for stakeholders to correct any misimpressions. It also makes it easy for stakeholders to revise or reinterpret their earlier comments. They don’t have to say they changed their minds; they can just say you’re a little off. Echoes should almost always have some kind of conditional language:

• “Let me see if I’ve heard you right….”
• “I think some people probably feel that….”
• “I wonder if some of what I’ve heard tonight means that….”

In an angry public meeting, it often makes sense to hold your tongue until your stakeholders have done some venting and are ready to hear from you – then summarize what they’ve been saying in one big tentative echo. In a calmer meeting, or a smaller conversation, you should do your echoing as you go along.

Note that you don’t have to agree with a stakeholder’s viewpoint in order to echo what the stakeholder said. It’s very useful to find things you can agree with, and I’m going to turn to that topic in a minute. But it’s crucial to echo even the things you disagree with. Save your disagreement for later, but don’t imply you agree either. If stakeholders are saying your company is dishonest, your probably don’t want to use an echo like, “Our company is dishonest”; that might imply you think so too. Nor do you want to start an argument: “Our company is not dishonest!” Nor do you want to pretend you didn’t hear what your stakeholders were telling you. “You think our company is dishonest” is a candid, straightforward echo that can’t be misinterpreted. But it’s awfully intrusive. Add some deflection, some conditional language, and some specifics, and you come up with something like this: “It sounds to me like some people here tonight think our company was dishonest when we said….”

Echoing is very empathic, but it’s not as empathic as listening. Don’t interrupt to echo. (I make that mistake a lot.)

6. Agreement: “I Think You’re Right about That”

The notion that agreeing with people is an empathic thing to do isn’t as obvious as it might seem.

Sometimes disagreeing is the empathic thing to do. The essence of empathy is to make other people feel understood and respected (respected in the sense that you have made the effort to “get” their position) – not necessarily agreed with. Facile or insincere agreement is going to come across as facile or insincere, not as empathic. I have a close family member who almost never expresses disagreement directly; when he thinks I’m wrong but doesn’t want to discuss it, he says, “You’re probably right.” I don’t experience that as empathy but almost as dismissal. He doesn’t care enough to argue.

If in fact you disagree with what a stakeholder is telling you, trying to fake agreement will actually get in the way of showing you understand and respect what he or she is saying.

Nonetheless, finding and focusing on areas of agreement is an empathic thing to do.

Of course it won’t help much if the areas of agreement are peripheral. Finding something unimportant to agree about may occasionally be a useful icebreaker in difficult negotiations, but often it comes across as sarcastic and trivializing: “Well, at least I can agree with you that 8 a.m. is too early for our next meeting.” You need to look for areas of agreement that will feel to your stakeholders like significant concessions – which means they may feel to you like significant losses.

Let’s assume your stakeholders are berating you. Maybe it’s a low-hazard, high-outrage risk controversy and they’re angrily (and fearfully) telling you how arrogant and dishonest they think your company or agency has been. Maybe it’s a high-hazard, high-outrage crisis situation and they’re fearfully (and angrily) telling you how badly they think your company or agency has prepared to cope with the crisis. Almost inevitably when other people are giving you a hard time, some of what they say is true; some of it is exaggerated but has at least a germ of truth; and some of it is nonsense – garbage.

When they’re done venting and it’s your turn to respond, which of these three do you tend to focus on? If you’re a normal human being, you focus on the garbage. “That’s garbage!” you explode. And then you give two or three examples of things your stakeholders said that are demonstrably untrue. Whose outrage are you managing when you do that? Your own. Who are you empathizing with? Yourself. To empathize with your stakeholders and address their outrage instead of yours, try to focus your response on the criticisms that have some merit. “Yes, we did screw X up. And we weren’t as candid as we should have been about Y.”

I’m not advising you to agree with criticisms that are false. That won’t work. (And you won’t do it anyway.) But if you listen hard for criticisms you can honestly agree with, odds are you’ll find some. If you want to do empathic risk communication, focus a large part of your response on those.

You shouldn’t let the areas of disagreement go completely unmentioned. If only to establish that you’re not conceding the truth of accusations you think are false, you need to find a way to correct the record (as you see it). It’s useful to have a corner of your website, for example, that addresses accusations you think are false and says why you think so. But at least as big a corner of your website should acknowledge accusations you think are sound. You may even decide to take up some of the points of disagreement in real time, responding to stakeholder comments at a public meeting. But at least as much of your response should be devoted to conceding the truth of some of what your stakeholders have said.

There is little room for empathy in an endless rebuttal. Be particularly careful not to find yourself busily rebutting minor factual errors and exaggerations when you should be conceding the truth of a basic proposition. I worked some years ago with the complaints department of a U.K. railway company that almost made a fetish of these sorts of rebuttals. A passenger who wrote that her train was 90 minutes late would get back a huffy letter documenting that it was only 83 minutes late – with no acknowledgment that 83 minutes is as unacceptable as 90. Nothing empathic about that!

In a typical risk controversy, your stakeholders are going to get a lot of technical things wrong. They’ll probably get most of the process things right. This is a rule with plenty of exceptions, of course. Still, more often than not they’re right (or mostly right) that you have been arrogant and unforthcoming, but wrong (or mostly wrong) that your dimethylmeatloaf emissions are giving the neighbors cancer. You will be tempted to spend all your time insisting that there’s no proof your emissions have had any effect on the neighborhood cancer rate. And to be sure, it’s a point you need to make. But if empathic risk communication is one of your goals, try to start by conceding that you have been arrogant and unforthcoming, or that you have sometimes been arrogant and unforthcoming, or at least that you might sometimes have come across to some people as arrogant and unforthcoming.

7. Kinds of Empathic Statements: A Typology

Havens divides empathic statements (as opposed to interpersonal statements; we’ll get to those next) into four categories: imitative statements, simple empathic statements, complex empathic statements, and extensions.

Imitative Statements

Imitative statements are a close cousin of what I earlier called echoes. Havens writes: “Thus ‘How can I decide?’ might be said to the doubtful person; ‘What hope is there?’ to the depressed one; or ‘Where does one find the courage?’ to fearful ones.” Notice how these statements are made less intrusive. They’re reframed as questions, and they’re deflected. “Where does one find the courage?” is a lot easier to hear than “You sound frightened.” Even so, Havens warns that imitative statements should be bland.

The goal is to comfort by our presence, not to startle by our prescience. Not everyone knows that minds can be read or that only a few people are so creative and unique that their thoughts could surprise anyone. Many people are deeply private; they believe their thought should never be known.…

[Making Contact, p. 28]

Here’s a slightly longer example:

One day Jeanne seemed especially still and distant. I mused aloud, “What is one supposed to do?” To my surprise, she crisply replied, “Right!” and after a long pause, “I don’t know what to do. I never know what to do.” I had put myself in the midst of her uncertainty, verbalized it for her, and shared her desperation by my tone. At the same time, there was no implication that she should know or decide, as many ways of calling attention to her indecision might have suggested.

[Making Contact, p. 31]

Similarly, you can echo your stakeholders’ emotions, their situation as they see it, without suggesting that they should see it or handle it differently. “A Superfund cleanup in the neighborhood – how to bear it?”

On the day I was drafting this section, doctors and parents in Perth, Australia were worried about an upsetting disease cluster. Four very young children had died suddenly from streptococcal illness; three of the four also had seasonal flu. Health Department officials openly (and appropriately) expressed their own frustration and anxiety in the media. Communicable disease director Paul Van Buynder, for example, said, “We are desperately trying to get to the bottom of what’s causing these illnesses.” But many local doctors, inundated with worried parents, didn’t want the parents to be concerned. Consider this typical quotation from the head of the Western Australia Council on General Practice, GP Steve Wilson: “If a child is sitting up, taking fluids, cooing and kaaing and looking fairly well, there’s no cause for concern, because the vast majority of these infections will just be straightforward viruses and not the flu.”

Australian media carried lots of technically correct but unempathic statements like Wilson’s – statements that failed to empathize with the understandable alarm of parents facing the specter of a rapidly progressing fatal illness of mysterious origin. How hard would it have been to say something like this instead?

It is scary to hear about these four sudden deaths of children. Of course it is alarming. How to tell whether a child’s minor symptoms might be the start of a nightmare! How can any parent be sure? Here are the symptoms parents should look for…. If those are the symptoms you’re seeing, get your child to the doctor NOW. But if a child is sitting up, taking fluids….

Instead of invalidating parents’ understandable fears, officials can put themselves into the midst of those fears, and only then offer helpful (and reassuring) guidance. Empathic imitation of the stakeholders’ feelings comes first, setting the stage for differential diagnosis (is it a routine cold or a flu that might morph into this fatal disease?) and reassurance (it’ll usually be the former).

Simple Empathic Statements

The simplest of simple empathic statements aren’t even statements, but gestures, facial expressions, and the like. There are also empathic sounds, of which the psychiatrist’s “hmmm” is an obvious example. Havens comments that you can often tell what school a therapist went to by these sounds; Jody tells me Les is famous among his students for the precise tone of his gentle, inquisitive “hmmm?”

The simplest way to show empathy in words is a single adjective: “Wonderful!” “Frightening!” According to Havens, “When we follow a person’s narrative in an empathic way, we find ourselves automatically producing these exclamations.” You just have to decide to express them rather than suppress them. Accented adjectives like “How awful!” are a little more dangerous. Havens warns that they can sound insincere and patronizing if they’re not spontaneous.

Translations go further, putting the stakeholder’s feelings into words. Once again, deflection is usually called for. “It was terrifying” or “wasn’t it terrifying?” or “Anyone would be terrified” is a lot less intrusive than “You were terrified.” But “you must have been terrified” may be okay, even though it uses “you,” because it validates the feeling rather than accusing the stakeholder of it. “You must have been terrified” pretty clearly implies that anyone would have been. Havens writes: “We are saying it [the feeling] is natural, understandable, and shareable.” Effective translations should also leave the stakeholder free to correct or deny your interpretation, or to confirm it, or to ignore it altogether.

Complex Empathic Statements

Havens’s chapter on complex empathic statements starts with “no one….” statements – “no one understands,” for example, or “no one did the right thing for you.” The “no one” formulation is in some ways a variant on deflection. Thus, “no one understands” postpones identifying who in particular has failed to understand. That’s probably more useful for a therapist whose patient feels mommy and daddy didn’t understand than for a company whose stakeholders feel management doesn’t understand. Don’t use “no one” to refer to your own failures and omissions.

“No wonder” statements don’t have that limitation. “No wonder you’re angry at us” (or, if you need to deflect, “No wonder some people are angry at us”) is a good way to get the anger into the room and validate it at the same time. The validation, of course, makes it less likely that stakeholders will deny the feeling. As Havens writes, “no wonder” is “a denial of denial.”

“God knows….” statements make it possible to say intrusive things without actually intruding. “God knows this community has endured enough already.” It’s God who knows, not you. “God forbid….” statements let you raise unpleasant possibilities while empathizing with your stakeholders’ desire not to consider them. “God forbid, if the chlorine sphere should explode….” These are also ways of expressing thoughts about which your stakeholders are ambivalent. Havens uses the example of a patient who both wants and fears to get out of a bad relationship. See how these three complex empathic statements variously capture the ambivalence:

• “God knows you must want to escape.”
• “God knows you must want to escape.”
• “God forbid you should want to escape.”

In early 2003, when the U.S. government raised the terrorism alert from yellow to orange, the Federal Emergency Management Agency reiterated its preparedness advice for citizens to stockpile some food, water, batteries … and duct tape, to help seal off a room in case of certain types of chemical or bioterrorist attacks. The media reported that a few people had actually sealed off some of their rooms. Homeland Security Secretary Tom Ridge needed to point out that this was a premature overreaction. Notice how he integrated complex empathic statements into his February 14 response:

God forbid, there may come a time when the local authorities or national authorities or someone will tell you that you’ve got to use them but, for the time being, we just don’t want folks sealing up their doors or sealing up their windows. Unfortunately, dealing with that kind of possibility is part of the new reality we live with.

While a few people may have overreacted to FEMA’s duct tape advice, millions responded with derision instead, and “duct tape” became the punch line of thousands of jokes. (Overreaction and derision are both predictable responses to scary precautionary advocacy.) As far as I know, Secretary Ridge didn’t offer an empathic response to the scoffers. He could have said something like this:

Even though there are situations where duct tape could save lives, it sounds so lame, so laughably inadequate. God knows how terrifying it is to think about another 9/11, not to mention things even worse than 9/11, like a widespread chemical or bioterror attack. And no one in the government has anything better to offer than duct tape? Of course duct tape is only one item on a list of recommended precautions. Still, I wish we could give people more and better ways to protect themselves. No wonder there are jokes on Letterman and Leno!

In August 2003, North Carolina had a case of Eastern Equine Encephalitis (EEE) that attracted significant media attention and frightened some people. State Epidemiologist Jeff Engel found empathic ways to resonate with people’s fear, validate people’s fear, educate people’s fear, and moderate people’s fear. Here he is in the August 24, 2003 Fayetteville Observer:

Dr. Jeff Engel, a state epidemiologist with Health and Human Services, said the state has documented “only 12 or 13 human infections since 1964.” The most in one year was three in 1989…. Though human infections are rare, Engel emphasized precautions.

“Fear is appropriate. I mean, my God, here you have a mosquito that can kill,” Engel said. “What we are trying to do through you guys, the media, is use that fear in a positive way. We are trying to get information out there.”

After the story ran, the local Wal-Mart sold out of insect repellant. Engel’s empathic statements generated preparedness, not panic.

Extensions

Extensions echo/imitate what your stakeholders are saying, but they add things your stakeholders aren’t saying but may well be thinking – often by extending what has been said in time or in space. If a patient says “It has been difficult,” for example, Havens writes that the therapist might say “It has been difficult for a long time” or “It is not only difficult at work but at home.” According to Havens, extensions let you probe the limits, extending until your stakeholders indicate you’re wrong. He illustrates this with a short dialogue:

P:     I did want to die.
H:    It may still be what you want.
P:     I don’t know.
H:    The pain continues.
P:     I feel so terrible.
H:    It may not be possible to imagine any time free of it.
P:     I thought if I died it would stop.
H:    That might be the only time.
P:     No, now I feel I will be better.

[Making Contact, p. 68]

Extensions make more of a contribution to the dialogue than Havens’s other sorts of empathic statements. They don’t just echo what the stakeholder is saying or not-quite-saying; they go on from there – in baby steps – to probe what else you think might be going on. Although Havens doesn’t say so, they’re a sort of hybrid between empathic statements and interpersonal statements (see #8 below).

When Havens writes about empathic statements, his focus is on how to get closer to other people’s unacknowledged feelings and how to express that closeness without intruding too far. Empathic statements neither endorse nor dispute what others are feeling, just gently get closer to what they are feeling. In risk communication, this helps get your stakeholders’ feelings closer to the surface and, with luck, “into the room” – but not quite “onto the table,” since that would probably be too intrusive. Getting stakeholders’ feelings into the room is useful partly because it helps your stakeholders feel understood, partly because it helps you address their feelings and the issues that give rise to them, and partly because it helps everyone disentangle stakeholders’ feelings from the substantive issues that the feelings may be distorting.

Sometimes it is possible to go beyond bringing your stakeholders’ feelings into the room; you can actually put them on the table. This is too intrusive for everyday use – but when it’s used appropriately it can be breathtaking. In the wake of the December 2004 tsunami, Malaysia’s New Straits Times published a superb January 17 editorial on the post-tsunami fear of eating locally caught fish. The editorial acknowledged (in a segment I’m not quoting) that eating the fish was only minimally risky, but argued at length that the fear was natural and inevitable. Notice the editorial’s use of deflection; it describes the public’s feelings very explicitly, but framed in terms of how “Everyman” and “people” are bound to be feeling – never “you” the reader:

Primal fear has dealt a double-whammy to the fishing industry. Finally putting out to sea again after repairing their boats and reassembling lives so rudely shattered and washed away by the December tsunamis, they now find no one wants to eat the fish they catch because … well, who knows where they’ve been, what they’ve seen – and eaten? The finger-in-a-fish-belly rumour that cut Penang’s fish sales in half in recent days was too obviously waiting to happen, and too predictable to be true. As with all urban myths and legends, someone was bound to find a way to express, in the simplest and most graphic way imaginable, Everyman’s deepest dread.

Let us not, therefore, dismiss these notions too disdainfully. The revulsion people feel at the very thought of eating an animal that has recently eaten a human being is not superstitious but primal. It is not irrational but sub-rational; it stems from the deepest and darkest recesses of the conscience, where resides, perhaps ironically, what it means to be human…. So this is a scare that must run its course….

[New Straits Times, January 17, 2005]

8. Kinds of Interpersonal Statements: Another Typology

(This next section is the shrinkiest and most complicated part of the column. If you’re already near your breaking point, you may want to skip to #9. Or I should say: Some readers who are already near their breaking point may want to skip to #9.)

Havens identifies three kinds of interpersonal statements: projective statements, counterassumptive statements, and counterprojective statements.

In Havens’s terms, all three are interpersonal rather than empathic because they operationalize skepticism rather than credulity. That is, they assert something different from what your stakeholders are trying to assert, rather than just repeating or clarifying or even extending what your stakeholders are trying to assert. They move the conversation to a different place. They are therefore more real, even more “I-Thou” than empathic statements; they are communications from one authentic person to another.

But in a broader sense interpersonal statements are still empathic, because the focus is still on your stakeholders, not on you or “the objective situation.” It’s a bit like the old joke: “Enough of me talking about myself, now you talk about me.” Empathic statements represent your attempt to understand the way your stakeholders see things. Interpersonal statements represent your attempt to explain your “take” on the way your stakeholders see things.

Empathic statements can take you only so far. To really explore how things look to your stakeholders, you have to put more of yourself on the line.

Projective Statements

Projective statements are the simplest form of interpersonal speech. The word “projection” here has nothing to do with Freudian projection (in which people attribute characteristics from one person or situation to another). A projective statement is simply a declarative sentence. You’re projecting some new content into the dialogue. But the purpose of a projective statement isn’t mostly to convey information. Its purpose, Havens says, is to evoke. “A mark has been placed and the patient is stimulated to add, correct, or erase.” A projective statement is a gentle probe.

Havens illustrates this with a short dialogue. The patient is talking about some past event. “It was a nice day in August,” the therapist says. The patient disagrees: “No, it was raining. I remember mother said it would be clear.” “Then your sister got mad,” the therapist prompts. This time the patient adds rather than disagreeing: “She said mother was crazy.”

As Havens points out, this is how people (empathic people, anyway) usually talk to each other.

People make empathic exclamations, ask questions, and listen in silence to reveries and associations; yet perhaps the most powerful engine of verbal intercourse is the statement of fact or possibility.

[Making Contact, p. 97]

Stakeholders who will quickly grind to a halt if you listen silently, who will get defensive or evasive if you ask questions, may tell you everything you need to know if you probe with projective statements. In Havens’s illustrative dialogue, for example, the projective statement “It was a nice day in August” leads naturally to the response, “No, it was raining. I remember mother said it would be clear.”

Two things are crucial about projective statements. First, as I have already said, they should keep the focus on your stakeholders. The value of your statement is in what it evokes, not what it says. And second, they should be hypothetical, tentative, conditional. You’re probing for a response, not taking over. You’re willing to be wrong, happy to be corrected, content even to have a particular probe ignored (you’ll try another). Your pursuit of truth – your stakeholders’ truth – is a series of successive approximations, like tacking a boat. If you have any doubts about whether this conditionality is clear to your stakeholders, add some explicitly conditional language to your projective statements:

• “Maybe it was a nice day in August.”
• “I’m guessing it was a nice day in August.”
• “It sounds like it might have been a nice day in August.”

Havens summarizes this eloquently: “Projective statements are free offerings meant to be taken, amended, or set aside…. The evocative powers of projective statements are directed both at finding the other and, by means of the hypothetical element, testing what has been discovered.”

Let me reiterate Jody’s excellent advice to listen to Terry Gross interviews on NPR. If you listen to Gross’s program, Fresh Air, listen for projective statements – and the responses they evoke from her interviewees.

Counterassumptive Statements

Your relationship with your stakeholders differs in many ways from a therapist’s relationship with a patient. But there are also similarities. Among the similarities is this one: Just as patients make false assumptions that can get in the way of the therapy, stakeholders often harbor false assumptions that similarly complicate their interaction with you. (Of course you often harbor false assumptions about your stakeholders that also get in the way.) In both cases the assumptions may be unstated, maybe even unexamined. In both cases confronting the assumptions directly will probably do more harm than good. “Counterassumptive statements,” Havens says, “shake assumptions without making them a matter for debate.”

I think this is sometimes the wrong approach for risk communication with stakeholders. In fact, I have argued that it is extremely difficult to correct a misimpression just by pointing out the contrary truth. See for example my discussion of the “Donkey” strategy in “Games Risk Communicators Play.” In our handout on “Anchoring Frame Fundamentals,” similarly, Jody and I argue:

You cannot correct misimpressions by ignoring them. You need to address them explicitly, acknowledging that they are widespread and why they seem convincing. Only then is it useful to explain why they are, surprisingly, mistaken. Take us with you from X to Y. Don’t ignore that we think X and just keep insisting Y-Y-Y-Y.

But when stakeholders’ false assumptions are barely conscious or deeply embarrassing, it may be more empathic to use counterassumptive statements.

One of the most common false assumptions in the patient-therapist relationship is the patient’s assumption that the therapist is incredibly wise and competent, even omniscient. A therapist who wants to shake that assumption might, Havens writes, “make it a point to forget the patient’s name” or “stumble clumsily while showing the patient out” or say something along the lines of “I hope I don’t make a bigger mess of things than your last doctor.” There are less extreme ways to accomplish the same goal. When Jody was seeing patients, she often emphasized regretfully what a weak and uncertain science psychiatry is, how far behind chemistry and physics, basically “in the dark ages.” I often say similarly deprecating things to my clients about risk communication. And I urge my clients to say similarly deprecating things to their stakeholders about, say, toxicology or epidemiology or quantitative risk assessment.

Any (or all) of three things may be going on when an epidemiologist says something self-deprecating about epidemiology at a public meeting:

• For those in the audience who overvalue epidemiology and expect miracles, she is using a counterassumptive approach to gently “shake” people’s mistaken belief that an epi study is likely to come up with a clear answer. Without directly challenging the false assumption and thus provoking either debate or denial, she is recalibrating people’s expectations. She will probably want to combine her interpersonal counterassumptive statements with empathic statements that gently help bring the false assumption into the room: “Communities sometimes think an epi study is sure to give them a definite answer. If only we could! It’s such a natural thing to hope. But….”
• For those in the audience who are contemptuous or suspicious of epidemiology (and may not be saying so), she is conceding that they are partly right. I’ll talk more about this sort of proactive acknowledgment in #9. People are of course likelier to cut her some slack if she acknowledges epidemiology’s limitations than if she seems to be overvaluing it herself. More importantly, they are likelier to notice where epidemiology can be of value.
• For those in the audience who are ambivalent about the strengths and weaknesses of epidemiology, she is riding the risk communication seesaw. Each time she says something favorable about her profession, ambivalent people move toward the negative seat on the seesaw; each time she says something critical, they move toward the positive seat. By oscillating from one to the other, she can slowly help them move toward the fulcrum – that is, toward a realistic appraisal of what epidemiology can and can’t accomplish.

Of course your stakeholders’ false assumptions aren’t just about whether epidemiology can save the day or whether you’re omniscient or incompetent. Your stakeholders may have false assumptions about whether you’re greedy or altruistic, for example, or whether you’re honest or dishonest. Or as I wrote near the start of this column, they may have false assumptions about whether you like them or trust them, or whether you’re a callous corporate lackey or an east coast elitist.

Your stakeholders’ false assumptions may not be about you at all, but about the situation, or even about themselves. Havens writes at length about dealing with patients’ false assumptions about themselves – patients who see themselves as inadequate or fragile or super-successful and want (or at least expect) the therapist to confirm that self-perception. The therapist’s job, he writes, is to “lightly undercut” the expectation. “The person expecting praise should feel a little unsure, and the person expecting to seem foolish should feel mildly praised.”

Counterprojective Statements

Counterprojective statements are arguably a special kind of counterassumptive statements. They aim to “lightly undercut” a specific sort of false assumption, one that started with someone else or some other situation and is now being “projected” onto you or the current situation.

Such psychological projections come up all the time in therapy. In fact, they are a big piece of how therapy works. Patients project onto the therapist conflicts from their past. For example, they may unconsciously expect the therapist to act as their parents once acted. The therapist uses the projection to help the patient work on the past conflicts; the goal is to disentangle the past from the present so the patient can put the past into perspective and move on, instead of reenacting the same plays over and over again.

Something similar happens sometimes in risk communication. People’s interaction with you probably isn’t their first contact with “evil, dishonest” corporations or “bureaucratic, incompetent” government agencies. You remind them of other times in their lives when they felt mistreated by big organizations, and perhaps also by more personal authority figures – including, yes, their parents. They may project onto you their reactions to these past contacts. And so you may have use for counterprojective statements.

Havens’s prototype is when somebody stubs a toe on a piece of furniture. The normal response to the pain of stubbing your toe includes “anger toward the object, blaming it, even wanting to kick the offending object again.” The anger then gets generalized and projected, so anybody standing nearby is likely to be resented. Havens continues:

The treatment of such small, paranoid psychoses is, first, empathy with the pain and then alliance against the object. A friend can be quickly made by kicking the offending object for the injured person…. Just as certain exclamations, “How painful!” or “That must have hurt,” are the prototypic form of empathic speech, so another exclamation, “That damn chair,” is the prototype of counterprojective speech.

[Making Contact, p. 126]

Or, in a therapy example, the patient is projecting onto Havens other people’s failure to provide needed help. He keeps putting the projection back where it belongs:

P:     Why don’t you clarify this? You know me very well. I’m confused.
H:    That’s right.
P:     I can’t do it alone.
H:    Neither your boss nor your girlfriend has clarified things either.
P:     They haven’t.
H:    Everywhere you look, no one helps.
P:     But you’re supposed to.
H:    I suppose your parents were too.
P:     They didn’t.
H:    No wonder you want someone to take their place.

[Making Contact, pp. 130–131]

In my safety communication consulting, I often point out to clients that adult employees started out as children who felt overprotected by their parents: “Don’t play with your feces” … “don’t touch the stove” … “don’t ride your tricycle in the street” … on and on to “don’t major in humanities” and “don’t marry that jerk.” Outrage at (parental) precautions is at least as fundamental in people’s psyches as outrage at risks. On the job, many employees project their childhood intolerance of parental protectiveness onto the safety manager, whose warnings are all too reminiscent of the mommy they thought they’d finally escaped.

The simple counterprojective solution is to remind employees where some of their irritation comes from: “You’ve been overprotected a lot.” Since that’s too intrusive, deflect it: “Sometimes people on the shop floor tell me I remind them of their parents, always warning them about some risk.” Or merge your counterprojective purpose with some of the empathic approaches discussed in the previous section: “Here comes that meddling safety manager [you] again! I’m always sounding like somebody’s mommy – be careful about this, be careful about that. God forbid I should leave you alone to get the job done!”

Though he classes counterprojective statements as “interpersonal” rather than “empathic,” Havens notes that counterprojective statements are also empathic, “because they involve the expression of the patient’s strongest feelings.” But counterprojective statements go on to help reattribute those feelings to their correct objects, rather than letting them stay misattributed to you.

I do a fair amount of work with mining and oil companies in developing countries – countries with a long history of colonial exploitation. When a western multinational corporation arrives in an African village, it is a safe bet that many stakeholders will project their feelings about colonialism onto the company. Of course the projection may be accurate; the company may in fact be yet another colonial exploiter. But assume for the moment that it isn’t. Assume that the company intends to behave responsibly and responsively, that it is capable of behaving responsibly and responsively, that it will share the profits from its resource development activities with the people whose resources are to be developed, that the village is really going to end up better off than if the resources weren’t developed (or at least better off than if they were developed by a different company). So the projection is inaccurate.

But the projection is nonetheless inevitable. To succeed in establishing a good relationship with the village and securing a “social license to operate,” the company will need to voice the loathing of colonialism that is animating the village and contaminating the interaction. It will need to say, not “That damn chair!” but something equivalent to “That damn colonialism!” It will need to make counterprojective statements.

But that’s not all it will need to do. The company will also need to acknowledge the ways in which it really does resemble the colonial exploiters of the past. I want to turn to that next.

9. Proactive Acknowledgment: “Some Things You Should Know about Me”

I have urged mining and oil companies in the situation described above to come as close as they can make themselves come to saying something like this:

For centuries people who looked a lot like me have stolen resources from people who looked a lot like you. So when we come here today asking your permission to let us drill, you’d have to be a fool not to feel a lot of anger and a lot of suspicion. Are we going to do it to you again? And I have to admit a painful truth: If we could do it to you again, there’s a good chance we would. A lot of people think that’s the nature of capitalism, to do whatever the company can get away with. I hope I wouldn’t be a part of anything like that, but there’s no reason why you should trust that I wouldn’t – and I can’t even claim that my company wouldn’t, if it could.

What’s important here is that whether we want to or not, we can’t act like colonial exploiters anymore. You’re too powerful – powerful in your own right and powerful in your allies, all the activists and NGOs and socially responsible shareholder groups that are watching us like a hawk. So the question isn’t whether or not we’re going to exploit you again – and get away with it again. We can’t get away with it anymore. The question is whether or not you’re going to forgive us for the history of past exploitation and let us do business with you.

There are still pros and cons to letting us come into your village, of course. It’s still debatable whether the benefit to your people will outweigh the harm. Even with a company that’s trying to be socially responsible and a community that’s empowered, resource development is a dirty and disruptive business. Some things haven’t changed: We come, we harvest your resources, we make money, we leave. Even though some of the control and some of the profits have moved from us to you, it can still feel a lot like colonial exploitation.

This is more than a counterprojective statement. It acknowledges that there is truth to the projection.

When Havens writes about interpersonal statements, he’s focusing on ways to address another person’s incomplete or inaccurate or poorly understood perceptions without directly confronting them, without overtly disagreeing. Unlike empathic statements – which echo, amplify, clarify, or extend what the other person is feeling and thereby help bring it “into the room” – interpersonal statements contribute another perspective: yours. Counterprojective statements in particular actually dispute what your stakeholders are feeling (and projecting onto you), though gently and indirectly so as not to launch a debate.

But what do you do when your stakeholders’ perceptions are accurate? Especially when they’re negative perceptions about you or your organization, I don’t think it’s enough just to bring those perceptions into the room. And it’s certainly not right to dispute them. You owe your stakeholders confirmation.

If they’re speaking their minds, no problem – then the empathic response is agreement (#6) or at least partial agreement. But often your stakeholders won’t speak their minds. They rightly (or somewhat rightly) see you as a colonial exploiter, but they’re too courteous or too embarrassed or too cautious or too strategic to say so. Then empathic risk communication requires you to say so for them.

This isn’t in Havens’s typology. For want of a better term, let’s call it “proactive acknowledgment.”

Suppose for example your state health department has withheld information about a local bird flu outbreak. You were pretty sure it was a low-pathogenic strain, not the dreaded high-path H5N1, but until you knew for sure you didn’t want to reveal the outbreak at all. Now it’s come out. Sure enough, it’s a low-path strain just as you expected. But even so, you can tell that people are feeling mistrustful, upset that your agency sat on the news, and worried that if it had turned out to be a high-path strain the delay could have been deadly. They’re not expressing their concerns overtly, but you can sense the ill-feeling.

We’ve talked about a variety of empathic ways to get that feeling into the room – empathic statements along the lines of “Some people might feel we withheld potentially important information.” But such statements leave unstated the fact that “some people” would be right! I think that fact deserves to be acknowledged. Honesty requires that it be acknowledged. Empathy requires that it be acknowledged. And your own self-interest requires that it be acknowledged – partly so you can more readily apologize, explain why you did what you did, and talk about what you might do differently next time; and partly so your stakeholders can more readily put their feelings and your behavior into perspective and begin the process of moving on.

It’s not all-or-nothing, of course. Your agency had its reasons for withholding the information; you may even feel that your reasons justified what you did (though I would tend to disagree). Since there are two sides to the issue, proactive acknowledgment doesn’t necessarily mean prostrating yourself in a total mea culpa. You might say something like this:

When we first learned of the outbreak, our agency made the debatable decision to wait to announce it until we could firmly establish which bird flu strain was involved. There are probably some people here today who feel that was the wrong decision. Some people inside our agency felt that way too; it wasn’t an easy decision. Even though we were pretty confident it was a low-path strain, if we’d turned out wrong we would have lost valuable warning time for nearby poultry farms. And being completely candid is usually the best policy.

But we were worried that announcing the outbreak without knowing the strain could start a shockwave of fear and end up having huge, harmful, and unjustified effects on poultry markets. So we waited. We turned out right this time – it was a low-path strain, with no human health implications. Still, by deciding to wait we forfeited some of the public trust we have earned in years past, and for that we apologize. And here’s what we plan to do next time a situation like this arises….

This example embodies many principles of risk communication and crisis communication – sharing dilemmas, acknowledging opinion diversity, apologizing for misbehaviors. From the perspective of empathy, what’s important about the example is that it doesn’t just bring stakeholders’ mistrust and disapproval into the room. It validates those reactions and expresses partial agreement with them.

My clients generally find proactive acknowledgment profoundly counterintuitive, almost incomprehensible. Albeit reluctantly, they get that it’s important to acknowledge the truth (or partial truth) of accusations that people are actually leveling against them, when in fact those accusations are true (or partially true). The only alternative would be a dishonest denial, and they know that’s unwise as well as wrong. And after a short struggle they get that it is useful to respond empathically to people’s unarticulated feelings, to bring those feelings into the room. The only alternative would be to let the feelings stay underground, where they’re likely to distort everything else that’s happening in the organization’s stakeholder relations. But they rarely get that it can make sense to plead guilty before they’re accused. Isn’t that doing their detractors’ dirty work for them?

And yet the principle that underlies proactive acknowledgment is one of the best-established principles in all of social science. It goes back to research in the 1950s by Hovland and others into one-sided versus two-sided argumentation. (As art rather than science, it goes back at least to Aristotle.) In a nutshell: When an audience is uninterested and uninformed, and likely to remain uninterested and uninformed, one-sided arguments are more persuasive than two-sided arguments. But when an audience is aware of information that supports the other side, or is interested enough to acquire such information later, then two-sided arguments are more persuasive.

In other words, you should plead guilty before you’re accused when you’re talking to people who are likely to hear you accused sooner or later – and especially when you’re talking to people who are already accusing you in their minds. When stakeholders are thinking but not saying something negative about you, a lot of their energy is inevitably focused there – energy that could be better used elsewhere. They keep thinking it, muttering it under their breath, no matter whether they’re trying to keep themselves from saying it or trying to find the courage to say it. When you acknowledge it proactively – say it for them – you free up that energy.

Once again, you’re not necessarily (and not usually) just plain pleading guilty. At worst you’re probably pleading guilty with extenuating circumstances. More likely still, you’re acknowledging a few senses in which you’re guilty, while clinging to your plea of overall innocence. Even lawyers, for whom one-sided communication comes most naturally, understand the value of admitting the other side’s best arguments in their summation – the jury’s already thinking about those arguments, and they’ll loom all the larger if they’re not conceded and put into context. And lawyers understand the value of mentioning the other side’s best arguments in their opening statement as well – the jury’s surely going to hear those arguments soon, and they’ll hurt less if they’ve already been previewed and put into context.

Yes, there’s a heritage of colonialism and your mining company might still act like a colonial oppressor if it still could, but the shoe’s on the other foot now. Yes, it wasn’t transparent to suppress news of that low-pathogenic bird flu outbreak and you can understand why people might mistrust your agency because of it, but you really didn’t want to provoke an unjustified international boycott of your state’s poultry.

Sometimes, of course, you’re completely in the wrong, and what’s called for really is a proactive total mea culpa. It would take something like saintliness to decide to own up proactively to misdeeds no one is ever likely to know about unless you come clean. But all it takes is wisdom to confess proactively before they inevitably catch you or when they already know you’re guilty. Parents know (and so do wise children) that proactive acknowledgment in such cases provokes forgiveness and reduces punishment.

But most of the time there is truth on both sides. Acknowledging the part of the truth that’s not on your side doesn’t come naturally. But it shouldn’t be that much of a stretch when you sense that your stakeholders already have that part of the truth in mind anyway and you realize that it will loom far larger if it isn’t acknowledged than if it is … or when you predict that your stakeholders are bound to find out eventually and you realize that they will feel totally betrayed if you weren’t the first to tell them.

I did a consultation today – literally in the middle of final editorial changes in this column – with an oil company whose newly hired CEO was involved in a big accident at his previous company. The major question on the floor was what the new CEO should say about the accident. I argued that he should raise the topic proactively, assuming it to be in stakeholders’ minds – and that once it’s under discussion he should never be the one to change the subject. And I argued that he should concede proactively, without waiting to be asked, that he feels responsible. Of course he can’t depict himself as legally responsible; he doesn’t believe he is, and he’s involved in litigation where he’ll be claiming he isn’t. But he needs to tell people what he has learned, what he would do differently if he had it to do again, and what he will do differently in his new job. Above all, he needs to confirm people’s unspoken belief that a tragic accident on his watch is a blot on his record.

Even when you are mostly in the right, and even when your stakeholders are mostly on your side, there are still points of divergence. I sometimes call these the “yes buts.” (The term as applied to risk communication originated with Caron Chess.) When my clients tell me their stakeholders are likely to support a particular position they’re taking, I ask them whether there are any aspects of the position that their stakeholders will see as negatives. Those are the “yes buts.” Unexpressed “yes buts” can often keep a plan from going forward, even when most people are mostly in favor.

If stakeholders are raising the “yes buts” on their own, fine; acknowledge them. But if for some reason stakeholders are harboring “yes buts” they’re not raising, I urge my clients to raise the “yes buts” themselves. Most importantly, I urge my clients to proactively acknowledge the validity of the “yes buts” that are valid. If you’re doing empathic risk communication, “Here are the downsides of our proposal….” is the sort of thing you should find yourself saying often. A client that has proposed a major new development in a forested area is currently preparing a public summary of its proposal. Of course the summary will explain how development will provide jobs and help the economy; of course it will cover the “conservation easements” that aim to balance the parts to be developed by guaranteeing that much of what’s left will remain forest forever. Of course the summary will explain why my client thinks its proposal is a good deal for the public. For the sake of empathic risk communication, the summary should also lay out the counterarguments – not just the ones my client can rebut, but also the ones it must simply concede are real disadvantages of what it wants to do.

In any future influenza pandemic, one of the most readily available strategies to slow the spread of the disease will be frequent and thorough hand-washing. This sort of hygiene advice will inevitably sound painfully inadequate, which it is – and in some cases (on the bus, for instance) difficult to implement as well. To maximize the impact of their hygiene recommendations, pandemic communicators will need to acknowledge the “yes buts” that people may be feeling (often without saying so, sometimes without even realizing it consciously):

• “I know this doesn’t sound like much of a response to the most serious national health threat in decades, but….”
• “At the risk of sounding like a kindergarten teacher or even your mother, let me suggest….”
• “I realize that it’s hard to do in some situations, but whenever you can….”
• “Here is an obvious but often-ignored infection-fighting strategy we try to drum into new medical students….”

If your stakeholders are ambivalent, as they often are, proactive acknowledgment also makes sense in terms of the risk communication seesaw. In talking with clients and prospective clients, for example, I often make a point of referring to myself as “expensive” or sometimes “exorbitantly expensive.” They have to know what I charge. If they’re privately aghast at my hourly rate, my proactive acknowledgment gets it into the room without forcing them to plead poverty or express disapproval – and that makes the rate a little less objectionable. Maybe it will help them feel more comfortable telling me my price is too high for them; maybe then they will counteroffer, or at the very least tell me “no thanks” instead of just disappearing. And if they’re ambivalent about my hourly rate, my proactive acknowledgment that it’s high increases the likelihood that they’ll say to themselves, “yeah, but maybe he’s worth it.”

Similarly, every time I write a long column (like this one) I make sure to mention at least once how damn long the column is. I’ve already decided not to keep it short; I want to say everything I’ve got to say, and I’m more committed to being thorough than to being user-friendly. But at least I can admit it proactively. It’s not like you won’t notice if I don’t mention it.

Okay. Let’s summarize:

• If it’s negative and it’s false and your stakeholders are accusing you of it – find a time and place to correct the record but don’t focus on it too much; searching for points of agreement (#6) is more useful.
• If it’s negative and it’s false and your stakeholders are thinking about it but not raising it overtly – use counterassumptive and counterprojective statements (#8) to try to correct the error without confronting it, and use deflected empathic statements (#3, #7) to get the feelings into the room.
• If it’s negative and it’s true and your stakeholders are accusing you of it – admit it. That’s agreement (#6).
• If it’s negative and it’s true and your stakeholders are bound to find out – admit it. That’s proactive acknowledgment (#9).
• If it’s negative and it’s true and your stakeholders are thinking about it but not raising it overtly – admit it. That’s proactive acknowledgment too (#9).
• If it’s negative and it’s true and your stakeholders will never know unless you tell them – see your clergyperson. (My opinion on ethical dilemmas isn’t a professional opinion; I’m a risk communication expert, not an ethicist.)

There’s one possibility not on this list: If it’s negative and it’s true and you learn that your stakeholders are about to accuse you of it. Opinions differ on this one. There’s a case to be made for just-in-time acknowledgment, preemptive acknowledgment. It’s a bit like turning yourself in moments before the cops knock on your door; maybe you’ll get a little time off for being proactive. But there’s also a case to be made for waiting till the accusation is made, since you know it’s coming, and then doing your acknowledgment thing. Sometimes stakeholders can find preemptive acknowledgment all the more infuriating. It can look like you had a slick, well-rehearsed plan to go public only if you were about to get caught.

One way to make this judgment call is to ask yourself what’s in your own mind. If you’re thinking, “that’ll take the wind out of their sails,” your preemptive intention is more hostile than apologetic. It’s probably better to give your critics the satisfaction of catching you fair and square – wait and let them raise the issue. (When they do, be sure to acknowledge and apologize promptly; there’s nothing to be gained by evading or protesting.) But if you’re thinking, “it’s time to own up and face the music,” then there’s a good chance your acknowledgment will be experienced as proactive, not preemptive.

10. Performatives: “I Hope the Situation Will Improve Soon”

Havens ends Making Contact with a short section on “Performative Language.” The concept of performative statements was introduced by philosopher J.L. Austin, particularly in his influential 1962 book, How To Do Things with Words. The basic concept is obvious once it’s pointed out: Some statements don’t state a claim but rather perform an action. The Wikipedia entry on “performative utterance” has a wonderful list of typical performatives. Here are some of them:

“I now pronounce you man and wife.”
“I christen you….”
“I apologize.”
“I accept your apology.”
“I divorce you, I divorce you, I divorce you.”
“I promise to be there.”
“This meeting is now adjourned.”
“The court is now in session.”
“I sentence you to death.”

Of course performatives don’t automatically happen. Havens quotes Austin’s example of the performative “I appoint you…,” which isn’t valid “when you have already been appointed, or when someone else has been appointed, or when I am not entitled to appoint, or when you are a horse.” Similarly, consider the performative “I apologize.” The apology may be insincere, or sincere but not believed, or believed but not accepted. Apologizing doesn’t necessarily resolve the other person’s grievance. Nonetheless, apologizing isn’t just claiming to apologize. It’s apologizing. It does what it says.

Some performatives are more complicated, because they both state a claim and perform an action. When a baseball umpire calls a runner “Out!” he is simultaneously judging that the ball reached the base ahead of the runner and causing the runner to go back to the dugout. And when a beloved spouse murmurs “I love you” she is simultaneously claiming to have a particular feeling and arousing that feeling (in herself and her husband both). Similarly, as Havens explains, “I admire you” is on one level a claim that may be true or false; but on another level it’s a performative, because it tends to make you feel admired. “I admire how generous you are” is more complicated still. It states two claims, that you are generous and that I admire generosity; as a performative it makes you likelier to feel generous and likelier to feel that your generosity is a good thing. “I hate how self-sacrificing you are” appraises the same behaviors very differently, and will have very different performative impacts.

Performatives are powerful interpersonal statements: They establish an “I” that is appointing, apologizing, umpiring, loving, or admiring. Performatives can also be powerfully empathic statements, especially when enacted on behalf of stakeholders. Probably the most powerful empathic use of performatives is to express hopes, wishes, regrets, fears, and worries.

“I hope the situation will improve soon” is a performative statement because the statement doesn’t just claim as a matter of fact that if you looked inside my mind you would find hope. It expresses the hope. It performs hopefulness. The statement itself hopes. (By contrast, the statement “I eat” doesn’t eat; it merely claims that I do.)

I am amazed at how seldom my clients tell their stakeholders their hopes and wishes. Even if your hopes and wishes are quite different from those of your stakeholders, expressing them is still humanizing; it establishes that you are a person who has hopes and wishes. If they are hopes and wishes your stakeholders share, then expressing them builds commonality. And if they are hopes and wishes your stakeholders share but cannot easily acknowledge, expressing them is profoundly empathic. “I wish we were back in a pre-9/11 world.” “I hope someday our enemies will agree to live and let live.” “If only we could just nuke ’em all.” (There are some hopes and wishes that your stakeholders daren’t acknowledge and you daren’t acknowledge either. One or more of these three may well be in that category.)

Achievable hopes are well worth expressing. In fact, every time you are tempted to make a promise, consider whether you’d be wiser to downgrade it to a hope. And every time you are tempted to say you’re “confident” about some good outcome (getting the fire under control quickly, for example), consider whether it would be more accurate, more credible, more sustainable, and more empathic to change “confident” to “hopeful.”

But expressing unachievable wishes is even more valuable than expressing achievable hopes. “I wish I could give you a definite answer” is more empathic than “I simply cannot give you a definite answer.” “I wish that damn accident had never happened” is more empathic than “I know you wish the accident had never happened.” “I wish we could take every conceivable precaution” is more empathic than “It’s foolish to imagine we could take every conceivable precaution” (and more honest than “we’re taking every conceivable precaution”).

Regrets are unachievable wishes about the past. And so expressing shared regrets is also extremely empathic. The actual word “regret” has lost its empathic power from overuse in legal contexts; “management regrets to inform you…” doesn’t sound regretful or empathic anymore. “I wish we had realized…” does. So does “If only we had realized….”

Finally, fears and worries are the flip side of hopes; they are what we hope won’t happen. Not surprisingly, then, expressing shared fears and worries is a powerful way to show empathy, especially if they are feelings that your stakeholders have trouble expressing for themselves. Choose your words carefully. If you say you’re “concerned” when stakeholders are terrified, it’s going to sound minimizing rather than empathic. If you say you’re “terrified” when stakeholders are merely concerned, you may frighten them more than you intended, or they may start worrying empathically about the state of your emotions. Needless to say, the first error is a lot more common than the second. But in a frightening situation, even official expressions of concern are more empathic than the conventional (and conventionally false) insistence that “There is no reason for concern.” Bottom line: When you are about to claim that there is no reason for concern, consider acknowledging instead that you (too) are concerned. Then ask yourself if you are actually worried. If you are, try to force yourself to say “worried.”

Performatives can be deflected and qualified as much or as little as you think best, depending on how “deniable” they need to be to avoid over-intrusiveness. Thus “I admire how generous you are” can morph into “Everyone admires that kind of generosity” or “It’s admirable how generous that was.” “I wish I could give you a more definite answer” can become “Many people wish there were a more definite answer” or “We all wish there were a more definite answer” or “I think we all probably wish there were a more definite answer.”

Some Examples from Earth

Empathic statements, counterassumptive statements, counterprojective statements, performative statements – all that obscure jargon from psychiatry and even philosophy. (Yes, glad you noticed. This is itself an empathic statement.) What does it mean down here on earth?

I hope you will collect your own down-to-earth examples – from media stories of risk controversies; from Terry Gross interviews; from your own interactions with family, friends, colleagues, and stakeholders. Figure out what makes them empathic. If you like thinking in categories, figure out what category they’re in – or make up a new category if you need one.

If you find (or create) a really good example of empathic risk communication and you don’t mind sharing it, please email it to me at peter@psandman.com. If I get a good collection, I’ll add an appendix of “Some Examples from Earth.”

---

Afterword by Jody Lanard

In the wondrous age of Google searches and tags, I would tag this column Risk Communication, Empathy, Havens, Buber, Sandman (although Buber is mentioned only briefly).

What is a reductionist seat-of-the-pants risk communication consultant like Peter Sandman doing on a list with psychiatrist Leston Havens and philosopher Martin Buber? What are Havens and Buber doing in a risk communication publication at all?

In the genealogy of risk communication, other important influences are far more obvious than Havens and Buber:

• The psychometric paradigm of risk perception, developed by Slovic et al., which underlies Sandman’s most famous formulation, “Risk = Hazard + Outrage.”
• The profoundly influential work of Kahneman and Tversky on the framing of decisions, prospect theory, and judgment under uncertainty (heuristics and biases).

But at its deepest roots, our philosophy of empathic risk communication springs from the ideas of Leston Havens and Martin Buber. In Peter’s column, he tries for the first time to bring Les Haven’s work to life for students and practitioners of risk communication.

It almost feels like an oxymoron, to try to operationalize empathy. Les Havens was brave and brilliant enough to do this in his 1986 masterpiece, Making Contact: Uses of Language in Psychotherapy. A Boston Globe reviewer called the book a “grammar of empathy.”

Les Havens was one of my most beloved teachers in the years surrounding the book’s publication. When I introduced Peter to Les’s work in 1985, I cautioned him, as my fellow students and I always cautioned each other, that Les’s teachings could not, should not, be turned into a cookbook or checklist for communicating.

Nevertheless, it often seemed like novices were doing exactly that, when they practiced using some of Havens’s difficult approaches. This sometimes led to scorn from more advanced practitioners or from closer “disciples” of Dr. Havens – a very unempathic response to abashed students taking brave baby steps!

In the 1960s at Princeton University, Peter passionately studied the work of Martin Buber with his beloved professor, Buber scholar Malcolm Diamond, and with philosopher Walter Kaufmann, who later translated Buber’s masterpiece, I and Thou. Another Buber scholar, Robert E. Wood, discussed the “fraught” issue of trying to analyze and use a complex structure such as Buber’s:

The attempt at analysis, though indeed fraught with the very serious danger of destroying the immediacy of experience, likewise puts us in a position of being able eventually to experience the immediate more profoundly…'. [T]here comes a time when, through persistence in analysis, along with attention to immediacy, immediacy itself is deepened. The musical score, as Buber observes, is not the same as the musical performance; and yet we may add that an understanding of the score may eventually enhance the appreciation of the performance.

— Robert E. Wood, in Martin Buber’s Ontology

This applies as well to the work of Leston Havens, and to future attempts of risk communicators to embrace this work.

I think it was courageous of Peter to take the risk of trying to operationalize empathy, to build guidelines but not a cookbook around the difficult thinking of Havens and Buber. I hope you will take the comparable risk: the risk of trying new approaches that might feel stilted or forced at first, even if they feel stilted or forced at first. I believe it will help your efforts at Making Contact.

Copyright © 2007 by Peter M. Sandman

Contact information page:    Peter M. Sandman