How To Give Orders Like a Man
by Deborah Tannen
The New York Times Magazine, August 28, 1994
A UNIVERSITY PRESIDENT WAS EXPECTING A VISIT FROM a member of the board of trustees. When her secretary buzzed to tell her that the board member had arrived, she left her office and entered the reception area to greet him. Before ushering him into her office, she handed her secretary a sheet of paper and said: “I’ve just finished drafting this letter. Do you think you could type it right away? I’d like to get it out before lunch. And would you please do me a favor and hold all calls while I’m meeting with Mr. Smith?”
When they sat down behind the closed door of her office, Mr. Smith began by telling her that he thought she had spoken inappropriately to her secretary. “Don’t forget,” he said. “You’re the president!”
Putting aside the question of the appropriateness of his admonishing the president on her way of speaking, it is revealing — and representative of many Americans’ assumptions — that the indirect way in which the university president told her secretary what to do struck him as self-deprecating. He took it as evidence that she didn’t think she had the right to make demands of her secretary. He probably thought he was giving her a needed pep talk, bolstering her self-confidence.
I challenge the assumption that talking in an indirect way necessarily reveals powerlessness, lack of self-confidence or anything else about the character of the speaker. Indirectness is a fundamental element in human communication. It is also one of the elements that varies most from one culture to another, and one that can cause confusion and misunderstanding when speakers have different habits with regard to using it. I also want to dispel the assumption that American women tend to be more indirect than American men. Women and men are both indirect, but in addition to differences associated with their backgrounds — regional, ethnic and class — they tend to be indirect in different situations and in different ways.
At work, we need to get others to do things, and we all have different ways of accomplishing this. Any individual’s ways will vary depending on who is being addressed — a boss, a peer or a subordinate. At one extreme are bald commands. At the other are requests so indirect that they don’t sound like requests at all, but are just a statement of need or a description of a situation. People with direct styles of asking others to do things perceive indirect requests — if they perceive them as requests at all — as manipulative. But this is often just a way of blaming others for our discomfort with their styles.
The indirect style is no more manipulative than making a telephone call, asking “Is Rachel there?” and expecting whoever answers the phone to put Rachel on. Only a child is likely to answer “Yes” and continue holding the phone — not out of orneriness but because of inexperience with the conventional meaning of the question. (A mischievous adult might do it to tease.) Those who feel that indirect orders are illogical or manipulative do not recognize the conventional nature of indirect requests.
Issuing orders indirectly can be the prerogative of those in power. Imagine, for example, a master who says “It’s cold in here” and expects a servant to make a move to close a window, while a servant who says the same thing is not likely to see his employer rise to correct the situation and make him more comfortable. Indeed, a Frenchman raised in Brittany tells me that his family never gave bald commands to their servants but always communicated orders in indirect and highly polite ways. This pattern renders less surprising the finding of David Bellinger and Jean Berko Gleason that fathers’ speech to their young children had a higher incidence than mothers’ of both direct imperatives like “Turn the bolt with the wrench” and indirect orders like “The wheel is going to fall off.”
The use of indirectness can hardly be understood without the cross-cultural perspective. Many Americans find it self-evident that directness is logical and aligned with power while indirectness is akin to dishonesty and reflects subservience. But for speakers raised in most of the world’s cultures, varieties of indirectness are the norm in communication. This is the pattern found by a Japanese sociolinguist, Kunihiko Harada, in his analysis of a conversation he recorded between a Japanese boss and a subordinate.
The markers of superior status were clear. One speaker was a Japanese man in his late 40’s who managed the local branch of a Japanese private school in the United States. His conversational partner was a Japanese-American woman in her early 20’s who worked at the school. By virtue of his job, his age and his native fluency in the language being taught, the man was in the superior position. Yet when he addressed the woman, he frequently used polite language and almost always used indirectness. For example, he had tried and failed to find a photography store that would make a black-and-white print from a color negative for a brochure they were producing. He let her know that he wanted her to take over the task by stating the situation and allowed her to volunteer to do it: (This is a translation of the Japanese conversation.)
On this matter, that, that, on the leaflet? This photo, I’m thinking of changing it to black-and-white and making it clearer. . . . I went to a photo shop and asked them. They said they didn’t do black-and-white. I asked if they knew any place that did. They said they didn’t know. They weren’t very helpful, but anyway, a place must be found, the negative brought to it, the picture developed.
Harada observes, “Given the fact that there are some duties to be performed and that there are two parties present, the subordinate is supposed to assume that those are his or her obligation.” It was precisely because of his higher status that the boss was free to choose whether to speak formally or informally, to assert his power or to play it down and build rapport — an option not available to the subordinate, who would have seemed cheeky if she had chosen a style that enhanced friendliness and closeness.
The same pattern was found by a Chinese sociolinguist, Yuling Pan, in a meeting of officials involved in a neighborhood youth program. All spoke in ways that reflected their place in the hierarchy. A subordinate addressing a superior always spoke in a deferential way, but a superior addressing a subordinate could either be authoritarian, demonstrating his power, or friendly, establishing rapport. The ones in power had the option of choosing which style to use. In this spirit, I have been told by people who prefer their bosses to give orders indirectly that those who issue bald commands must be pretty insecure; otherwise why would they have to bolster their egos by throwing their weight around?
I am not inclined to accept that those who give orders directly are really insecure and powerless, any more than I want to accept that judgment of those who give indirect orders. The conclusion to be drawn is that ways of talking should not be taken as obvious evidence of inner psychological states like insecurity or lack of confidence. Considering the many influences on conversational style, individuals have a wide range of ways of getting things done and expressing their emotional states. Personality characteristics like insecurity cannot be linked to ways of speaking in an automatic, self-evident way.
Those who expect orders to be given indirectly are offended when they come unadorned. One woman said that when her boss gives her instructions, she feels she should click her heels, salute, and say “Yes, boss!” His directions strike her as so imperious as to border on the militaristic. Yet I received a letter from a man telling me that indirect orders were a fundamental part of his military training. He wrote:
Many years ago, when I was in the Navy, I was training to be a radio technician. One class I was in was taught by a chief radioman, a regular Navy man who had been to sea, and who was then in his third hitch. The students, about 20 of us, were fresh out of boot camp, with no sea duty and little knowledge of real Navy life. One day in class the chief said it was hot in the room. The students didn’t react, except perhaps to nod in agreement. The chief repeated himself: “It’s hot in this room.” Again there was no reaction from the students.
Then the chief explained. He wasn’t looking for agreement or discussion from us. When he said that the room was hot, he expected us to do something about it — like opening the window. He tried it one more time, and this time all of us left our workbenches and headed for the windows. We had learned. And we had many opportunities to apply what we had learned.
This letter especially intrigued me because “It’s cold in here” is the standard sentence used by linguists to illustrate an indirect way of getting someone to do something — as I used it earlier. In this example, it is the very obviousness and rigidity of the military hierarchy that makes the statement of a problem sufficient to trigger corrective action on the part of subordinates.
A man who had worked at the Pentagon reinforced the view that the burden of interpretation is on subordinates in the military — and he noticed the difference when he moved to a position in the private sector. He was frustrated when he’d say to his new secretary, for example, “Do we have a list of invitees?” and be told, “I don’t know; we probably do” rather than “I’ll get it for you.” Indeed, he explained, at the Pentagon, such a question would likely be heard as a reproach that the list was not already on his desk.
The suggestion that indirectness is associated with the military must come as a surprise to many. But everyone is indirect, meaning more than is put into words and deriving meaning from words that are never actually said. It’s a matter of where, when and how we each tend to be indirect and look for hidden meanings. But indirectness has a built-in liability. There is a risk that the other will either miss or choose to ignore your meaning.
ON JAN. 13, 1982, A FREEZING COLD, snowy day in Washington, Air Florida Flight 90 took off from National Airport, but could not get the lift it needed to keep climbing. It crashed into a bridge linking Washington to the state of Virginia and plunged into the Potomac. Of the 79 people on board, all but 5 perished, many floundering and drowning in the icy water while horror-stricken bystanders watched helplessly from the river’s edge and millions more watched, aghast, on their television screens. Experts later concluded that the plane had waited too long after de-icing to take off. Fresh buildup of ice on the wings and engine brought the plane down. How could the pilot and co-pilot have made such a blunder? Didn’t at least one of them realize it was dangerous to take off under these conditions?
Charlotte Linde, a linguist at the Institute for Research on Learning in Palo Alto, Calif., has studied the “black box” recordings of cockpit conversations that preceded crashes as well as tape recordings of conversations that took place among crews during flight simulations in which problems were presented. Among the black box conversations she studied was the one between the pilot and co-pilot just before the Air Florida crash. The pilot, it turned out, had little experience flying in icy weather. The co-pilot had a bit more, and it became heartbreakingly clear on analysis that he had tried to warn the pilot, but he did so indirectly.
The co-pilot repeatedly called attention to the bad weather and to ice building up on other planes:
Co-pilot: Look how the ice is just hanging on his, ah, back, back there, see that?
. . .
Co-pilot: See all those icicles on the back there and everything?
Captain: Yeah.
He expressed concern early on about the long waiting time between de-icing:
Co-pilot: Boy, this is a, this is a losing battle here on trying to de-ice those things, it [gives] you a false feeling of security, that’s all that does.
Shortly after they were given clearance to take off, he again expressed concern:
Co-pilot: Let’s check these tops again since we been setting here awhile.
Captain: I think we get to go here in a minute.
When they were about to take off, the co-pilot called attention to the engine instrument readings, which were not normal:
Co-pilot: That don’t seem right, does it? [three-second pause] Ah, that’s not right. . . .
Captain: Yes, it is, there’s 80.
Co-pilot: Naw, I don’t think that’s right. [seven-second pause] Ah, maybe it is.
Captain: Hundred and twenty.
Co-pilot: I don’t know.
The takeoff proceeded, and 37 seconds later the pilot and co-pilot exchanged their last words.
The co-pilot had repeatedly called the pilot’s attention to dangerous conditions but did not directly suggest they abort the takeoff. In Linde’s judgment, he was expressing his concern indirectly, and the captain didn’t pick up on it — with tragic results.
That the co-pilot was trying to warn the captain indirectly is supported by evidence from another airline accident — a relatively minor one — investigated by Linde that also involved the unsuccessful use of indirectness.
On July 9, 1978, Allegheny Airlines Flight 453 was landing at Monroe County Airport in Rochester, when it overran the runway by 728 feet. Everyone survived. This meant that the captain and co-pilot could be interviewed. It turned out that the plane had been flying too fast for a safe landing. The captain should have realized this and flown around a second time, decreasing his speed before trying to land. The captain said he simply had not been aware that he was going too fast. But the co-pilot told interviewers that he “tried to warn the captain in subtle ways, like mentioning the possibility of a tail wind and the slowness of flap extension.” His exact words were recorded in the black box. The crosshatches indicate words deleted by the National Transportation Safety Board and were probably expletives:
Co-pilot: Yeah, it looks like you got a tail wind here.
Captain: Yeah.
[?]: Yeah [it] moves awfully # slow.
Co-pilot: Yeah the # flaps are slower than a #.
Captain: We’ll make it, gonna have to add power.
Co-pilot: I know.
The co-pilot thought the captain would understand that if there was a tail wind, it would result in the plane going too fast, and if the flaps were slow, they would be inadequate to break the speed sufficiently for a safe landing. He thought the captain would then correct for the error by not trying to land. But the captain said he didn’t interpret the co-pilot’s remarks to mean they were going too fast.
Linde believes it is not a coincidence that the people being indirect in these conversations were the co-pilots. In her analyses of flight-crew conversations she found it was typical for the speech of subordinates to be more mitigated — polite, tentative or indirect. She also found that topics broached in a mitigated way were more likely to fail, and that captains were more likely to ignore hints from their crew members than the other way around. These findings are evidence that not only can indirectness and other forms of mitigation be misunderstood, but they are also easier to ignore.
In the Air Florida case, it is doubtful that the captain did not realize what the co-pilot was suggesting when he said, “Let’s check these tops again since we been setting here awhile” (though it seems safe to assume he did not realize the gravity of the co-pilot’s concern). But the indirectness of the co-pilot’s phrasing certainly made it easier for the pilot to ignore it. In this sense, the captain’s response, “I think we get to go here in a minute,” was an indirect way of saying, “I’d rather not.” In view of these patterns, the flight crews of some airlines are now given training to express their concerns, even to superiors, in more direct ways.
The conclusion that people should learn to express themselves more directly has a ring of truth to it — especially for Americans. But direct communication is not necessarily always preferable. If more direct expression is better communication, then the most direct-speaking crews should be the best ones. Linde was surprised to find in her research that crews that used the most mitigated speech were often judged the best crews. As part of the study of talk among cockpit crews in flight simulations, the trainers observed and rated the performances of the simulation crews. The crews they rated top in performance had a higher rate of mitigation than crews they judged to be poor.
This finding seems at odds with the role played by indirectness in the examples of crashes that we just saw. Linde concluded that since every utterance functions on two levels — the referential (what it says) and the relational (what it implies about the speaker’s relationships), crews that attend to the relational level will be better crews. A similar explanation was suggested by Kunihiko Harada. He believes that the secret of successful communication lies not in teaching subordinates to be more direct, but in teaching higher-ups to be more sensitive to indirect meaning. In other words, the crashes resulted not only because the co-pilots tried to alert the captains to danger indirectly but also because the captains were not attuned to the co-pilots’ hints. What made for successful performance among the best crews might have been the ability — or willingness — of listeners to pick up on hints, just as members of families or longstanding couples come to understand each other’s meaning without anyone being particularly explicit.
It is not surprising that a Japanese sociolinguist came up with this explanation; what he described is the Japanese system, by which good communication is believed to take place when meaning is gleaned without being stated directly — or at all.
WHILE AMERICANS BELIEVE THAT “THE SQUEAKY wheel gets the grease” (so it’s best to speak up), the Japanese say, “The nail that sticks out gets hammered back in” (so it’s best to remain silent if you don’t want to be hit on the head). Many Japanese scholars writing in English have tried to explain to bewildered Americans the ethics of a culture in which silence is often given greater value than speech, and ideas are believed to be best communicated without being explicitly stated. Key concepts in Japanese give a flavor of the attitudes toward language that they reveal — and set in relief the strategies that Americans encounter at work when talking to other Americans.
Takie Sugiyama Lebra, a Japanese-born anthropologist, explains that one of the most basic values in Japanese culture is omoiyari, which she translates as “empathy.” Because of omoiyari, it should not be necessary to state one’s meaning explicitly; people should be able to sense each other’s meaning intuitively. Lebra explains that it is typical for a Japanese speaker to let sentences trail off rather than complete them because expressing ideas before knowing how they will be received seems intrusive. “Only an insensitive, uncouth person needs a direct, verbal, complete message,” Lebra says.
Sasshi, the anticipation of another’s message through insightful guesswork, is considered an indication of maturity.
Considering the value placed on direct communication by Americans in general, and especially by American business people, it is easy to imagine that many American readers may scoff at such conversational habits. But the success of Japanese businesses makes it impossible to continue to maintain that there is anything inherently inefficient about such conversational conventions. With indirectness, as with all aspects of conversational style, our own habitual style seems to make sense — seems polite, right and good. The light cast by the habits and assumptions of another culture can help us see our way to the flexibility and respect for other styles that is the only best way of speaking.
