Technology in Society 31 (2009) 342–349

Contents lists available at ScienceDirect

Technology in Society journal homepage: www.elsevier.com/locate/techsoc

The politics of airplane production: The emergence of two technological frames in the competition between Boeing and Airbus
Alexander Z. Ibsen*
University of Arizona, Department of Sociology, P.O. Box 210027, Tucson, AZ 85721-0027,United States

a b s t r a c t
Keywords: Airplanes Boeing Airbus Two-party democracy Frames Technological philosophy

Economic models of technological innovation, as well as modern sociological approaches to the study of organizations, predict that two-actor markets will eventually evolve into one dominant technological logic. Why is it, then, that the only two global manufacturers of large commercial airplanes have developed diametrically opposed technological philosophies? Based on secondary historical sources, this article employs a theory of twoparty democracies from political science and the theory of sociotechnical frames to explain why Boeing pilots are allowed ultimate command of their aircraft whereas Airbus confers this authority to the flight computer. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction For anyone who has ever flown on a large airplane owned by an airline based in an affluent country, the chances are almost exactly 50% that the plane will be made by Boeing Commercial Airplanes, and 50% that it will be made by Airbus S.A.S. At the same time, it is 100% certain that it will not be made by anyone else. Most passengers probably cannot tell the difference, nor will they care about the type of aircraft they boarded; indeed, they are more interested in flight comfort, safety, and reliability, and it makes little difference whether the plane was manufactured by Boeing or Airbus. The truth is, however, that there are significant differences in the cockpits of airliners built by the two companies. The Boeing pilot is in full command of the airplane and its flight computer, whereas in an Airbus airplane the computer has ultimate control over the pilot. The differences are both visual and technical, and highlight two very different technological philosophies that have gradually emerged in modern commercial aviation. This paper utilizes the political theory of two-party democracies as well as social constructivist theories of sociotechnical change to explain the emergence of these two radically different technological philosophies. The two systems will be broadly introduced before the theoretical perspectives adopted here are presented. Relying on insider testimonies reproduced in published works, the paper attempts to reverse-engineer the two companies’ trajectories that have led to such divergent technologies in contemporary aviation. Although the different technological philosophies have been widely commented on separately, no attempt has been made to analyze the two strategies within one overarching framework. Consequently, one of the most important high-tech industries has escaped academic attention with regretful loss of important insights. 2. Two technological philosophies With the launch of Airbus’s A-320 family of single-aisle aircraft in 1988, and with Boeing’s introduction of the B-777 widebody plane six years later, both companies divested themselves of decades of mechanical flight control systems in favor of
* Tel.: þ1 520 621 3531; fax: þ1 520 621 9875. E-mail address: ibsen@email.arizona.edu 0160-791X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.techsoc.2009.10.006

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

343

fully computerized fly-by-wire technology. All ensuing types of aircraft from both producers did, and will in the foreseeable future, rely on this type of flight control. Functionally speaking, there are few differences between Boeing and Airbus. They both introduce computer interpretation and transformation of information between the pilot’s commands and the aircraft’s physical reactions. In digital fly-by-wire, the pilot’s motions on handles and levers in the cockpit are transmitted to the primary flight control computers via control surfaces, and are further introduced to the flight actuators only after the computers have assessed the command as compatible with the plane’s design limits as well as its current course in the air. A curious principle of aviation is that systems of aircraft design and flight control must choose a position on a stability/control continuum. Less stable aircraft are more agile and capable of variable performance in the air. On the other hand, demands on pilots increase as the system approaches more control and less stability. With fly-by-wire, a plane is allowed to give up considerable stability in favor of performance and control due to relaxed pilot-work intensity by computers. A NASA historian comments: ‘‘Aeronautical engineers employed computers in flight control systems not because they represented a new technology and were ‘progress for progress’ sake’, but because they were part of a solution to the flight control problem’’ [1, p. 128]. This particular technology was first introduced by NASA in the Apollo program in an exercise module used to prepare astronauts for lunar landings. Thereafter fly-by-wire was used almost exclusively by American commercial and military pilots, culminating in the introduction of the first digital (as opposed to analog) computer system for the F-8 project at the Dryden Flight Research Center in the early 1970s [1, pp. 57–69]. The militaristic growth and development of American fly-by-wire evolved from the availability of U.S. government research funds for such projects. European air travel also had an early introduction to fly-by-wire technology in the Anglo-French Concorde, which was fitted with an analog version of the computer system. On both sides of the Atlantic, therefore, the technology was well-known and pursued by researchers. And while it had a civil origin in Europe, Boeing has from the beginning maintained close connections with military aeronautics [2]. Table 1 provides an overview of the current flight control systems of Boeing and Airbus. Details will not be worked through here, except for those pertaining to the two technological philosophies. As Table 1 illustrates, during normal flight, the computers in both systems interject between pilot and plane actuators to make sure the aircraft remains airborne. This is secured by flight control calculation, which prevents it from stalling, loosing load factor, being short of speed, or banking too strongly. Details differ between the two versions, but the basics are the same. The true difference appears when we move from normal control law to secondary or alternate law. In the Boeing system, there are no longer any restrictions on the pilot’s input whereas some still apply for the Airbus pilot. A Boeing pilot is free to choose this control law at any time during flight; it only automatically activates when multiple systems in an Airbus machine fail. The same is true for the direct control law: both aircraft will under this law completely eliminate the computerized command interjection, and pilots will navigate in the same way they would have steered a fully mechanical planedexcept, the Airbus pilot cannot initiate this control by him/herself whereas a Boeing pilot can. In the jargon of the airline industry, the Airbus version is said to employ ‘hard’ envelope protection, whereas the Boeing uses ‘soft’. In addition to computer differences, the cockpits of the two airliners also look distinctly different. By dispensing with certain mechanical elements, digital fly-by-wire aircraft have no need for the classical instruments and navigation equipment that was needed to maneuver older airplanes. This is obvious in Airbus cockpits in the A-320 family and later models, which are fitted with liquid crystal displays rather than analog control panels; also the aircraft is navigated by one-hand control sticks that look identical to the joysticks used in video games. In contrast, Boeing’s 777 model has maintained the traditional

Table 1 Comparison of the two flight control systems. AIRBUS Normal Law Alternate Law Direct Law Conventional Airplane Load Factor Limitation Pitch Attitude Protection High Angle of Attack Protection Speed Protection Bank Angle Protection Yaw Control BOEING Normal Law Pitch Attitude Protection High Angle of Attack Protection Thrust Asymmetry Compensation Speed Protection Bank Angle Protection Yaw Control Sources: A-320 flight manual. Flight controls. Denver, CO: Jeppesen; 1993. B-777 flight manual. Flight controls. Denver, CO: Jeppesen; 1996. Secondary Law No Protections Direct Law Conventional Airplane Load Factor Limitation Mechanical Back-Up No Computers Functioning, Sidestick Inoperative

Speed Protection Yaw Damping Only

344

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

steering yoke, which looks and even feels like the conventional stick on mechanical aircraft. This is because it includes sophisticated feedback machinery that makes it behave as if the pilot’s input was directly transferred to the actuators. This function is called ‘feel’ in aviation. Therefore, the B-777 yoke is nothing more than a gigantic joystick that pretends to be a conventional steering yoke. Hence, the two manufacturers offer different implementations of the same technology. Airbus makes pilots the handlers of flight computers. This is reflected in Northwest Captain Kenneth Waldrip’s statement, made at the introduction of the A-320 to the American market: ‘‘It’s a dream to flydbut you’d better make sure that the pilot flying it understands that computer’’ [3, p. 1534]. This does not mean that greater computer skill is required from Airbus pilots than from Boeing pilots; it simply means that Airbus pilots need to be aware that the flight control system will not permit them to exercise any command they want. By comparison, if there are warnings, or there is shaking in the yoke’s ‘feel’ system, or other problems may arise, Boeing pilots can decide on and initiate any command freely. At the launch of the B-777, a brochure advertising the new aircraft stated Boeing’s technological philosophy: ‘‘The pilot is the final authority for the operation of the airplane’’ [4]. The statement was, in all likelihood, meant to counter any anxiety on the part of potential purchasers about the new and highly computerized model. Any system of automation, such as digital flight control, needs to balance numerous factors relating to both humans and machines. Fundamentally, the pre-set interpretation of human input by a machine must, in some way, be matched by a nonfixed interpretation of the machine’s behavior by the human. There is an inherent asymmetry in this flow of information: whereas feedback must be enhanced to help the human, the automated technology does not need feedback [5]. In the air, feedback has to take into account not only the interplay between pilot and computer, but also the plane’s actual physical behavior. It is this latter requirementdand its solutiondthat has caused the two companies to chose different paths.

3. A global duopoly After Lockheed Martin pulled out of the airplane production business in the early 1980s and McDonnell Douglas was acquired by Boeing a decade later, only Boeing and Airbus were left to manufacture the largest commercial airplanes, resulting in one of only a few global duopolies. The industry of airplane manufacturing is highly capital intensive due to the great demand for expensive technology and the first-class expertise of the manpower involved, which helps to explain why there are no new market entrants. In markets where perfect substitutes are available, economists assume that rational companies avoid direct price wars and try to use non-price methods to differentiate their products instead [6]. Such a situation is likely to occur in duopolistic markets. When duopolistic production involves technology-intensive manufacturing, economic theory further predicts that the competitors will engage in technological ‘leapfrogging’ by alternately supplanting each others’ inventions [7]. Hence, economic models of duopolies expect only temporary differences in technological philosophiesdessentially only for a brief period after a new innovation has taken place. This has not been true in modern aviation. Rather, aviation technology lasts for decades. For instance, except for some upgrading of software and flight display in the flight deck, the famous B-747 ‘jumbo jet’ continues to rely on technology from the late 1960s. Therefore, manufacturers offer products that incorporate a decades-long spectrum of inventions. It has been documented that Airbus and Boeing respond to technological innovations from competitors with investment that yields price cuts, and vice versa, rather than following suit in technological adoptions [8]. In other words, the ‘leapfrogging’ phenomenon seen in other industries is not found in aviation. Another theory comes from sociology. When organizations are subject to similar kinds of market pressure, New Institutional theory assumes that the broader environment influences contestants equally and simultaneously, with the result that all market actors adopt similar organizational strategies [9]. The market for commercial jets is a good example: although passengers are the ones who purchase air travels, the consumption of aircraft is really done by the airlines. With a few exceptions, mainly in the low-cost carriers, airlines operate with fleets of planes of different sizes and performance abilities. Therefore, Boeing and Airbus compete for the same buyers who keep their purchases for many years. The fact that the technological strategies are so different runs against the predictions of New Institutional theory. It is possible that political factors distort the picture so that the market alone cannot be held responsible. Indeed, given the political importance of employment creation, export revenue, and national prestige of both Boeing and Airbus, it is not surprising that both companies have caused numerous headlines. At times the market in which the two giants function seems more politically than economically driven. Every few years, one or both companies are the focus of trade disputes from either side of the Atlantic [10–13]. A final agreement between the Americans and the Europeans might never be reached due to the highly complex politics involved. The political involvement in aircraft manufacturing is not surprising. Economically speaking, attempts by governments to smooth out the competitive disadvantages facing their home company is expected when trade is international [14]. In fact, by studying the few cases of successful aircraft manufacture, commentators have reached the conclusion that only companies that have enjoyed extensive state sponsorship and domestic help internationally have prevailed [15]. Political involvement has direct economic consequences to the producers. For example, price hikes that occurred as a result of political disputes have been documented in the air jet market [16]. However, there is nothing in the political environment that explains the adoption of opposite technological philosophies. Essentially, the question as to whether to keep the pilot in ultimate command or remand this privilege to the computer is

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

345

a question of strategy that does not rest on experience or on political pressure, since both Boeing and Airbus have access to the same technology. What is different is how the two companies have chosen it to function. The technological developments in the aircraft industry cannot, therefore, be accounted for by either economists’ models, sociological explanations, or simply by pointing to political rivalry. Surely all these factors matter for certain parts of the industry, but with regard to the question of the choice of man or machine, none offers a satisfactory answer. We shall have to look for it elsewhere. 4. The politics of airplane production The claim here is that the duopoly of aircraft production resembles the political struggle within a two-party political system. In a classic model from political science, Anthony Downs [17] reversed the conventional relationship between elections and party ideologies. Instead of assuming that parties win elections based on the appropriateness of their agenda, his theory gives elections preeminence, with the implication that all party action is aimed at maximizing votes [p. 35]. Political ideology serves two roles for democratic parties in this theory. First, it serves the heuristic purpose of making the party appear consistent across different political actions. In order to be identified as an ideology in the first place, the alleged purpose motivating party activity must be stable and internally consistent: ‘‘ideologies are never internally contradictory’’ [p. 113]. The other purpose is to eliminate uncertainty. By adhering to certain principles of action, voters have grounds to predict the likelihood of the outcome of their choice during elections. On the other hand, voters will be divided in the degree to which they are uncertain of the results from election. Those most certain will also constitute the most fervent political interest group. Parties, therefore, chose ideological affiliation in alignment with those groups that have the strongest belief in the outcome from successful elections. In effect, ‘‘uncertainty forces rational governments to regard some voters as more important than others’’ [p. 95]. There is an important parallel to duopolies in this model. According to Downs, both parties in two-party democracies try to appeal to the majority of votersdor the average voterdin order to capture the largest base of support [p. 68]. Both parties will therefore go to great lengths to try to resemble each other as much as possible in political appeal. To distinguish one party’s candidate from the other, the party will develop a systematic ideology that is only modestly different from its competitor’s. It follows from the majority principle in politics that both the government and the voters are interested in only ‘‘marginal alterations in the structure of government activity’’ [p. 53]. This distribution of the majority principle is, however, only fulfilled as long as there is a large measure of political consensus in the population of voters [p. 114]. The market for mid- and large-size airplanes meets these theoretical requirements well. There is great consensus among market actors as to what constitutes an attractive product, namely, a safe and economically efficient aircraft. The majority opinion is that safety must never be compromised; a dictum adhered to with equal strength at both Airbus and Boeing. Lastly, although there are only two suppliers, there are several interest groups that have a stake in the product design, just as there are many potential interest groups within a democratic system. Fig. 1 attempts to corroborate the claim that the basic two-party strategies of democracies applies to the duopoly of airplane manufacture. When considering seat capacity, the figure shows that neither company has tried to specialize in only one segment of the market. Both companies have, throughout their commercial duel, responded to every launch by the other by offering compatible, albeit slightly different alternatives. By using the jargon of political science, the attempt by both companies to offer aircraft for the entire seat capacity spectrum is equivalent to the fight for the ‘average voter’. 5. Technological frames In his exposition on sociotechnical change, Wiebe Bijker [18] offers as a demand for a theory of technological development that it situates the object of study within a social context of both users and producers [p. 47]. All existing technology is an accomplishment within a web of social actors who accept the claims to problem solving, goals, and assumptions of the invention. This accomplishment is what Bijker calls a ‘technological frame’ [pp. 124–25]. Neither one person nor even one side of the market alone is fully able to design this technological frame. Rather, the accomplishment of a viable technological frame is settled in relationship between several social groups, leaving no one fully in charge of the outcome. Relevant groups in modern airplane manufacturing are the two producers (Boeing and Airbus), airline carriers, engine manufacturers, national governments, pilots, and air transportation agencies. All of these contribute to any technological choices and to developments from both Boeing and Airbus. 5.1. Airbus: technological frame of dependence The adoption of digital fly-by-wire with hard envelope protection was certainly not a fad limited to one particular Airbus model, the A-320 family. In fact, all later Airbus models did, and will continue to use the same technological philosophy. Airbus affirms its system’s adherence to the strictest safety requirements; above all, however, the economic benefits emerging from the new plane’s better in-air performance, easier maintenance, lower weight, and better fuel consumption seem to have played an even greater role [19]. The economic benefits accrue to those who pay for manpower, maintenance, fuel, and repairdthe airlines. This section will show how Airbus’s flight control system is part of a technological frame that has been shaped to appeal to the airlines.

346

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

Seat Capacity
800

600

400

200

0 1970 1980 1990 Year of Introduction / Boeing / Airbus 2000 2010

Fig. 1. Seat Capacity as an illustration of duopoly strategy. Note: The Boeing families are, chronologically, B-737; B-747; B-757/B-767; B-777; B-787. The Airbus order is A-300/A-310; A-320; A-220/A-340; A-380; A-350. Sources: www.boeing.com/commerical; www.airbus.com/en/aircraftfamilies/productcompare/

When introduced to American carriers, the A-320 was harshly attacked by flight crews. Pilots’ unions on both sides of the Atlantic were opposed to Airbus’s system, mainly because it reduced the number of jobs, but also due to initial suspicions regarding safety with the hard envelope protection [12, p. 163]. In the words of a United pilot at the time: ‘‘This sort of sophistication [hard envelope protection] should be left to the single-seat fighter, where it appropriately belongs and where the pilot can bail out in a hurry when the inevitable occursda privilege neither I nor my passengers enjoy’’ [20]. Regardless which decade one looks at, pilots and their associations constitute a strong interest group in aviation. For instance, Robert Crandall, former CEO of American Airlines, told Jean Pierson, then managing director of Airbus, that because American Airlines’ pilots were opposed to the new cockpit, the company could not buy the A-320, despite American’s interest [21, p. 98]. A survey of 132 pilots of advanced aircraft found that pilots endorsed a human-centered philosophy and saw fully automated flight control systems as creating a heavier mental workload. In particular, pilots expressed a desire to maintain control of the most complex parts of their job, such as communicating, managing, and planning, especially in high workload situations [22]. The so-called ‘human-centered avionics’ approach to pilot-work has maintained that automation has the unintended tendency to tune out small errors and create opportunities for large ones; to diminish the workload in the relaxed parts of the flight and add to it in the busiest phases; and to amplify the importance of ongoing communication between crew members even as the immediate incentives to communicate are muted [4]. It was apparent that Airbus could not rely on support from the people who would actually operate the new technology in the air. However, after the Airline Deregulation Act was adopted in 1978, the aviation industry in the United States changed dramatically, and competition between carriers suddenly became an important factor. With a growing number of potential purchasers, rather than a few large ones, the desires and needs of individual companies had to be reckoned with in order to successfully introduce new aviation technology. Although Airbus is a European company, as Gerard Blanc (Airbus’s executive vice president for operations at the time of the A-320 launch) testified: ‘‘All of our big turning points were in the U.S. market’’ [21, p. 100]. And, those turning points were discovered with the airlines. It is due to the airlines’ desire for competition and lowered prices that they have pitched the two giants against one another [21, p. 224]. In the American market, much of the acceptance of the A-320 was not a matter of the plane’s technological achievements but due to Boeing’s reluctance to comply with American carriers’ desire for a modernized version of the B-737. According to Airbus spokesman Robert Alizart, ‘‘Boeing should have killed this upstart. If Boeing had produced a clean sheet of paper the A-320 would never have become Airbus’s bread and butter’’ [21, p. 100]. However, competition alone would probably have resulted in nothing more than disappointment for Airbus had the company not had something to offer besides low sales prices. The answer lies in the technological frame Airbus negotiated. First of all, the A-320’s fly-by-wire with hard envelope protection was completely new to the market. This feature itself appealed to airlines interested in the technological prestige of having the first mover advantage with a new machine. Most important, though, was the fact that Airbus introduced its new technology with the promise that future aircraft would utilize the same flight control system and cockpit configuration. This principle of commonality, well-known in the airplane industry, offers carriers the advantage of focusing on one type of technology throughout all their equipment. It gives airlines the advantage of buying multiple parts from the same supplier, the

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

347

opportunity to engage with only one form of pilot training, with all the benefits that come from specialized know-how. In the words of an industry commentator: ‘‘Cockpit commonality has been the cornerstone of cockpit design in Airbus aircraft’’ [23]. Airbus’s success at the American market was a direct result of their conscious strategy to establish fleet commonality. In interviews with Northwest Airlines representatives, John Newhouse [21] found that American carriers who purchased the A-320 did so because Airbus explicitly announced that future modelsdbeginning with the A-330 and A-340dwould be built with the same cockpit configuration. ‘‘This commonality in the Airbus family is known to have tilted a number of sales campaigns toward Airbus’’ [p. 102]. In fact, the A-380 Superjumbo, launched almost 20 years after the A-320, flies with almost the exact same technology as that of its predecessor. Not only does this signal the prescience of the A-320 technology, but also Airbus’s persistence in adhering to its philosophy of fleet commonality, and the wisdom of this early strategy. In real terms, the benefit to airlines of fleet commonality is clearly revealed in a February 15, 2008 press release, issued by EADS (Airbus’s parent company). It explains that the European Aviation Agency and the Federal Aviation Authority jointly approved reduced training requirements for flying the Superjumbo for pilots who were certified for all of Airbus’s digital fly-by-wire models (A-320 family, A-330, and A-340). In other words, within less than two weeks, pilots who used to fly planes carrying no more than 130 passengers could be certified to airplanes carrying as many as 850. EADS states that this gives airlines significant cost savings since training times can be halved compared to standard rating courses. How is this possible? EADS explains: ‘‘Reduced transition training is only possible due to the unique flight operational commonality between Airbus fly-by-wire aircraft. Commonality is a fundamental design criterion for Airbus as demonstrated by the fact that [reduced transition training] is available now for all combinations of Airbus fly-by-wire aircraft in service’’ [24]. Airbus’s fleet commonality design is a technological frame that has fostered loyalty from many airlines. It is a technological frame because the technical details of the flight control system do more than simply facilitate the introduction of new models to the market. Instead, the technological frame actively shapes future company strategy and success. Even the choice of developing new models has been partially made with a view to preserving the technological frame. For instance, Hanko von Lachner, Airbus’s general secretary, said: ‘‘An argument in favor of the A-380 was Boeing’s 777. It was a better airplane than our A-340. But we calculated that the A-380 would help to protect the A-340. An airline that bought the A-380 would be unlikely to buy the Boeing airplane’’ [21, p. 155]. Airbus’s technological frame is therefore one of dependence. The company anticipated possible future technological trends, developed a flight control system it would stick to without yielding, and thereby made airlines reliant on their products if they wanted to make money on the equipment already in their fleet. True, the hostile pilots had to be converted, but once several large airlines were swayed, their flight crew became specialized in the same technological frame of dependence as their employer. Today, half of the pilots in several countries are only able to operate an Airbus plane. And as a result, it is harder to find incendiary or negative remarks about digital fly-by-wire with hard envelope protection. As expected from candidates in a two-party democracy, Airbus chose to affiliate with the one group in the duopoly that was likely to appreciate the company’s technological ideology, namely, the airlines. 5.2. Boeing: technological frame of accommodation The reason Boeing chose a different technological philosophy lies in the company’s decision to affiliate its technological frame with pilots. When digital fly-by-wire technology seemed inevitable, Boeing chose to implement its own version, even though it needed a technological frame that was different from the one Airbus had introduced. It took Boeing eight years to produce its response: the B-777. Karl Sabbagh [25] followed the B-777 project from within the company, and he related, ‘‘John Cashman, Boeing’s chief test pilot on the 77, was instrumental right from the beginning in making sure that the fly-by-wire system reflected the views of pilots rather than relying on engineers’’. This led to the decision to ‘‘let the pilot make the final decision rather than the fly-by-wire computers’’ [p. 149]. ‘Softer’ envelope protection, which can be overridden, reflects pilots’ general attitude to the safety of digital fly-by-wire. For instance, the Journal of the Airline Pilots Association published a piece written by a member pilot in 2000 which instructed plane crash investigators to ‘‘first assume that the FCS [flight control system]dand not the pilotdinduced it [the accident] until proven otherwise,’’ when investigating accidents involving aircraft fitted with fly-by-wire [26]. Regarding the implementation of ‘feel’ feedback in instruments and retaining the traditional yoke, Sabbagh [25] reports that ‘‘Boeing pilots who worked on the system tried very had not to turn the 777 into a plane that is flown by reading things rather than feeling them’’ [p. 153]. Pilots’ desires were also taken into account by Boeing in maintaining the traditional yoke. Sabbagh explains that the designers of the 777 fly-by-wire system wanted pilots to feel at home, and they built artificial feedback into the system that gave useful information to the pilot about how far s/he had pulled the lever. Apparently, there ‘‘was an almost religious fervor in the way the system’s designers left a degree of freedom for the pilot in the system’’ [p. 153]. It is interesting to note that whereas today’s pilots might favor a cockpit that resembles traditional aircraft, laboratory experiments suggest that individuals without any flight experience are more reliable when navigating with an untraditional sidestick [27]. Only two new models have been developed by Boeing that implement digital fly-by-wire flight controldthe B-777 and the new B-787 Dreamliner (scheduled for operation sometime in 2010). The latter will contain virtually the same digital fly-bywire as its predecessor, and the cockpit configurations will be almost identical. This includes ‘feel’ feedback in some instruments, as well as a yoke instead of a sidestick [28,29]. Neil Adams, manager of business development at Rockwell Collins Systems Division (the company that furnishes Boeing’s cockpits on the 777 and 787), explained: ‘‘Boeing has tried to make the

348

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

cockpit of the B-787 identical to that of the B-777, so a pilot can go in [to the cockpit] blindfolded, touch it, and it feels exactly the same’’ [29, p. 51]. As a result, B-777 pilots can obtain certification for the B-787 in five days or less. At first glance, it looks as if Boeing’s technological frame is identical to that of Airbus in that Boeing, too, aims for airlines’ loyalty through fleet commonality. Closer inspection reveals this is not the case. Of course, the American company has to appeal to airlines just as its European competitor. The important thing to bear in mind, though, is that Boeing’s technological frame was developed before the final product was introduced to the market. By allowing pilots into the development of their system, airlines were later presented with a technological solution that was furnished in accordance with the opinions of the pilots. This process resulted in a technological frame that did not attempt to introduce something radically new, but instead something that was compatible with what already existed. This is noteworthy because Boeing explicitly tried to set limitations on the technology they mastered. Here too, the testimony of the test pilot in the 777 project is illuminating. John Cashman explained: ‘‘We really want to keep the airplane flying just like he’s [the pilot] always flown it. We don’t want him to have to learn new techniques. We’re trying to keep the airplane so the pilot, when he climbs into it, doesn’t have to learn anything new’’ [25, p. 159]. Boeing’s technological frame therefore means establishing commonality between its new models and older onesdand even with smaller aircraft made by other non-Airbus companies. Outside of the military, digital fly-by-wire does not exist in any of the smaller regional jets (as of 2009), or in any of the older larger models. In developing Boeing’s fly-by-wire system and its cockpit design, Adams from Rockwell Collins stated: ‘‘What we are trying to do is create the ‘feel’ of a mechanical cable system. It allows the pilots to feel like they are flying a 1950 aircraft in 2007’’ [29, p. 50]. The direct benefits of this technological philosophy do not really get to the airlines. Airline management neither enjoy the right to override flight control systems nor are they the ones to feel the feedback in cockpit instruments that resemble older mechanical aircraft. Boeing’s technological frame is therefore one of ‘accommodation.’ Pilots’ needs and desires are diligently implemented in the flight deck by the company, and pilots are allowed to command the plane fully as they see fit. Airlines are not forced into technological dependence, as they are by Airbus, but receive new aircraft that behave as close as possible to smaller regional jets and older large aircraft. Leaving aside all marketing strategies and tactics adopted by the company, we notice that Boeing’s technological frame has been adjusted to meet the desires of the flight crew. The ideological alliance with this latter group distinguishes Boeing’s policy from that of Airbus, and explains the choice to abandon strong envelope protection and the retention of more traditional flight deck instruments with tactile feedback built in. When the systems were launched, Airbus was looking ahead at what would be the future technological norm, whereas Boeing looked back at what already existed. 6. Conclusion This paper has argued that marketing strategies and technical development alone cannot account for the emergence of different technological adoptions for commercial airplanes. Instead, attention has to be given to the ideological alignments of those who manufacture the aircraft. Politics in two-party democracies (insofar as voters share a general agreement of desirable outcomes from elections) often involve efforts to become as similar to the opponent as possible, while still striving to maintain a unique identity. This identity is secured by ideological commitments with certain groups of voters. Based on testimonies by some authors who have been privy to the technical planning processes of both Boeing and Airbus, this paper has shown that in today’s duopoly of airplane manufacture, both companies forged ideological alliances, but each with a different social group. Boeing diligently maintained its commitment to the desires of pilots, whereas Airbus engaged in new alliances with the airlines. The idea of ‘technological frames’ was used to address the social context in which any technological application gets credibility and receives its final form. Boeing’s frame was described as one of ‘accommodation,’ and Airbus’s was characterized as a frame of ‘dependence.’ After more than a decade of coexistence between these two philosophies of flight control, the initial debate as to the best way has calmed, which must surely be attributed to the fact that both systems have been proved to work safely. Also, the situation has yet to occur where it can undeniably be determined that either company’s flight control system would have prevented an accident befalling a plane from the competitor. The technological frames have both managed to settle and exist side by side. The question now is: is it possible that the debate will reappear? It seems likely that the next great restructuring of the airline business will again raise the issue of automation. Within the next decade, or so, air travel in the United States might be permitted to proceed at each pilot’s discretion, which is called ‘free flight’ air traffic control. Today planes must follow certain fixed routes, more or less like cars on interstate highways. However, recent advances in global positioning satellites, new ground/air communication links, enhanced collision-avoidance systems onboard planes, and powerful automation could make it possible to let pilots fly the shortest route between two points. This, of course, has the potential of reducing fuel consumption and is therefore highly attractive to airline carriers. As discussed, both of the manufacturing giants will have to participate in new politics if ‘free flight’ control is introduced. Air traffic controllers are likely to resist that idea, given the likelihood of some job losses. Pilots and airlines, on the other hand, might favor the arrangement, although probably for different reasons: pilots might tend to prefer arrangements that increase their decision power in the air, whereas airlines are most interested in ways to minimize expenses. The decision for Boeing and Airbus is with whom they will collaborate for the next technological frame. Whatever the outcome of future decisions about ‘free flight’ and an onboard flight control system, the technological philosophy will undoubtedly again occupy center stage. In the words of one commentator: ‘‘Understanding what human/technology capabilities would exist in a free flight world is essential to evaluating its safety’’ [30, p. 845].

A.Z. Ibsen / Technology in Society 31 (2009) 342–349

349

Acknowledgements The author is grateful to Doctor Theodore Beneigh at Embry-Riddle Aeronautical University for help to understand the technical aspects of modern flight control systems. References
[1] Tomayk JE. Computers take flight: a history of NASA’s pioneering digital fly-by-wire project. Washington, DC: National Aeronautics and Space Administration; 2000. [2] Lawrence PK, Thornton DW. Deep stall: the turbulent story of Boeing commercial airplanes. Aldershot, UK/Burlington, VT: Ashgate; 2005. [3] Mitchell W. Flying the electric skies. Science 1989;244:1532–4. [4] Sweet W. The glass cockpit: pilots who work in glass cockpits throw no stones, despite some deep ambivalences. Institute of Electrical and Electronics Engineers (IEEE) Spectrum 1995;32:30–8. [5] Norman DA. The ‘problem’ with automation: inappropriate feedback and interaction, not ‘over automation’. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 1990;327:585–93. [6] Beckman SR. Cournot and Bertrand games. Journal of Economic Education 2003;34:27–35. [7] Giovannetti E. Perpetual leapfrogging in Bertrand duopoly. International Economic Review 2001;42:671–96. [8] Tombak MM. Strategic asymmetry. Journal of Economic Behavior and Organization 2006;61:339–50. [9] DiMaggio PJ, Powell W. The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. American Sociological Review 1983;48:147–60. [10] Carbaugh RJ, Olienyk J. Boeing–Airbus subsidy dispute: a sequel. Global Economy Journal 2004;4:1–9. [11] Carbaugh RJ, Olienyk J. Boeing–Airbus subsidy dispute: an economic and trade perspective. Global Economy Quarterly 2001;2:261–82. [12] Lynn M. Birds of prey: Boeing vs. Airbus, a battle for the skies. New York: Four Walls Windows; 1998. [13] Love W, Sandholtz W. David and Goliath: Airbus vs. Boeing in Asia. In: Aggarwal VK, editor. Winning in Asia, European style: market and nonmarket strategies for success. New York: Palgrave; 2001. p. 187–224. [14] Garcia Pires AJ. Losers, winners and prisoner’s dilemma in international subsidy wars. CEPR Discussion Papers 2006;5979. [15] Hira A, de Oliveira LG. Take off and crash: lessons from the diverging fates of the Brazilian and Argentine aircraft industries. Competition and Change 2007;11:329–47. [16] Irwin DA, Pavcnik N. Airbus versus Boeing revisited: international competition in the aircraft market. Journal of International Economics 2004;64: 223–45. [17] Downs A. An economic theory of democracy. New York: Harper & Row; 1957. [18] Bijker WE. Of bicycles, bakelites, and bulbs: towards a theory of sociotechnical change. Cambridge, MA: MIT Press; 1997. [19] Briere D, Traverse P. Airbus A320/A330/A340 electrical flight controls: a family of fault-tolerant systems. Twenty-Third Annual International Symposium on Fault-Tolerant Computing. 1993: 616–623. [20] Fulford GA. Letter. Science 1989;245:582–3. [21] Newhouse J. Boeing versus Airbus: the inside story of the greatest international competition in business. New York: Vintage Books; 2008. [22] Tenney YJ, Rogers WH, Pew RW. Pilot opinions on cockpit automation issues. International Journal of Aviation Psychology 1998;8:103–20. [23] Ian P. Avionics for a colossus. Avionics Magazine 2000;24:20–2. [24] European Aeronautic Defense and Space Co. (EADS). Reduced pilot training now approved for the A-380. Press Release, 2008. [25] Sabbagh K. 21st century jet: the making and marketing of the Boeing 777. New York: Scribner; 1996. [26] Stowe S. Fly-by-wire: a primer for aviation accident investigators. Air Line Pilot 2000;69:18–21. [27] Beringer DB. Applying performance-controlled systems, fuzzy logic, and fly-by-wire controls to general aviation. DOT/FAA/AM-02/7, 2002: 1–8. [28] Ramsey JW. Boeing 787: integration’s next step. Avionics Magazine 2005;29:20–9. [29] Ramsey JW. The Dreamliner, in control. Avionics Magazine 2007;31:46–55. [30] Barnett A. Free-flight and en route air safety: a first-order analysis. Operations Research 2000;48:833–45. Alexander Z. Ibsen is currently a Ph.D. candidate in Sociology at the University of Arizona.