Flyvbjerg - rationality

Flyvbjerg - rationality - final eve/Wharf, ' jg annual?...

Info iconThis preview shows pages 1–16. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 16
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: final eve/Wharf, ' jg annual? “flan; 9190/ 2 Rationality, body,‘ and intuition in human learning Our task is to broaden our reasoning to make it capable of grasping what, in ourselves and others, precedes and exceeds reason. [Maurice Marlena-Pam)! "J Context is central to understanding what social science is and can be. This chapter asks, “What role does context play in human knowledge and skills?” Philosophy of science and epistemology typically pose queStions such as: “What is knowledge?”; “What can we know?”; “Under what conditions can we know that We know?” Here we will approach the question of knowledge by askin the more dynamic question: “How do people acquire knowledge and s fills?” It is by addressing this question that We begin to understand the problem of context. The intention here is not to outline and analyze all possible ways in which people acquire knowledge and skills, nor shall we review the many schools and theories that exist in this area. Rather we will deal with a single phenomenology of human learning as formulated by Hubert and Stuart Dreyfus.1 This particular phenomenology has been chosen be— cause it is especially useful for understanding the linkage between knowl- edge and context, and because it directly addresses the question of whether knowledge about human activity can be context-independent. The answer to this latter question is decisive for an understanding and response to two fundamental epistemological questions in the study of human activity: “Are theory and epistemology possible in social science?” “Can social and political science be scientific in the same sense as is natural science?” The first part of the chapter reviews the phenomenology of hlurnan learning, the so-called Dreyfus model. We will then discuss the model’s implications for social science. lll Why social science has failed as science Competence and virtuosity in human learning Some years ago in the United States, group of paramedics. V cardiopulmonary resus Five of the six were in sixth was a paramedic e an experiment was conducted on a ideo films were made of six persons administering ¢itation (CPR) to victims of acute heart failure. Xpen'enced trainees just learning CPR, while the with long experience in emergency life—saving techniques. The films W with practical experience 9 tors in life-saving technigues. Each sub ersons shown in the films would you choose to resuscitate you if you were the victim of such an accident?” group of experienced paramedics, paramedic from the film percent of the cases. Finally, tion: “Who of the six p resuscitation had poorer or the students, choosing the cases.2 What form of rationality led the instruct performance? And what amedics’ well—developed be dealt with in the follo (I) Novice r (2) Advanced beginner (3) Competent perfonner (4) Proficient performer (5) Expert They are levels, say Dreyfus and Dreyfus, , terms they consist of reco and performing in the pr given level do better than achieve the highest level in re shown to three groups of subjects: paramedics , students being trained in this field, and-instruc- Among the 90 percent chose the one experienced s. The students chose “correctly” in Only 50 and perhaps surprisingly, the instructors in results than either the experienced paramedics the experienced paramedic in only 30 percent of mechanisms lay behind the experienced par- ability to choose correctly? These questions will Wing discussion. Detailed phenomenold people pass through sevena “skills” are understood II e.g., building a house, heing socially adept, studies, all after the degrie into a varying number of five levels in the human-1e gical studies of human learning indicate that lphases or levels in the learning of skills, where 0 range from the technical to the intellectual; analyzing a text. Various e of detail, have divided the learning process such levels. The Dreyfus model operates With arming process: because in phenomenological izable, qualitatively different ways of acting cess of learning a given skill. Individuals at a individuals at the previous level. Not all people a given field. Some fields, such as ehess, guitar playing, or surgery, are characterized by only a small fraction of n0vices hecmning experts. In oth 61' areas, such as bicycling and driving, a large Rationality, body, and intuition in human learning 1 1 number of novices reach the expert level. Let us examine the levels one at a time. I . N om'ce As a novice, the individual experiences a given problem and a given situation in a given task area for the first time. During instruction the novice learns what various objective facts and characteristics of the: situ- ation are relevant for the perforthance of the skill. The novice learns to recognize these facts and characteristics when they appear. On this basis, the novice also learns rules for action. Facts, characteristics, and rules are defined so clearly and objectively for the novice that they can be recog— nized without reference to the concrete situation in which they occur. On the contrary, the rules can be ge eralized to all similar situations, which the novice might conceivably corilfront. At the novice level, facts, charac— teristics, and rules are not dependent on context: they are context- independent. Let us take, for example, someone learning to drive a car. The student driver learns facts about speed and shifting gears, and he or she learns the rule that when the speed exceedh a certain level, you must shift gears. Both the fact (speed) and the lrule (gear shift at certain speed) are independent of the concrete sitluation. In principle, shifting gears is executed as the same type of logical information process as within a digital computer. Later on, when the n vice has shifted gears many times and has achieved a higher level in the cliearning process, the gear shift situation is recalled as analogous to prior situations. Shifting gears now occm‘s by reflex, without direct use of contht-independent facts and rules. Novices judge their skills by evaluating how well they follow the rules they have learned. When novices have learned a handful of rules for a given skill, however, performing the skill becomes so complex and de- mands so much concentration that it impedes continued improvement of performance. For example, the ability to speak and to listen to advice declines in relation to the number of rules which the novice learns and must remember to use. The firs rules are necessary for gaining initial experiences, but the rules quickly ecome a barrier to the learning process and must be put aside in order for the novice to advance. 2. Advanced beginner The beginner advances from the first level in the learning proceSs by achieving real—life experience, in contrast to the often deliberative and protected learning situations of the first level. Via these further experien— i 2! Why social science has failed as science ees, the advanced beginner learns to recognize relevant elements in relevant situations. ReCCEgnition occurs because the advanced beginner sees similarities in relat on to prior examples of the same situation. Gaining experience cons Recognition is concrete context which plays the lists in a cumulative recognition of similarities. and dependent on context, and it is precisely decisive role, for it is context which becomes increasingly more important as one proceeds up the levels of the learning process. For the advanced beginner, the basis for action may contain elements which are both situation advanced~beginner level concrete situation of the of speed, and will in fact situation. In this sense, bend or ignore the rules. or especially weak positio independent rules. And . rules to identify the smell text-independent, explicit 3. Competent peifo With more experience, th individual sees in a con individual lacks a feeling the individual is unable to l 0 a1 and context-independent. A driver at the can thus shift gears on the basis of bOth the motor sounds and the context-independent rule use both indicators according to‘ the specific tuational behavior involves knowing when to good chess player recognizes eSpeciailyistrong sin a concrete situation without use of context- ere is no one who needs to combine facts and of freshly brewed coffee, Dreyfus and Dreyfus say. Personal experience via trial-and-error is more important than con— , verbally formulated facts and rules. rmer e number of recognizable elements, which an irete situation, becomes overwhelming. The what elements are important. In other words, prioritize.°At this stage, individuals learn from themselves and from others to apply a hierarchical, prioritizing procedure I for decision-making. By first choosing a goal and a plan with which to ‘ organize the information about the concrete situation, and then process- t in g only those factors relevant to achieving the goal and plan, lthe individ- ‘ ual can simplify his or her task and obtain improved results. A professor of nursing explains the problems her interns had with making the transition from the initial, rule—based levels in the learning process to the kind of prioritizing behavior and overview which charac— terizes competence: I give instructions to the new graduate, very detailed and explicitiinstructions: when you come in and first se the baby, you take the baby’s vital signs and make the physical examination, an you check the LV. sites, and the ventilator and make sure that it works, and on check the monitors and alarms. When I would any this to them, they Would d exactly what I told them to do, no matter what else was going on . . . They coul ’t choose one to leave out. They couldn’t choose which one was the most impo ant . . . They couldn’t do for one baby the things Rationality, body, and intuition in human learning that were most important, and leave the things that weren’t as important] unlil later on . . . If I said, you have to do these eight things . . . they did those things. and they didn’t care if their other kirj was screaming its head ofi". When they did realize, they would be like a mule be ween two piles of hay.3 Via goals, plans, and the setting of priorities, the student nurses learn to deal with a smaller set of key factbrs instead of the total knowledge abOut the actual situation. The competent nurse, in centrast to the beginner, does not go automatically from atient to patient in a preset sequence, but continually evaluates the paiients’ need for attention and care and arranges his or her routine according to these evaluations. The per— former’s behavior “flows” and becomes better adapted to the concrete situation. Selecting a plan is not simple, and not without problems for competent performers. It takes time and re uires deliberation. There are no objec- tive procedures for choosing a p an similar to the novice’s context-linde- pendent choice of facts and appllication of rules. Besides, the choice of plan has wide-ranging consequences for actions and results in- a way which the choice of other elements seldom has. The lack of ten-afirma for the choice of plan, combined vdith the competent performer’s need to have a plan, produces a new, important relationship between performer and surroundings: a relationshi of involvement. The novice and the. advanced beginner have only lit-tilted responsibility for the consequences- of their actions, these actions bding predetermined by learned elements and fixed learned rules. Exoludi g a gross error, a bad result will therefore appear as having been caused y inadequately specified elements and rules. Actions and results will thus stand in an external relation to the beginner: they can be justified and given a rational explanation in relation to objective facts and abstract rules. Competent performers, on the other hand, are personally involved in their actions. The competent perfOrmer, after having struggled with the problem of selecting a plan, feels respon— sible for the consequences of the choice precisely because selecting a plan cannot be done objectively, but must nevertheless be carried out in order to be able to act competently. Hence, the actions of the competent performer comprise an elemen ' retatio ' t. ‘As we shall see, the ability to make ese judgments becomes crucial at the upper levels of the learning process. It is this ability, according to Dreyfus and Dreyfus, which constitutes the core of true human expertise. Cognitivists and others who conceive of thinking as logical informa— tion—processing and analytical] problem-solving concern themselves mainly with the kind of thinkin petent performer” stage. Herb view. In his attempt to underst processes which take place at the “com.— rt Simon is a leading exponent of this and how people select plans, goals, and l 4 Why social scie ce has failed as science strategies, Simon and his colleagues have convincinglylillustrated how people confronted wi unknown tasks in unfamiliar situatiorts act as analytical problem-sol ers. The cognitivists, however, tend to generalize these results as beingv lid for all intelligent behavior. People are generally seen as problem-solvi g beings who follow a sequential model of reason- ing consisting of “e1 ments—rules—goals—plans—decisions.”“ It is this model which the co itivists have attempted to simulate in computers and in various probl m-solving models, in “expert systems” and in artificial intelligence. eir extrapolation yields good results when the models are applied to ell-defined tasks with well-defined solutions. The cognitivists have had uch less success, however, when the tasks and solutions are less well defined. According to Dreyfus and Dreyfus, the poor results reflect the lack of evidence for the cognitivisth’ assertion that humans can act intelli ently only by acting as analytical problem-solvers. There are other kinds f intelligent behavior, assert Dreyfus and Dreyfus, which appear especial y among those individuals who lare either very proficient or experts in their fields.5 In many of our dail activities, we can see phenomehologijcally that humans do not exclusi ely act as conscious problem-solvers, i.e., choos- ing goals, plans, and c mbining elements according to miles for reaching goals. When we ride bicycle, recognize faces on the street, or talk to our neighbors, we do ot appear to be solving problemsi Of ecurse, we may be operating unc nsciously as logical information‘ rocetssors and problem—solvers, but a we will see, this does not have tote the case and there is no evidence to support this claim. The fundamental error of the cognitivists is that they exclude any other possibility. In contrast to the co petent performer, genuine human experts exhibit thinking and behavior at is rapid, intuitive, holistic, interpretive, and visual and which has 0 immediate similarity to the slow, analytical reasoning which Chara erizes rational problem-solving arid the first three levels of the learning recess. 0n the contrary, it seems that there is a fundamental and qual tative jump from analytical prohlem—szolving to genuine, human expert so. This jump must be made in order for:someone to be really adept at p rforming a given skill. Stuart Dreyfus, the main architect behind the fiv ~leve1model for human leaminga is a competent chess player. However he has remained at the “competent” level and cannot improve becau he finds himself unable to make} the qualitative jump to the next level 0 “proficiency,” and he says he will never become an “expert.” Dreyfus e aborates on the possible causes: I was always good at ma matics and took up chess as an outlet for that analytic talent. At college, where I captained the chess team, my players were mostly mathematicians and mos , like me, at the competent level. At this point, a few of l Rationality, body, and imiuition in human learning = my teammates who were not math maticians began to play fast chess at the to | c 1:1 five or ten minutes a game, and a1 0 eagerly to play over the great games of 1hr- grandmasters. I resisted. Fast chest was no fun for me, because it didn’t give me time to figure out what to do. I fodnd grandmaster games inscrutable, and since the record of the game seldom if ever gave rules and principles explaining themselves, I felt there was nothing I could learn from the games. Some of my teammates who through fast chess and game studying acquired a great deal of concrete experience have gone on do become masters. As I look around at my mathematical academic colleagues, most of whom play chess and none of whom have gotten beyond my own competent level, I see how our view of chess as a strictly analytic game has cut us off from absorbing concrete chess experience. While students of mathematics and related topics predominate in the population of young peopleienthusiastic about chess, you are as likely to find a truck driver as a mathematician among the world’s best players. You are more likely to find an amateur psychologist or a journalist. In a way I am glad that imy analytic approach to chess stymied my progress, because this helped me to see that there is more to skill than reas ning.6 l When I asked Stuart Dreyfus i an interview where in the body a chess player feels that a move is rightie told me, “in the whole body. In the pit of the stomach?” It is similar, says Dreyfus, to asking where do you feel you are hungry when you are hungry. “You can’t say that your brain thinks it is hungry,” continues Dreyfus, “you experience your whole body as craving and the chess player hhs the same type of experience.”3 Dreyfus explains in the interview that when chess players play one-second-a-move chess, they describe a strange sehsation that their hand is playing and they are not. “Their hand is just moving pieces as fast as it can and they almost feel as if their detached brain looks down at their hand playing chess,” says Dreyfus, “so the whole body is even in that picture.”9 , l Stuart Dreyfus touches here tin two important general points.§irst,?n exclusive use of analytical ratiohality tends to impede further i ‘ e- ment in human performance belcause of analytical rationality’s 810W rea- soning and its emphasis on rdles, principles, and universal solutions. ficzofi‘cl,_bodily involvement, spiced, and an intimate knowledge of cori— cféte' cases in the form of good examples is a prerequisite for true exper— tise. We will return to these factbrs repeatedly in the following chapters. Doctors and nurses say that eitperiences from working in an emergency ward are important for developing skills in clinical practice. The emer- gency room’s patients are often acute cases with a broad range of different problems. Often there is no time to retrieve all the information one might want about the patient, and doctors on night duty will often not be able to obtain immediate aid from their more experienced colleagues. Doctors and nurses in an emergency room are therefore forced to think on their feet, i.e., to act quickly and to:- utilize spontaneously their experiences l 10 Why social science has failed as science from similar, prior siluationsfil‘hinking on their feet contributes to the development of intuition and judgment; prerequisites for becoming a good clinician. The emergency room situation contrasts with a ward for internal medicine, fdr example, where the doctor has more time, the patients have been thiere longer, the case histories are more detailed, the illnesses less acute and the outcomes more predictable. 4. Proficient performer: beyond analytical rationality In the first three levels; the performer of a given skill has made a conscious choice of both goals and decisions after having reflectedthoroughly over various alternatives, if the individual has not simply followed rules. Dreyfus and Dreyfus dall this procedure the “Hamlet model” of decision- making.10 In contrast ltO this model, decision—making for the proficient performer is more continuous and is not sequential in the same way. Proficient performers tjend to be deeply involved in their actions and have evolved their perspective on the basis of prior actions and experiences. This perspective enalbles certain key features of a situation to stand out, while others recede into the background. New actions and experiences change the predomina ‘ t features, plans, and expectations, and with it the actions. No objective cihoice or conscious evaluation of appropriateness takes place, which is the case in selecting elements, rules, and plans. The choice is simply made,l that much is clear phenomenologically speaking. And this seems to happen because the proficient performer has experi* enced similar situations earlier. Via spontaneous interpretation and intu- itive judgment the memory of these situations generates plans corre- sponding to plans which have worked before. Similarly, of earlier situations releases expectations about actions, which correspond to those actions carried out in sitnilar situations earlier. c The proficient performer understands and organizes her or his tasks intuitively, but intermittently continues to reflect analytically over what will happen. Elements alrid plans from the performer’s experiences, which appear as intuitively important, are evaluated and combined analytically with the help of rules for reaching decisions about the most appropriate actions. Deep intuitive linvolvement in performance thus interacts with analytical decision-making. To use one of Dreyfus and Dreyfus’s examples, a proficient marketing manager keeps herself oiiented about the market situation for her produCt by reading and listening to everything in her area, from formal reports to gossip in the field. One day the manager can intuitively decide that a problem or sales possibility exists, and that a new sales strategy should be considered. The manager then initiates a study of the situation hnd may 5 a B 0 Q Rationality, body, and intuition in human learning I l even take great pride in carrying out a sophisticated scientific Illill'lit‘ll analysis, while overlooking the Equally important ability to be able ml identify the existence of the problem or possibility intuitively; this despilcl the fact that it was the manageri’s intuition which led to the marketingl study being initiated in the first place. i 5. Expert“ The proficient performer gradually achieves intimate experience from different situations, all of which touch upon the same goal and the same perSpective, but which demandldifferent tactical decisions. The profi- cient performer then perhaps achieves a level in which it is not OHIYl situations, which are recognized intuitively, but also -— synchronicallly and holistically — the relevant decisions, strategies, and actions. According to Dreyfus and Dreyfus, this is the level of genuine, human expertise and is characterized by efiortless performance. It is the level of virtuosity. l Expert soecer players assess the moment for dribbling or the possibility to score a goal by the entire visual situation in front of them, together with the sensations in their bodies releasing memories of earlier situations, where dribbling or attempts at sdoring have succeeded. There is nothingl which indicates that soccer players utilize general rules to combine vari- ous facts about their own and iheir opponent’s positions, movement, speed, etc., and then select a course of action on this basis. Intuitive, holistic, and synchronous action Es now at the center. I y In normal, familiar situations,lreal' experts do not solve problems and do not make decisions. They just do what “works.” This does not mean that experts never think Consciq‘usly, nor that they always do the right thing. When there is time, and When much is at stake, experts will also deliberate before they act. Their deliberation, however, is not based on calculated problem solving but {in critical reflection over the intuition, which the expert applies. Evenlafter this reflection, there will remain situations where the expert’s decisions do not work. Unforeseen events may occur. And when one expert confronts another in competition, as in a championship chess or tennis match, only one of them can win. f Compared to rational decision-making, intuitive decision-making has been neglected as an object for scientific study, perhaps because science tends to emphasize analytical rationality as its own too]. Ultimately it is a‘ question of what constitutes scidnce, and whether it is possible to study phenomena such as intuition and synchronicity scientifically. Yet, we are‘ familiar with most of these phelnomena in their nonscientific form —f seeing What needs to be done in an instant — when we perform in a craft, a sport, or making music. 1 l H Why social sdjence has failed as science ‘3 Where science does not reach, art, literature, us comprehend the reality in which we live.'2 Freud, Who in was a pioneer for research into human learning, “valuable allies . . . flwho in] their knowledge of th advance of us everyd 1y peOple, not yet opened up fo science. for they draw upon sources which ’“3 The late Danish novelist Han Nielsen described viiituoso expertise in soccer and used a Iabe virtuoso soccer play r, which says it all: “soccer angel” (fa Here are some “angefis” in action: l We get a free kick, just “jithin the other team’s penalty zone, just to th goal, and I take it, self-assuredly waving the others off, with the sed ii I and no one else kn opponents stand up in a; wall in hour of me goal, as I perhaps also had first thought, them, far to the right, like an extension of th approach, while everything is focused on as if to kick directly. whim I get to it, defensive wall, and the hall would have the same moment, Franke has made his way around and has rushed place where it would haste landed. It never does, he catches it in the right leg, half-gliding iti o the goal with his left. No one else is able to has happened before hell‘itfts his arms. Soccer players tend to ave this kind of thing with them from home, it over and over again d g training. Franke and I have perhaps done similar before, but never practiced it as something specific in this wa exchange a word before I‘ take the free kick happens during the run- stands, I just play him lilie he has to be played when he has position where he has suddenly p anything we are so awar ‘ shared knowledge, from i become reality, and it is p an ego . . . It lies priorito, she, it, in themselves . . . Standing there, first genuinely surprised, then intensely happ , kno then and there that this very moment in the grey, luminous teammates just away fro me in a bunch around Franks . . moment I will always re ember . . . The fantastic thing is that the successful in precisely this nonchalant, effusive way. This is wh and Franke, too; in any case he opens the bunch around him ever so slow toward me, while we eye ehch other, and gives me a little slap on the a receipt, '4 ‘ ‘ . ucti but suddenly, Fratike stand Y: , not even a telling glance, of that it can become clear for us in adva e perspective of the bodies and the eyes ’or to our being able to speak about it as a la or outside, sentences which must contain 1 an ay eve Experts Operate from a mature, holistic well—tried unders intuitively and without ponscious deliberation. comes primarily from experiences on one’s own body and is in th and narrative often help many ways thus saw writers as e mind . , . are far in we have s-Jorgen I for the dboldengel) . e right of the VC move- ows what needs to be done. The in order to block a shot aimed at the S next to eir wall. This has happened during my me, and I keep running toward the ball I instead kick it in a very flat arc, over the taken the turf a few meters behind it. In toward the air with his grasp what working on i something We don’t pp, completely natural, he just stands there where he ed himself ositioned himself, the thought doesn’t even become nee. It is a ready to nguage and I, you, he, lying right ning, the . it is precisely this goal was at surprises me, rm, like a tanding, Intuitive understanding is way at p t l Rationality, body, and intuition in human learning I "i Eone with the performer. Experts do not see problems as one thing :nmli solutions as something else, they do not get anxious about the futurci while they act, they do not make plans. Their skills have become so muchj a part of themselves that they are no more aware of them than they are oil eir own bodies. i 3 It is important to emphasize tihat when Dreyfus and Dreyfus me the; word “intuition” they do not mean some kind of guesswork, irrationality; or supernatural inspiration, as the cognitivists often describe it, usually as' a preface to a critique. For Dreyfus and Dreyfus intuition is a property: which each individual uses in eyieryday life. The cognitivists, limited as; they are to explanations in terms f rule-based processes, can explain only. competent human performance. ey are so far unable to integrate hu—: mans’ expert performances into their models. Take something as mun-! dane as riding a bicycle. Someone able to ride a bike has not formulated a set of rules, which, if followed, can teach somebody else to ride a bicyele.‘ How could we, for example, “teach” the difference between nearly falling and the need to lean over in order to turn a corner? How do we explain the best response to being off bala ce? Bicyclists can bicycle because they: have the necessary know—how, allhieved via practical experiences, invari-i ably accompanied by a few childhood scrapes and bruises. Experience: cannot necessarily be verbaiized, intellectualized, and made into rulesi Therefore, the cognitivists have a difficult time understanding it. : Sitting at a computer, a virtuoso secretary “is one” with the machine! and does not think over what finger does what on the keyboard. A virtuoso car driver is one with th car. If an American attempts to drive in; a left—hand—drive country such ajEngland, however, the experience is one of stepping backwards in the learning process: formerly effortless, unrefd lected driving becomes stiff and dependent on the conscious deliber- ations and decisions of the begihner. It becomes a problem to make a rightturn or to drive through a: rotary. The same happens to hospital doctors who transfer to a new unfamiliar ward. Hence, there exist rota— tion arrangements as part of training young doctors. Studies of pilots’f learning processes indicate thatfi'lljovice pilots “fly their planes,” while as experienced pilots “they fly.” e separation between person and ma-i chine, subject and object disappears. One scientific domain, which puts virtuoso expertise into sharp and instructive contrast with rational competence, is the area of “artificial intelligence” and “expert systems.” The field so far has been strongly procedural and rule-bound because both artificial intelligence and ex- pert syStems must be programn’ied.i5 When designers of expert systems seek to replicate in their systems the decisions of professionals such as doctors, geologists, chemists, pharmacologists, and Stockbrokers, thesq i | 20 Why social science has failed as science professionals have a difficult time explaining to the system designers what they are doing in terms of specific procedures and rules. Donald Schon has described the! problem: “When [the professional practitioner] tries, on rare occasions, to say what he knows — when he tries to put his knowing into the form of knowledge — his formulations of principles, theories, maxims, and rules of thumb are often incongr'uent with the understanding and know-how implicit in his pattern of pra.ctice."“5 e The Dreyfus model enables us to understand why: virtuosos simply do not use rules. They reco ize thousands of cases directly, holistically, and intuitively on the basis f their experience. In the preface to the revised paperback edition of the r Mind over Machine, on the basin of experience from neural networks, [ireyfus and Dreyfus downgrade the importance of storing memories and recognizing similarities. Instead, because of her or his experience, “the expert holistically discriminates among classes of situations and associates with these classes appropriate responses” (em- phasis in original).” The rules for expert systems are formulated only because the systems denlland it. They are characteristics of the systems, but not of the real experts. Research shows that heuristic expert systems, being rule—based, are unable to go further than level three in the learning process. The heuristic systems cannot make the qualitative jump to levels four and five and therefpre never become as skillful as hUman experts. This conclusion also applies when the systems are compared with the behavior of the same experts who gave the rules to the system builders. In this sense, the term “eatpert systems” is a misnomer. In. terms of the Dreyfus model they are nolmore than “competent systems." 'Only in areas which are content-independent, which can be strictly separated from daily understanding and from change, and which have well—defined problems with clear rules for their solution, only in these rare areas, and in tasks where brute comp tational number crunc mg can so ve ems, will expert systems succe d as well or better than human experts. ests of existing expert systems support this conclusion. Rationality, irrationali , arationality The'five levels in the learning process can be briefly summai-ized as follows. 5 i (l) Novices act on the basis of context—independent elements arid rules. (2) Advanced beginners also use situational elements, which ey have learned to identify and interpret on the basis of their man ex lerience from similar situations. 1 (3) Competent performers are characterized by the involVed choice of i “a )2 Rationality, body, and intu'tion in human learning goals and plans as a basis for their actions. Goals and plans tn‘ci used: to structure and store mass s of both context-dependent andicun--- text—independentinformatio . Proficient performers identi problems, goals, and plans intuitively (4) from their own experientiall based perspective. Intuitive choice is; checked by analytical evalua ion prior to action. Finally, experts’ behavior 3 intuitive, holistic, and synchrjonic, (5) understood in the way that a given situation releases a pictiire of I problem, goal, plan, decisio , and action in one instant and with no division into phases. This is the level of true human expertisei. Ex-' perts are characterized by a owing, effortless performance, unhinu: dered by analytical deliberati ns. The Dreyfus model contains a qualitative jump from the three first to the' fourth and fifth levels. The jump implies an abandonment of rule—lbased thinking as the most important trasis for action, and its replacement by context and intuition. Logically biased action is replaced by experieritially based action. Dreyfus and Dreyfus provide several conceptual and empirical! of their model. They also describd how man learning fit with it. They conclude something other than calculated,l ana- e often hear the opposite. From the examples to illustrate the validity the results of recent research in h that intelligent action consists o lyrical rationality, even though perspective of the Dreyfus mod 1, analytical rationality is a limitejd ra-. wer levels in the peRormance of aiskill,i tionality: it is appropriate to the l but not to high-level performanc The best performances withi a given area require a qualitatively different expertise based on intui ion, experience, and judgment. A's yet, there exist no computer program which have succeeded in capturin' and simulating this expertise. Intuit on is the ability to draw directfy on one’s own experience — bodily, e otional, intellectual - and to recognize similarities between these exper ences and new situations. Intuition is internalized; it is part of the in ividual. Existing research provides no evidence that intuition and jud ent can be externalized into rulejs and 0 explanations, which, if followe , lead to the same result as intuitive: behavior. Such externalization is possible only for analytical rationality, that is, for those skills which cha acterize the lower levels in the learning process. I That conventional rationality s not the ultimate outcome of human. learning procc3ses does not mea , however, that one necessarily ends in. processes indicates that the cohven—' tional Opposition between ration ity and irrationality is inadequate for an, irrationality. Research in leami 22 Why social s¢ience has failed as science understanding of wlhat actually happens when individualsi understand and act. In order t bridge this gap, Dreyfus and Direyfus invoke the e concept “arational.’ The word “rational,” from the Latin ratio, means to calculate or reason. Rationality in the West has becoime identical with analytical thinking, that is, with conscious separation oflwholes into parts. ° Arational behavior, in contrast, connotes situational behavior without the conscious analytical division of situations into parts and evaluation ac- cording to context-independent rules. Dreyfus and Dreyfus link increas- ing levels of skill acq isition with a relatively decliningllevel of analytical rationality: “001'anth performance is rational; proficiency is transi— tional; experts act a11ationally.”18 In the present context, the interesting point is that the Dreyfus model and “arationality” .accond a central importance to context in the development of knowledgei and skills. As will be shown in the next chapter, this has radical implications for social scrence. ' The Dreyfus mod can be criticized for being slightly mechanistic and insensitive to issues 0 creativity, innovation, and poweri19 HOWever, such weaknesses do not etract from our use of the model in the current context. The argume t in the following chapters is based on only a single property of the modei: and this property is convincinglygestablished in its original form; namely, the qualitative jump from the niodel’s first three stages to the two last stages, that is, from rule-based, context-indepen— dent to experience-based, situational behavior. Other properties of the model are irrelevant for our purposes. ! On closer examina ion, the qualitative difference between rule-based and experience-basedibehavior shows itself to have radical consequences, l in that every rule-based, rational mode of conceiving of human activity — i be this activity scien ific, practical, or didactic— collapses when con- \ fronted with the Dreyi‘us phenomenology. This is the mddel’s critical and dcconstructive perspective, a perspective which caused Iiirgen Haber- mas, after having heard Hubert Dreyfus present the model to him at Frankfurt University, to exclaim, “you are talking about skills like ham- mering and playing chess, but What you really want to 4:10 is undermine ' Western society.” To which Dreyfus replied, “you are right, that’s exactly what it comes to?“ i 5 The Dreyfus model shows how the rational mode of ihinkihg is inad- equate for comprehending the total spectrum of human iactivity, both in relation to human everyday activities and to rare virtuosb performances. Inslesd, the rationalist: perSpective focuses on those properties of human :lt‘l ivity by which humans most resemble machines or Weberian bureau- cmls: rule—based delib ration based on formal logic. 'l he Dreyfus mode has not only critical implications, hoyvever. Its “(it lit l1 ma] value — whic in this context is more important+is constructive. Rationality, body, and intui The model makes clear that what not lie in the rationalists’ emphasi phenomena. These are importani Rather, the rational fallacy consis ltion in human learning We could call the “rational fallacy“ lLll u-r. i on analysis and rationality as impul'lnnl , also according to the Dreyfus model. ts of raising analysis and ratiOnality into the most important mode of operation for human activity, and allowing these to dominate our view of human activity: so much so that ether? equally important modes of human understanding and behavior are made invisible. The Dreyfus model dbes not present a situation of “ditheri rationality or intuition” but of both of them in their proper context: the i position of intuition is not beyond rationality but alongside it, comple-l mentary to it, and insofar as we speak of experts, above rationality.§ The i model specifies that What is need d in order to transcend the insuffilcienti rational perspective is explicit integration of those properties characteristic : of the higher levels in the learning process which can supplement and take over from analysis and rationality. These properties include context,i judgment, practice, trial and error, experience, common sense, intuition, and bodily sensation. ' Context, experience, and intuition In the introduction to this chapter I described a study in which a group of paramedics had been asked, “wh would you choose to revive you you had been victim of an accident?” he experienced paramedics, that is, the practitioners, knew what was good for them and chose the experienced rescuer even though this individual appeared in only one of the six video: films shown. Practical experien¢e consists precisely in an individual’s. ability to readily recognize skill and virtuoso expertise. Teachers in riescu-i ing life were especially unable to identify the expert paramedic, theyl were even worse than the group of inexperienced trainees. After the review of the Dreyfus model we can understand why, The teachers attempted to identify a c mpetent rescuer by looking for individ- uals who best followed the rules students in CPR. The teachers’ nique was simply to follow the rulles. They tended to identify the inexperi- enced students on the films as “ e teachers themselves had taughii their concept of “good” resuscitation itech-. 00d” because, as novices, they closely: and consciously followed the ml 5 they had learned. Being novices, the; students could do little else. In 7 percent of the cases the teachers could not identify the experienced res er because this individual, being truly; experienced - an expert — had go e beyond rule-based behavior. i The example would be trivi if the problem of the dominance oil rule—based rationality over practi 21 experience pertained only to teaching in the health sector or only in th United States. Regrettably, the pervai siveness of the rational paradig to the near exclusion of others is 12-1 Why social science has failed as science problem for the vast majority of professional educationl and especially in practical fields such as engineering, policy analysis, management, plann— ing, and organizatiofi. All are professions where practical skill occupies central importance ut has been threatened by epist¢mic science and 1 didactics. Law is an exception. The practice of law caninot be decontex— tualized to the same degree as other disciplines and has therefore never been made “scientific” to the same extent. : As for the teachers in heart-lung resuscitation, the rule«basled, rational mode of thinking ge erally constitutes an obstacle to food results, not because rules and ratlilonality are problematic in themse Ves, but because the rational ecti e has been elevated fr m bein to bein sufficient, even excl ' ' e scholarly , experience, anii intuition, even though these phenomena and, ways of being are at least as important and necessary for good results as are analysis, rationality, and rules. In part, this is the problem Nietzsche points to when he stresses that “the growth of consciousness bec¢mes a danger”, the faculty of cohsciousness may marginalize those fac lties making true human expertise possible.21 It is also one reason that ietzsche is highly critical of cen a1 tehets in the thought of Socrates o regarded explicit rational und rstanding as the ' highest human acco plishment. “Socrates was a misimderstanding,” i Nietzsche writes, “rat' nality at any cost . . . in opposition to the instincts, has itself been no more than a form of sickness.”22 As an antidote to Socrates, N ietzeche s gests that the central task for hu an beings is not the Socratic one of m king knowledge cerebral and rail}: ut instead one of making it bodi and intuitive. In Nietzsche’s owh words what is central is “the task of z commuting knowledge and making it instinctive,” a task Nietzsche regrets “is only beginning to dawn on thd human eye and is not yet clearly discernible”23 (emphasis in original). The Dreyfus model helps make this task clear. j i The conclusion that rationality may endanger sensitivity to context, experience, and intuitibn is important for teaching, and teaching can be directly compared with the model for human learning:r However, the conclusion also applies to scientific research, even though it demands a more complex argumentation. In the following two chapters, I will deal with the implications of arationality to social science, using the con- clusions from this chap er to evaluate social science theory' and methodol- ogy. We will see a wh 1e gamut of key scientific notions collapses when subjected to the model 3 critical pempective. Then the cdnstrdctive per— spective creates the point of departure for the development of an alterna- tive concept of social science, one based on context, iudgrnent, and practical knowledge. 3 .':.I E. 0" ...
View Full Document

This note was uploaded on 01/16/2009 for the course HONR Hon taught by Professor Unknown during the Fall '09 term at Pacific.

Page1 / 16

Flyvbjerg - rationality - final eve/Wharf, ' jg annual?...

This preview shows document pages 1 - 16. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online