Accueil
"Few ideas are so bizarre that no evidence can be found to support them. Reasons to believe are quite easy to find. A much more telling test is to search for reasons not to believe, to seek evidence that does not fit. In good science the most persuasive ideas are those that survive repeated attempts to prove them wrong."
"Science thrives on the free exchange of ideas and on intellectual competition. It stagnates when, as happened in the Middle Ages under the Catholic Church and under Stalin in the Soviet Union, an outside agency tries to impose on scientists an orthodoxy that is not rooted in the work of the discipline. In the nineteenth century some geneticists argued that characteristics that individuals acquired during their lifetimes could be passed on through their genes. French biologist J.-B. Lamarck believed that the giraffe owed its long neck to the habit of stretching to feed on tree leaves. The counter-case would suppose that 'long-necked-ness' was already a genetically encoded quality and that those giraffes that possessed it had a better chance of survival than those that did not. So genetic stock changes through 'natural selection' rather than learning. By the 1920s the Lamarckian view had been largely abandoned. However, it survived in the Soviet Union, where the natural-selection alternative was thought to be too close to the logic of capitalism and hence politically unacceptable. T. D. Lysenko used his political position to have Lamarckism incorporated in official Communist philosophy, and those geneticists who opposed him were either forced to recant or exiled to Siberia. Only in the 1950s did Soviet biology recover from Lysenko's influence. Sadly, the official sponsorship of the wrong side of the argument blighted not only Soviet biology but also the Soviet economy. The ideologically justified rejection of 'bourgeois' genetics cut the Soviet Union's agriculture off from the great advances in crop development enjoyed in the West."
"If we begin our description of the sociological enterprise by saying that it should model itself on the methods of the physical sciences, we cannot proceed very far before recognizing some fundamental limits to such imitation.
The first is that social scientists can rarely construct experiments. While researching the Ulster Defence Association (UDA) and the Ulster Volunteer Force (UVF), the two main loyalist terrorist organizations in Northern Ireland, I became interested in how certain people had come to occupy important leadership roles. Having found out everything I could about the leaders (and about those who might have been regarded as leadership material but never made it), I came to a number of tentative conclusions. Contrary to what one might expect of terrorist organizations, it was not personal viciousness that kept leaders in office. Out of some thirty cases, I could find only two people who had ruled by fear. One of them was murdered by his own people as soon as his more senior protectors had lost office; the other would have been had he not been arrested and imprisoned. What was much more important than naked coercion was the talent of being able to persuade and reconcile. However, this skill seemed to be common among UDA and UVF leaders across the twenty-five years of the organization, which left unexplained a major difference in the background of those who commanded the UDA and UVF in the 1970s and those who replaced them in the mid-1980s.
While diplomacy was a general requirement, social status was important for the first decade but not thereafter. The first generation of leaders were almost always people who had occupied some sort of community leadership role before the Troubles broke out and the Protestants of working-class areas started to organize themselves into vigilante groups. They had held office in trade unions, community associations, the Ulster Unionist party, and housing associations. The men who came to prominence in the late 1980s were very different. Most had grown up in the terrorist organizations and had come to the fore because they were 'operators' - ruthless killers and planners of terrorist actions or of such necessary subsidiary activities as fund-raising by bank-robbing, extortion, and drug-dealing.
The differences between the generations led me to the following conclusion. In a new enterprise, where no one has experience or can point to a track-record, general marks of status or competence are called on to determine leadership. In modern educational jargon, leadership is taken to be a transferable skill. But once an enterprise has been going for long enough for large numbers to have gained experience of its core activities (in this case, planning or carrying out murders and related criminal acts), then it becomes possible to judge potential candidates for leadership on the core skills of the organization. So the attention of members shifts from very general marks of competence (such as having been prominent in some other community activity) to more specific task-related attributes.
This explanation could be quite wrong. What matters for my purposes here is how I could further test my idea. The chemist studying bromide reactions could have devised further experiments that held constant what were taken to be extraneous variables and focused changes on just those things that were thought to be central. But I could not, for experimental purposes, take a previously stable society and start a minor civil war. No reader should need persuading that the pursuit of social-scientific knowledge cannot justify terrorism. Even if I had had no ethical scruples, it would have been impractical. I had neither the wealth nor the power to start a small war and motivate people to take part in it.
However, let us imagine that both ethical and practical obstacles had been overcome. Creating my own terror group would still not have produced data comparable to those from the repeated experiments of the bromide chemist, because my terror group would not have been the same as the 'naturally occurring' ones I wished to understand. There are two problems. One is that artificial experiments in the social sciences have a fundamentally different relationship to the real world than chemistry experiments because the social experiment is not a facsimile of the naturally occurring: it is itself a novel social event. The other issue is that social life appears to be too complex to be broken down into simple component parts that can then be examined in isolation.
So one major difference between the natural and social sciences is that the ideas of the latter cannot normally be rigorously tested by being subjected to experiments that isolate the features of human action that interest us from the complexities of ongoing life. However, we can and often do perform quasi-experiments in which we try to compare the action that interests us in a variety of settings that are mostly similar but different in just one or two key ways. The work of Rosabeth Kanter on Utopian communities provides a good illustration. She wanted to know why some communes succeeded while others failed. Her extensive reading of the history of such groups and her own involvement in the communes of the 1960s had given her some general ideas about which features of such engineered societies might work.
So she began with some hypotheses, derived from previous scholarly work and shaped by her own unsystematic observation, and then sought a test of those ideas. To avoid the effects of differences in the communities being swamped by differences in their surrounding societies, she concentrated on communes that had been formed in one country within a relatively short time period: the United States between 1780 and i860. She managed to identify ninety such communities: eleven 'successes' that had survived twenty-five years (the conventional view of a generation) and seventy-nine 'failures' that had not lasted a quarter of a century. She concluded that, although there was no short list of properties that were present in all the successes and uniformly absent from the failures, there were characteristics that were common in almost all of the communities that had survived one generation and were rare in the failures. The successes demanded various forms of sacrifice (such as abstinence from sex, alcohol, and dancing) from their members. They had world-views that drew hard lines between the good people of the commune and the rest of the world. They had very strict definitions of membership and rigorous membership tests. New members were required to prove their commitment by investing a great deal of time and money in the enterprise, which in turn made it costly to defect. Almost all the successes bolstered this psychic and social separation from the world with geographical isolation. Kanter concluded that commitment was not a mysterious phenomenon that had to precede the formation of a Utopian community. Rather it was a social property that could be engineered by the deliberate use of what she called 'commitment mechanisms'.
Researchers since have modified Kanter's conclusions. I have argued that it is easier to engineer commitment to some sorts of belief-systems than to others. Those political philosophies and religions that vest supreme authority in the individual are far more difficult to organize than those that can evoke some higher power. Conservative Catholics and Protestants can form successful communities ; liberal Protestants and devotees of the New Age cannot. However, here I am more interested in Kanter's method than in her conclusions. She very ably demonstrates that, while we cannot experiment as easily as the natural scientist, with some imagination we can find 'naturally occurring' data, examples from real life, to simplify social phenomena. Social scientists routinely do this with large-scale social surveys. Imagine we want to know what effect gender has on political preferences. We could ask large numbers of men and women how they voted in the 1997 election that brought Labour to power in Britain after eighteen years of Conservative rule, and compare the answers.
However, if we stopped there we would learn very little, because other characteristics such as income, levels of education, race, and religion also affect political preferences. So we would ask our men and women further questions that allowed us to assign them labels for levels of income, years in formal education, ethnic identity, religious affiliation, and so on. We could then use statistical methods to work out which of these characteristics, either on its own or in combination, has the greatest effect on voting behaviour.
While such research is illuminating, its conclusions are always tentative and probabilistic. We can say with confidence that working-class people are more likely to lean to the left politically than the upper classes. But there are enough exceptions to that proposition to stop us treating it as if it were a natural law. In the 1950s it was possible to identify a group of 'deferential workers' that, though we might want to call it 'objectively' working class, was none the less extremely conservative in its politics and supposed that the upper classes would make a better job of running the country than the representatives of the workers. In the 1980s Margaret Thatcher's brand of conservatism (laissez-faire on economics but authoritarian on social issues) drew strong support from sections of the working class in the prosperous south-east of England. So we start with a simple expectation and find that it needs to be refined. Simple divisions of people by type of occupation (such as manual and nonmanual work) are not powerful predictors of voting. So we further divide class or we add other considerations, but we find that our propositions never move beyond probabilities.
Some sociologists take such failures as encouragement to become more sophisticated in the definition, identification, and measurement of what are taken to be the causes of social action. While improvement in those three areas is all to the good, sociology's failure to produce Maws' reflects far more than its relative immaturity. After a century of scientific sociology, the it's-early-days- yet' defence sounds rather thin. More research and more sophisticated methods of analysing the data we collect will make us better informed, but we will never discover the laws of human action because people are not like atoms.
The subject matter of the social sciences is conscious sentient beings who act out of choice. At this stage we do not need to get bogged down in well- rehearsed arguments about the extent to which people are really 'free'. All we have to recognize is that, whatever the sources of uniformity in human behaviour (and more of that later), they are not 'binding' in any absolute sense. The most oppressive regime may constrain us so tightly that we can choose only between conformity and death, but we can still choose the latter. This distinguishes us utterly from the subject matter of the natural sciences. Water cannot refuse to have its volatility increased as it is heated. With pressure held constant, water cannot boil at 1oo°C for four days and then refuse to do so on the fifth day. People can. Even the lowest worm can turn.
This leads us to recognize that what counts as explanation in the social sciences is quite unlike explanation in physics or chemistry. We explain why the kettle boils by citing the general laws of pressure, temperature, and volatility. Because the water has not decided to boil (a decision that it could change on some other occasion), we do not need to refer to the consciousness of the water. If we wish only to identify some very broad regularities of human behaviour, then we can treat social characteristics like the variables of natural science and propose, for example, that unskilled workers are more likely than businessmen to vote socialist, but if we wish to explain why that is the case then we have to examine the beliefs, values, motives, and intentions of the people in question. Because the human consciousness is the engine that drives all action, the social sciences have to go further than the natural sciences. When the chemist has repeatedly found the same reactions in his bromides, he stops. Identifying the regularity is the end of that search. For the social scientist it is only the beginning. Even if we found that everyone in a particular situation always did a particular thing (and such strong regularities are almost unknown), we would want to know why."
"The sociologist's interest in beliefs, values, motives, and intentions brings with it concerns unknown in the natural sciences. In order to understand people, we need to solicit their views or 'accounts' of what they are doing. Furthermore, we can take the same point back one step and note that it is not just understanding that requires some interest in motives. Even identifying the social act we wish to understand requires attention to motives. Let us go back to our pan of water. There are ways of defining when a liquid changes to a gas that make no reference to its state of mind. But the actions of people cannot be identified simply by observing them. Or, to put it another way, the action itself is not enough. Suppose we are interested in how people interact in public places. We could sit at a table in a crowded railway station and watch and take notes. But if we confined ourselves only to what was visible, we would learn little. We could note 'man facing platform raises arm in air and moves it from side to side'. We could not say 'man waves to greet incoming passenger', because that second description gives a particular interpretation to the physical action. He might actually be trying to relieve a trapped nerve."
"The accounts that people give can be both honest attempts to reconstruct past motives and performances through which they pursue present interests. In some settings the distortion is obvious. We can be sure that the story people give of their actions during their defence in court, or in pleading for mitigation after admitting guilt, will be quite different from the version they give to friends and family after they have been found 'Not guilty' or avoided a custodial sentence. The person telling the story has interests in the outcome of the telling and the court itself requires stories to be told in an unusually stylized manner. I am not saying that the formal courtroom version is false and the informal version true. What I am saying is that giving an account is itself a social activity and not merely an explanation of earlier activities."
"What people say and what they actually did might be linked in four ways. First, respondents may not recall or understand their motives. Secondly, they may recall or understand all too well but deliberately dissemble. The nineteenth-century industrialist J. P. Morgan hit on an important point about the desire to appear decent and honourable when he said: 'For every act there are two reasons: a good reason and the real reason.' Thirdly, whatever the level of self-understanding and willingness to be honest, the setting for the giving of accounts may exert such an influence that we cannot with confidence use what people say as a guide to their previous mental states. Individual variety might be funnelled into apparent consensus [...] Between these last two types, we can place a fourth case: collective dissembling. Often a group of people share what for them are good reasons for their actions but routinely 'explain' what they do by calling on a publicly more acceptable language of justification. For example, doctors may ration expensive treatments for renal failure or lung cancer by informal moralizing about what sort of person deserves their attention, but then avoid having to defend such reasoning by claiming that decisions were made solely on grounds of the likelihood of the intervention being successful."
"So far the differences between the natural and social sciences have been discussed to the disadvantage of the latter. I would now like to suggest a very different conclusion. Consider the position of racehorse trainers. Long experience may give them confidence that they can put themselves in the horses' hooves. But social scientists begin with the great advantage of sharing biology, psychology, and a great deal of culture with their subjects. I have never been a member of a terrorist organization and I have oriented my life around avoiding assassination and arrest. But I can find in my own experience causes to which I have been strongly attached, events that have caused me great fear and anger, and actions of which I am proud and others of which I am profoundly ashamed. Even when those we study seem as distant as citizens of a foreign country, there is enough in our common humanity to create countless border-crossings. We may misunderstand, but there will be opportunities to clear up confusion. Whatever analytical purchase is lost by us not being able to experiment is amply regained by our ability to sustain extended conversations with our subjects. I could not test my ideas about terrorist career structures experimentally, but I could, directly and indirectly, raise them with my respondents."
"We could describe sociology as the study of social structures and social institutions, and sociological work is often divided into such topics as the class structure of modern societies, the family, crime and deviance, religion, and so on. However, to list what we study gives no sense of what is distinctive about the way we do it. Like a charm bracelet, this account of sociology will hang a number of substantial observations from a central thread made of the following strands: reality is socially constructed, our behaviour has hidden social causes, and much of social life is profoundly ironic."
"An easy way to dismiss the more extreme forms of biological determinism is to point to the many ways we deliberately reject instincts. There may be a will to live but we can commit suicide. There may be a will to reproduce, but women can choose not to have children and still live apparently fulfilled lives. There may be a sex urge, but celibacy is possible. The claims for biology are further weakened if we note the considerable cultural differences in what might be instinctual. Not only do people kill themselves but the suicide rate differs from one society to another, as does the frequency of childlessness. Whatever part instinct plays in our lives, it is complicated by cultural variations."
" [Regulation is] a moral force, a shared culture that specifies what we can desire and how we can attain those goals, takes the place of the biological straitjacket. To fill the gap left by what Gehlen called 'instinctual deprivation', people create social frameworks. Some parts of those frameworks may be fixed in formal law. The bulk of it is merely conventional. There is no law that says that white-collar workers in management positions should wear dark two-piece suits, but every aspirant to senior management knows how to dress. At its most effective, the straitjacket is applied not to the outside of the body but to the inside of the mind. We are socialized in the culture so that important elements of it become embedded in our personalities.
If we can see the importance of culture in giving a framework within which the individual can achieve contentment, the third problem of world-openness - coordinating joint action should be even more obvious. Where, as with ants and bees, communication and coordination are themselves biological, there is no difficulty. One ant does not need to interpret the signals given off by another. It responds automatically to the secretions."
"Human biology does nothing to structure human society. Age may enfeeble us all, but cultures vary considerably in the prestige and power they accord to the elderly. Giving birth is a necessary condition for being a mother, but it is not sufficient. We expect mothers to behave in maternal ways and to display appropriately maternal sentiments. We prescribe a clutch of norms or rules that govern the role of mother. That the social role is independent of the biological base can be demonstrated by going back three sentences. Giving birth is certainly not sufficient to be a mother but, as adoption and fostering show, it is not even necessary !
The fine detail of what is expected of a mother or a father or a dutiful son differs from culture to culture, but everywhere behaviour is coordinated by the reciprocal nature of roles. Husbands and wives, parents and children, employers and employees, waiters and customers, teachers and pupils, warlord and followers ; each makes sense only in its relation to the other. The term 'role' is an appropriate one, because the metaphor of an actor in a play neatly expresses the rule-governed nature or scripted nature of much of social life and the sense that society is a joint production."
"We make life manageable by creating social institutions that do for us what instincts do for other animals. By routinizing programmes of action and either painting them onto the 'backcloth' or writing them into a script, we can leave free for creative improvisation and conscious choice an area that is small enough for individuals and groups to manage without becoming overwhelmed.
However, though we may on calm reflection see the virtues of allowing large parts of our lives to follow well-worn paths, modern people periodically feel themselves frustrated by the impersonality and predictability of life. As Laurie Taylor and Stan Cohen illustrate in Escape Attempts, we often try to distinguish between the social roles we play and the real 'us'. Like Sartre's waiter, we perform in such a way as to show to our audiences that we are more than, and can rise above, our roles as managers, civil servants, bus drivers, fathers, and loyal spouses. We may use hobbies, holidays, and weekend trips to establish a persona separate from our place in the paramount reality of everyday life."
"Even when objective stimuli are implicated in our actions, it is our interpretations of those stimuli that affect our behaviour. Consider the way we 'get drunk'. It is unlikely that there are major differences in the way that the Australian Aboriginal farmer, the New York businessman, the Scottish medical student, and the Italian child metabolize alcohol, but there are huge variations in how these peoples behave when they drink. I do not mean just that cultures differ in attitudes to drunkenness, though they do. What is acceptable for a fishing crew returned from a week in the north Atlantic is not what one would expect at a business lunch in Tokyo. I mean that the 'overlay' of culture is such that different peoples expect alcohol to affect them in different ways and as a result do indeed feel different. The amount of alcohol that in one context produces staggering, incoherent speech, and incontinent giggling can in another produce quiet reflection and feelings of peace. Or, to put it another way, we learn what to expect and by and large find it. As Howard Becker argued in his seminal essay 'Becoming a Marijuana User', the same objective sensation could be interpreted as elation or nausea and learning to feel the former rather than the latter is a crucial part of becoming a dope smoker."
"The realm that interests sociologists is neither 'all in the mind' nor entirely external to our consciousness: it is inter-subjective. Things that people imagine, provided they are imagined similarly by large enough numbers of people, can have an enduring and even oppressive reality indistinguishable from the 'objective' world. In considering how we explain our actions, the American social psychologist W. I. Thomas wrote that, if people define situations as real, then they are real in their consequences. The man who believes his house is on fire will run from it. That the house does not burn down proves his belief was wrong, but none the less, to understand why the man evacuates his house what matters is his belief and not the 'truth'."
-Steve Bruce, Sociology. A Very Short Introduction, Oxford University Press, 1999.
"Science thrives on the free exchange of ideas and on intellectual competition. It stagnates when, as happened in the Middle Ages under the Catholic Church and under Stalin in the Soviet Union, an outside agency tries to impose on scientists an orthodoxy that is not rooted in the work of the discipline. In the nineteenth century some geneticists argued that characteristics that individuals acquired during their lifetimes could be passed on through their genes. French biologist J.-B. Lamarck believed that the giraffe owed its long neck to the habit of stretching to feed on tree leaves. The counter-case would suppose that 'long-necked-ness' was already a genetically encoded quality and that those giraffes that possessed it had a better chance of survival than those that did not. So genetic stock changes through 'natural selection' rather than learning. By the 1920s the Lamarckian view had been largely abandoned. However, it survived in the Soviet Union, where the natural-selection alternative was thought to be too close to the logic of capitalism and hence politically unacceptable. T. D. Lysenko used his political position to have Lamarckism incorporated in official Communist philosophy, and those geneticists who opposed him were either forced to recant or exiled to Siberia. Only in the 1950s did Soviet biology recover from Lysenko's influence. Sadly, the official sponsorship of the wrong side of the argument blighted not only Soviet biology but also the Soviet economy. The ideologically justified rejection of 'bourgeois' genetics cut the Soviet Union's agriculture off from the great advances in crop development enjoyed in the West."
"If we begin our description of the sociological enterprise by saying that it should model itself on the methods of the physical sciences, we cannot proceed very far before recognizing some fundamental limits to such imitation.
The first is that social scientists can rarely construct experiments. While researching the Ulster Defence Association (UDA) and the Ulster Volunteer Force (UVF), the two main loyalist terrorist organizations in Northern Ireland, I became interested in how certain people had come to occupy important leadership roles. Having found out everything I could about the leaders (and about those who might have been regarded as leadership material but never made it), I came to a number of tentative conclusions. Contrary to what one might expect of terrorist organizations, it was not personal viciousness that kept leaders in office. Out of some thirty cases, I could find only two people who had ruled by fear. One of them was murdered by his own people as soon as his more senior protectors had lost office; the other would have been had he not been arrested and imprisoned. What was much more important than naked coercion was the talent of being able to persuade and reconcile. However, this skill seemed to be common among UDA and UVF leaders across the twenty-five years of the organization, which left unexplained a major difference in the background of those who commanded the UDA and UVF in the 1970s and those who replaced them in the mid-1980s.
While diplomacy was a general requirement, social status was important for the first decade but not thereafter. The first generation of leaders were almost always people who had occupied some sort of community leadership role before the Troubles broke out and the Protestants of working-class areas started to organize themselves into vigilante groups. They had held office in trade unions, community associations, the Ulster Unionist party, and housing associations. The men who came to prominence in the late 1980s were very different. Most had grown up in the terrorist organizations and had come to the fore because they were 'operators' - ruthless killers and planners of terrorist actions or of such necessary subsidiary activities as fund-raising by bank-robbing, extortion, and drug-dealing.
The differences between the generations led me to the following conclusion. In a new enterprise, where no one has experience or can point to a track-record, general marks of status or competence are called on to determine leadership. In modern educational jargon, leadership is taken to be a transferable skill. But once an enterprise has been going for long enough for large numbers to have gained experience of its core activities (in this case, planning or carrying out murders and related criminal acts), then it becomes possible to judge potential candidates for leadership on the core skills of the organization. So the attention of members shifts from very general marks of competence (such as having been prominent in some other community activity) to more specific task-related attributes.
This explanation could be quite wrong. What matters for my purposes here is how I could further test my idea. The chemist studying bromide reactions could have devised further experiments that held constant what were taken to be extraneous variables and focused changes on just those things that were thought to be central. But I could not, for experimental purposes, take a previously stable society and start a minor civil war. No reader should need persuading that the pursuit of social-scientific knowledge cannot justify terrorism. Even if I had had no ethical scruples, it would have been impractical. I had neither the wealth nor the power to start a small war and motivate people to take part in it.
However, let us imagine that both ethical and practical obstacles had been overcome. Creating my own terror group would still not have produced data comparable to those from the repeated experiments of the bromide chemist, because my terror group would not have been the same as the 'naturally occurring' ones I wished to understand. There are two problems. One is that artificial experiments in the social sciences have a fundamentally different relationship to the real world than chemistry experiments because the social experiment is not a facsimile of the naturally occurring: it is itself a novel social event. The other issue is that social life appears to be too complex to be broken down into simple component parts that can then be examined in isolation.
So one major difference between the natural and social sciences is that the ideas of the latter cannot normally be rigorously tested by being subjected to experiments that isolate the features of human action that interest us from the complexities of ongoing life. However, we can and often do perform quasi-experiments in which we try to compare the action that interests us in a variety of settings that are mostly similar but different in just one or two key ways. The work of Rosabeth Kanter on Utopian communities provides a good illustration. She wanted to know why some communes succeeded while others failed. Her extensive reading of the history of such groups and her own involvement in the communes of the 1960s had given her some general ideas about which features of such engineered societies might work.
So she began with some hypotheses, derived from previous scholarly work and shaped by her own unsystematic observation, and then sought a test of those ideas. To avoid the effects of differences in the communities being swamped by differences in their surrounding societies, she concentrated on communes that had been formed in one country within a relatively short time period: the United States between 1780 and i860. She managed to identify ninety such communities: eleven 'successes' that had survived twenty-five years (the conventional view of a generation) and seventy-nine 'failures' that had not lasted a quarter of a century. She concluded that, although there was no short list of properties that were present in all the successes and uniformly absent from the failures, there were characteristics that were common in almost all of the communities that had survived one generation and were rare in the failures. The successes demanded various forms of sacrifice (such as abstinence from sex, alcohol, and dancing) from their members. They had world-views that drew hard lines between the good people of the commune and the rest of the world. They had very strict definitions of membership and rigorous membership tests. New members were required to prove their commitment by investing a great deal of time and money in the enterprise, which in turn made it costly to defect. Almost all the successes bolstered this psychic and social separation from the world with geographical isolation. Kanter concluded that commitment was not a mysterious phenomenon that had to precede the formation of a Utopian community. Rather it was a social property that could be engineered by the deliberate use of what she called 'commitment mechanisms'.
Researchers since have modified Kanter's conclusions. I have argued that it is easier to engineer commitment to some sorts of belief-systems than to others. Those political philosophies and religions that vest supreme authority in the individual are far more difficult to organize than those that can evoke some higher power. Conservative Catholics and Protestants can form successful communities ; liberal Protestants and devotees of the New Age cannot. However, here I am more interested in Kanter's method than in her conclusions. She very ably demonstrates that, while we cannot experiment as easily as the natural scientist, with some imagination we can find 'naturally occurring' data, examples from real life, to simplify social phenomena. Social scientists routinely do this with large-scale social surveys. Imagine we want to know what effect gender has on political preferences. We could ask large numbers of men and women how they voted in the 1997 election that brought Labour to power in Britain after eighteen years of Conservative rule, and compare the answers.
However, if we stopped there we would learn very little, because other characteristics such as income, levels of education, race, and religion also affect political preferences. So we would ask our men and women further questions that allowed us to assign them labels for levels of income, years in formal education, ethnic identity, religious affiliation, and so on. We could then use statistical methods to work out which of these characteristics, either on its own or in combination, has the greatest effect on voting behaviour.
While such research is illuminating, its conclusions are always tentative and probabilistic. We can say with confidence that working-class people are more likely to lean to the left politically than the upper classes. But there are enough exceptions to that proposition to stop us treating it as if it were a natural law. In the 1950s it was possible to identify a group of 'deferential workers' that, though we might want to call it 'objectively' working class, was none the less extremely conservative in its politics and supposed that the upper classes would make a better job of running the country than the representatives of the workers. In the 1980s Margaret Thatcher's brand of conservatism (laissez-faire on economics but authoritarian on social issues) drew strong support from sections of the working class in the prosperous south-east of England. So we start with a simple expectation and find that it needs to be refined. Simple divisions of people by type of occupation (such as manual and nonmanual work) are not powerful predictors of voting. So we further divide class or we add other considerations, but we find that our propositions never move beyond probabilities.
Some sociologists take such failures as encouragement to become more sophisticated in the definition, identification, and measurement of what are taken to be the causes of social action. While improvement in those three areas is all to the good, sociology's failure to produce Maws' reflects far more than its relative immaturity. After a century of scientific sociology, the it's-early-days- yet' defence sounds rather thin. More research and more sophisticated methods of analysing the data we collect will make us better informed, but we will never discover the laws of human action because people are not like atoms.
The subject matter of the social sciences is conscious sentient beings who act out of choice. At this stage we do not need to get bogged down in well- rehearsed arguments about the extent to which people are really 'free'. All we have to recognize is that, whatever the sources of uniformity in human behaviour (and more of that later), they are not 'binding' in any absolute sense. The most oppressive regime may constrain us so tightly that we can choose only between conformity and death, but we can still choose the latter. This distinguishes us utterly from the subject matter of the natural sciences. Water cannot refuse to have its volatility increased as it is heated. With pressure held constant, water cannot boil at 1oo°C for four days and then refuse to do so on the fifth day. People can. Even the lowest worm can turn.
This leads us to recognize that what counts as explanation in the social sciences is quite unlike explanation in physics or chemistry. We explain why the kettle boils by citing the general laws of pressure, temperature, and volatility. Because the water has not decided to boil (a decision that it could change on some other occasion), we do not need to refer to the consciousness of the water. If we wish only to identify some very broad regularities of human behaviour, then we can treat social characteristics like the variables of natural science and propose, for example, that unskilled workers are more likely than businessmen to vote socialist, but if we wish to explain why that is the case then we have to examine the beliefs, values, motives, and intentions of the people in question. Because the human consciousness is the engine that drives all action, the social sciences have to go further than the natural sciences. When the chemist has repeatedly found the same reactions in his bromides, he stops. Identifying the regularity is the end of that search. For the social scientist it is only the beginning. Even if we found that everyone in a particular situation always did a particular thing (and such strong regularities are almost unknown), we would want to know why."
"The sociologist's interest in beliefs, values, motives, and intentions brings with it concerns unknown in the natural sciences. In order to understand people, we need to solicit their views or 'accounts' of what they are doing. Furthermore, we can take the same point back one step and note that it is not just understanding that requires some interest in motives. Even identifying the social act we wish to understand requires attention to motives. Let us go back to our pan of water. There are ways of defining when a liquid changes to a gas that make no reference to its state of mind. But the actions of people cannot be identified simply by observing them. Or, to put it another way, the action itself is not enough. Suppose we are interested in how people interact in public places. We could sit at a table in a crowded railway station and watch and take notes. But if we confined ourselves only to what was visible, we would learn little. We could note 'man facing platform raises arm in air and moves it from side to side'. We could not say 'man waves to greet incoming passenger', because that second description gives a particular interpretation to the physical action. He might actually be trying to relieve a trapped nerve."
"The accounts that people give can be both honest attempts to reconstruct past motives and performances through which they pursue present interests. In some settings the distortion is obvious. We can be sure that the story people give of their actions during their defence in court, or in pleading for mitigation after admitting guilt, will be quite different from the version they give to friends and family after they have been found 'Not guilty' or avoided a custodial sentence. The person telling the story has interests in the outcome of the telling and the court itself requires stories to be told in an unusually stylized manner. I am not saying that the formal courtroom version is false and the informal version true. What I am saying is that giving an account is itself a social activity and not merely an explanation of earlier activities."
"What people say and what they actually did might be linked in four ways. First, respondents may not recall or understand their motives. Secondly, they may recall or understand all too well but deliberately dissemble. The nineteenth-century industrialist J. P. Morgan hit on an important point about the desire to appear decent and honourable when he said: 'For every act there are two reasons: a good reason and the real reason.' Thirdly, whatever the level of self-understanding and willingness to be honest, the setting for the giving of accounts may exert such an influence that we cannot with confidence use what people say as a guide to their previous mental states. Individual variety might be funnelled into apparent consensus [...] Between these last two types, we can place a fourth case: collective dissembling. Often a group of people share what for them are good reasons for their actions but routinely 'explain' what they do by calling on a publicly more acceptable language of justification. For example, doctors may ration expensive treatments for renal failure or lung cancer by informal moralizing about what sort of person deserves their attention, but then avoid having to defend such reasoning by claiming that decisions were made solely on grounds of the likelihood of the intervention being successful."
"So far the differences between the natural and social sciences have been discussed to the disadvantage of the latter. I would now like to suggest a very different conclusion. Consider the position of racehorse trainers. Long experience may give them confidence that they can put themselves in the horses' hooves. But social scientists begin with the great advantage of sharing biology, psychology, and a great deal of culture with their subjects. I have never been a member of a terrorist organization and I have oriented my life around avoiding assassination and arrest. But I can find in my own experience causes to which I have been strongly attached, events that have caused me great fear and anger, and actions of which I am proud and others of which I am profoundly ashamed. Even when those we study seem as distant as citizens of a foreign country, there is enough in our common humanity to create countless border-crossings. We may misunderstand, but there will be opportunities to clear up confusion. Whatever analytical purchase is lost by us not being able to experiment is amply regained by our ability to sustain extended conversations with our subjects. I could not test my ideas about terrorist career structures experimentally, but I could, directly and indirectly, raise them with my respondents."
"We could describe sociology as the study of social structures and social institutions, and sociological work is often divided into such topics as the class structure of modern societies, the family, crime and deviance, religion, and so on. However, to list what we study gives no sense of what is distinctive about the way we do it. Like a charm bracelet, this account of sociology will hang a number of substantial observations from a central thread made of the following strands: reality is socially constructed, our behaviour has hidden social causes, and much of social life is profoundly ironic."
"An easy way to dismiss the more extreme forms of biological determinism is to point to the many ways we deliberately reject instincts. There may be a will to live but we can commit suicide. There may be a will to reproduce, but women can choose not to have children and still live apparently fulfilled lives. There may be a sex urge, but celibacy is possible. The claims for biology are further weakened if we note the considerable cultural differences in what might be instinctual. Not only do people kill themselves but the suicide rate differs from one society to another, as does the frequency of childlessness. Whatever part instinct plays in our lives, it is complicated by cultural variations."
" [Regulation is] a moral force, a shared culture that specifies what we can desire and how we can attain those goals, takes the place of the biological straitjacket. To fill the gap left by what Gehlen called 'instinctual deprivation', people create social frameworks. Some parts of those frameworks may be fixed in formal law. The bulk of it is merely conventional. There is no law that says that white-collar workers in management positions should wear dark two-piece suits, but every aspirant to senior management knows how to dress. At its most effective, the straitjacket is applied not to the outside of the body but to the inside of the mind. We are socialized in the culture so that important elements of it become embedded in our personalities.
If we can see the importance of culture in giving a framework within which the individual can achieve contentment, the third problem of world-openness - coordinating joint action should be even more obvious. Where, as with ants and bees, communication and coordination are themselves biological, there is no difficulty. One ant does not need to interpret the signals given off by another. It responds automatically to the secretions."
"Human biology does nothing to structure human society. Age may enfeeble us all, but cultures vary considerably in the prestige and power they accord to the elderly. Giving birth is a necessary condition for being a mother, but it is not sufficient. We expect mothers to behave in maternal ways and to display appropriately maternal sentiments. We prescribe a clutch of norms or rules that govern the role of mother. That the social role is independent of the biological base can be demonstrated by going back three sentences. Giving birth is certainly not sufficient to be a mother but, as adoption and fostering show, it is not even necessary !
The fine detail of what is expected of a mother or a father or a dutiful son differs from culture to culture, but everywhere behaviour is coordinated by the reciprocal nature of roles. Husbands and wives, parents and children, employers and employees, waiters and customers, teachers and pupils, warlord and followers ; each makes sense only in its relation to the other. The term 'role' is an appropriate one, because the metaphor of an actor in a play neatly expresses the rule-governed nature or scripted nature of much of social life and the sense that society is a joint production."
"We make life manageable by creating social institutions that do for us what instincts do for other animals. By routinizing programmes of action and either painting them onto the 'backcloth' or writing them into a script, we can leave free for creative improvisation and conscious choice an area that is small enough for individuals and groups to manage without becoming overwhelmed.
However, though we may on calm reflection see the virtues of allowing large parts of our lives to follow well-worn paths, modern people periodically feel themselves frustrated by the impersonality and predictability of life. As Laurie Taylor and Stan Cohen illustrate in Escape Attempts, we often try to distinguish between the social roles we play and the real 'us'. Like Sartre's waiter, we perform in such a way as to show to our audiences that we are more than, and can rise above, our roles as managers, civil servants, bus drivers, fathers, and loyal spouses. We may use hobbies, holidays, and weekend trips to establish a persona separate from our place in the paramount reality of everyday life."
"Even when objective stimuli are implicated in our actions, it is our interpretations of those stimuli that affect our behaviour. Consider the way we 'get drunk'. It is unlikely that there are major differences in the way that the Australian Aboriginal farmer, the New York businessman, the Scottish medical student, and the Italian child metabolize alcohol, but there are huge variations in how these peoples behave when they drink. I do not mean just that cultures differ in attitudes to drunkenness, though they do. What is acceptable for a fishing crew returned from a week in the north Atlantic is not what one would expect at a business lunch in Tokyo. I mean that the 'overlay' of culture is such that different peoples expect alcohol to affect them in different ways and as a result do indeed feel different. The amount of alcohol that in one context produces staggering, incoherent speech, and incontinent giggling can in another produce quiet reflection and feelings of peace. Or, to put it another way, we learn what to expect and by and large find it. As Howard Becker argued in his seminal essay 'Becoming a Marijuana User', the same objective sensation could be interpreted as elation or nausea and learning to feel the former rather than the latter is a crucial part of becoming a dope smoker."
"The realm that interests sociologists is neither 'all in the mind' nor entirely external to our consciousness: it is inter-subjective. Things that people imagine, provided they are imagined similarly by large enough numbers of people, can have an enduring and even oppressive reality indistinguishable from the 'objective' world. In considering how we explain our actions, the American social psychologist W. I. Thomas wrote that, if people define situations as real, then they are real in their consequences. The man who believes his house is on fire will run from it. That the house does not burn down proves his belief was wrong, but none the less, to understand why the man evacuates his house what matters is his belief and not the 'truth'."
-Steve Bruce, Sociology. A Very Short Introduction, Oxford University Press, 1999.
