Daniel Courgeau

Trois moments de l'histoire de la démographie et de la probabilité sont analysés au prisme de la connaissance savante: la naissance et l'évolution de ces deux sciences au XVIIe siècle, leur impact dans les débats ( XVIIIe - XXe siècle) sur l'inoculation et la vaccination de la variole; enfin la polémique à la fin du XXe siècle entre les démographes et l'interprétation faite par la presse de leurs travaux. Nous examinons les diverses interprétations auxquelles ces disciplines ont été confrontées dans différents contextes car, si leur objet semble à la portée de tous, elles nécessitent cependant un traitement plus complexe pour acceder au niveau de discilines scientifiques.

Par son libellé même, l'épistémologie des sciences reconnait l'existence, au sein de l'ensemble des connaissances humaines, d'une connaissance savante. De nos jours, elle est exclusivement élaborée dans les milieux universitaires et articulée en disciplines relativement autonomes. Elle est soumise à des canons exigeants en matière de rigueur formelle et de valeur empirique, qui s'avèrent satisfaits au moins dans les sciences de la nature. La connaissance vulgaire est d'emblée définie en négatif par rapport à la science qui est l'objet de toutes les intentions et jouit de tous les privilèges. Elle apparait comme un magma hétérogène de connaissances même si elle peut être positionnée à son tour en savoirs de divers types en fonction de son universalité et de sa technicité. L'influence directe de la connaissance savantes sur la connaissance vulgaire est bien cernée dans certains domaines, mais à peine abordée dans d''autres. l'apparition de régularités sociales ayant force de loi à partir des actions individuelles fait l'objet même des sciences sociales, mais se révèle peu robuste. Ce bouclage de la connaissance sur elle-même semble donc plus hypothétique que véritablement réaliste. Cet ouvrage examine les relations entre connaissance savant et connaissance vulgaire, dans les diverses sciences sociales et en phhilosophie.

L'examen de divers exemples permet de montrer que, si la cumulativité des connaissances en démographie est évidente, celle des paradigmes est loin d'être linéaire. Ainsi, le taux de masculinité des naissances conduit à des interprétations de plus en plus complexes du XVIIe au XVIIIe siècle, alors que sa mesure demeure identique. De même, l'additivité des effets de la nature et de la culture, à la base de la génétique du comportement, n'est en fait pas vérifiée et a conduit à la mise ne place de modèles non additifs. Ensuite, le passage du paradigme longitudinal au biographique montre comment il a été nécessaire de lever les hypothèses de base du premier pour pousser plus avant l'analyse. Enfin, l'intégration du paradigme transversal au biographique montre que l'analyse multiniveau qui en résulte n'est en rien une addition des deux précédents. La pluralité des paradigmes permet ainsi d'en donner une vue stéréoscopique, qui restitue les faits démographiques dans une cumulativité no linéaire.

La notion de cumulativité du savoir exprime simplement la possibilité de comparer deux états du savoir alternatifs ou successifs, et d"évaluer si on en sait plus dans l'un que dans l'autre. Si le savoir pouvait se décliner sous forme de propositions parfaitement validées et mutuellement indépendantes, cette comparaison serait relativement aisée, un état du savoir étant plus riche qu'un autre s'il admet des propositions en plus. Cependant, même en adaptant une telle vision axiomatique, le savoir est formé de propositions auxquelles le degré d'adhésion est variable, ce qui rend la comparaison beaucoup plus délicate.

Chapter 9 offers a general conclusion to the whole volume. We return to the various topics discussed and propose a synthesis. The first obvious conclusion is that it is impossible to grasp human life in its entirety; we can only capture a small portion of it. The choice of the portion is therefore essential in order for a social science to analyze it. It is up to researchers to define the limits of that portion, which are often dictated by their field of scientific expertise, the survey’s cost, and its feasibility. We start with a detailed examination of the theories that, across the centuries, assigned a major but not exclusive role to demography. Before the nineteenth century, economics and population science were often practiced in tandem: the populationists believed that an increase in population would be matched by an increase in wealth, while a greater number of authors argued that population depended on wealth. In the late eighteenth century, new concepts emerged, most notably linked to the French Revolution. The notion of the perfectibility of the human species appeared in the writings of many authors such as Godwin and Condorcet, while Malthus developed a theological conception that went in the opposite direction. In the nineteenth century, the industrial revolution elaborated economic theories that made little room for demography. The theory of industrialism crowned the notion of perfectibility, whereas the theories of Proudhon and Marx were essentially economic. It was not until the twentieth century that Landry theorized the demographic revolution—freed from the grip of economics—through the notion of the rationalization of life. Most recently, the late twentieth and early twenty-first centuries have seen the emergence of three main theories that have been adopted in many social sciences but are far from having won unanimous acceptance among demographers: systemic theory, developed by von Bertalanffy and applied to demography, most notably by Meadows; agent-based models, developed in demography by Billari and Prskawetz; and viability theory, developed by Aubin and applied to demography by Bonneuil. We go on to examine theories of memory and show that artificial intelligence, despite its successes, cannot constitute a theory of human memory. We must turn to neuroscience to see a productive application of mechanistic theories by Bechtel, Craver, and others, who try to describe how the parts of a mechanism are organized to produce human memory—while voicing doubts about the theories’ exhaustivity. We then discuss systemic theory, and in particular Maturana and Varela’s autonomy theory, which focuses on the organism as a whole and on how memory allows it to activate self-maintenance. These two approaches, which developed independently at first, have encountered each other very recently. To conclude this book, we show how, in combination, they offer a more effective approach to more general biological phenomena, while hermeneutics resist to a more scientific approach of human life.

Chapter 8 looks at the problems relating to memory and its transmission, which are crucial in the collection of life histories. We focus on the results of psychological and psychoanalytical studies on human autobiographical memory since their beginnings in the late nineteenth century. Over time, psychology and psychoanalysis developed different schools, which placed varying emphasis on the importance of studying memory. They include structuralism, functionalism, psychoanalysis, behaviorism, cognitive psychology, and evolutionary psychology. We deal separately with psychoanalysis, whose connections to neuroscience—discussed in greater detail in Chapter 9—are significant. The structuralist school, which used introspection, makes no contribution to the study of autobiographical memory, but it does offer interesting findings on the memorization of syllables alone, not of life stories. The functionalist school is barely interested in biographical memory. Its only representative, Galton, provided some elements—some important, others debatable—which we analyze. The behaviorist school, which enjoyed great success in the United States in the first half of the twentieth century, totally rejected the study of mental imagery. In the end, it was the cognitive school that made memory the focus of its study in 1950. We show its many contributions to this field, with the use of data from official registers to verify major life events, and the use of neuroimaging. While evolutionary psychology contributes little to the study of memory, psychoanalysis takes us into new territory. This discipline, introduced by Freud in 1895, relied on Cajal’s discovery of neurons in 1888 as possible storehouses of memory. We describe and critique the way in which Freud incorporated this notion into psychoanalysis with the aid of the unconscious, and compare this with the picture provided by today’s neuroscience. The chapter’s conclusion looks at the challenges to the psychological approach generated by the replication crisis and to the broader notion of “statistical significance.”

In Chapter 7, we turn to real biographies of famous figures, trying to identify which aspects of their lives are highlighted and which are left out, depending on the period in which the events occurred. The aspects on which ancient biographers focused were clearly very different from those recorded today. Another crucial issue is the veracity of the facts gathered. Ancient biographies were often written long after the person’s death, when the memory of that life had faded considerably and was influenced by external events. For biographies written during a person’s lifetime, the biographer’s interests are more important to keep in mind than those of the subject. These interests intermingle with the story told, turning the account into a view reflected by a distorted mirror. In the case of autobiography, we must pay critical attention to the aspects most frequently emphasized by the writer. The more recent interest in the life stories of more ordinary people should also be examined. The biography by Thomas and Znanieki (1919) of Polish peasants who migrated to the United States offers a perfect example, followed by many other similar studies. At the same time as the “comprehensive” approach, another more “explanatory” approach to life stories emerged with the recourse to population sciences. We extensively describe the evolution of the paradigms adopted by these sciences and the relevance of the “explanatory” approach, which we compare with the “comprehensive” approach. We then discuss the methods used by the social sciences to analyze the outline of a human life recorded in one or more interviews and depending on whether the survey is retrospective or prospective. While the approaches to event histories may differ substantially from one science to another, the basic material is the same: the collection, in an interview, of a very small portion of the life of one or more persons. We describe these different methods in detail, including the statistical tools used for the analysis and the scientific foundations of the methods.

Chapter 6 looks at the earliest forms of life stories, dating from the invention of writing in the third millennium B.C.E. We then follow the evolution of these stories over time. We may qualify them as imaginary, although they originate in what are often real lives but magnified into epics, myths, novels, and other narratives that nourish our minds. First, we discuss the notion of genre, which aims to define more precisely the different ways of inventing or transforming lives so as to express deeper reflections on the society and culture in which they take shape. We begin with the Greek philosophers, Plato and Aristotle, who defined epic, tragedy, and comedy by their characteristics. The Middle Ages saw the appearance of a new genre: the novel. In our time, the focus has shifted away from the formal characteristics of genres to their underlying mental processes and the devices they use to bring readers to reflect on and live in their society. Here, we encounter the notion of “comprehension” proposed by hermeneutic philosophy as the opposite of “explanation.” To better understand this approach, we take a closer look at how three examples of the genres were introduced and used. First, the Epic of Gilgamesh, which enables us to understand Mesopotamian civilization; second, Sophocles’ tragedy Oedipus Tyrannus, which give us an understanding of divine causality and human freedom among the Greeks; third, the romances on the life of Henri de Joyeuse, which give us an insight into Christian thought in seventeenth-century France. The chapter concludes with a fuller discussion of the notion of “comprehension.”

Chapter 5 takes a broader view, showing first how these approaches—which regard individual lives as predestined—are not truly scientific but are “idols,” to use Bacon’s term for such investigations. At the same time, we verify that astronomy and genetics qualify as genuine sciences, again in the Baconian sense, and that we can state their main axioms. However, as the “idols” still enjoy widespread acceptance, we must explore why so many people continue to believe in them. We move on to a more general view of religious practices, which facilitate these beliefs in supernatural forces. Cicero’s analysis, produced in a context of polytheistic religion, gives us a better insight into the motives for such beliefs. In more recent periods, monotheistic religions have prevailed, and psychology, bio-cultural anthropology, and cognitive sciences have tried to identify the factors driving these faiths. But the detailed examination of the findings does little to explain the complexity of these phenomena—and even less to account for atheism, whose prevalence is far from negligible. We conclude the chapter with a discussion of the current concept of freedom, which—some argue—offers people a way out of these beliefs, and an ethical discussion on the different approaches presented in this first part.

Chapter 4 focuses on a more recent period, from the late nineteenth century to the first half of the twentieth, when Galton’s eugenics was increasingly recognized and practiced not only by specialized organizations but also by extreme political regimes. The new discipline sought to predict our future life from our genetic endowment at birth. As with astrology versus astronomy, eugenics came to compete with genetics—a product of Mendel’s laws—and we examine how the two disciplines evolved in the early twentieth century. Eugenics became a formidable concept with the advent of fascist and Nazi regimes, which sought to impose the supremacy of certain “races” by eliminating other ethnic groups. World War II ended these regimes, making the promotion of eugenics impossible. However, many former eugenicists such as Osborn continued to pursue the same goals under another label: hereditarianism. We describe the different forms of this theory and show the reasons for their scientific and political failure.

This Chapter 3 takes a closer look at how astrology and astronomy, once indistinguishable, were deemed capable of predicting future events or the future of a human life. The two disciplines emerged in remote antiquity, most notably in Mesopotamia but also in ancient Egypt. By observing celestial phenomena such as the paths of planets and stars, it was believed that one could foretell future events both in human societies and in individual lives. The disciplines spread to the Greek world in later centuries in the form of horoscopes predicting the fate of a given person. However, the scientific revolution that took hold in the sixteenth century introduced a clear distinction between astrology as a divination art and astronomy as a science describing the movements of celestial bodies. Astronomy became a leading science, achieving major discoveries such as those of Kepler and Newton, whereas astrology was increasingly rejected. Yet, in the twentieth century and especially since the 1950s, there have been ever more attempts at a scientific demonstration of the validity or invalidity of astrology, and we examine their findings. We also try to identify the reasons why astrology, despite its rejection on scientific grounds, retains a strong worldwide appeal.

This Chapter 2 presents an overview of the approaches, some of whose more specific aspects are addressed in greater detail in the following chapters. We begin by noting the widespread practice of the divination arts in both past and present, pointing out their prevalence around the world. We distinguish between the “-logies” and the “mancies,” the former making greater use of techniques closer to those of modern science than the latter. We go on to give the reasons for the closer examination of astrology in Chapter 3. Next, we discuss the history of eugenics, from its ancient roots such as texts by Plato and Aristotle, to the Enlightenment and the writings of early nineteenth-century French physicians. This introduces Chapter 3 on eugenics. Lastly, we examine the various meanings of “freedom,” a concept that Chapter 4 addresses only in its current meaning. The notion emerged in Greece in the eighth century B.C.E. and later in Rome. Its definition at the time was very different from the one assigned to it by the monotheistic religions of the early modern age, i.e., free will. The scientific revolution that developed in the sixteenth century further modified the concept. Descartes gave it a philosophical dimension, while classical liberalism and the industrial revolution gave it a political dimension.

This book addresses the challenge of understanding human life. We compare our life experience with the attempts to grasp it by astrologers, eugenicists, psychologists, social scientists, and philosophers. How have these various disciplines sought to give substance to an experience at once so intimate and so universal? The main opposition in the list above lies between understanding and misunderstanding. For example, the astrologers’ and eugenicists’ approach, fully accepted in their day, is now largely viewed as a form of misunderstanding. To show why this is so, we examine their methodology. For practitioners of the other disciplines, their understanding may be limited by various methodological problems they encounter but are trying to overcome. We shall explore these issues as well.

This book constitutes the first review of the techniques of event history analysis in demography. During the 1971s, the amount of reliable data made available as a result of surveys increased enormously. At the same time, statistical and computational techniques developed to allow the new data to be handled. This treatment of the subject gives a detailed treatment of the techniques of event history data analysis. It provides not only a thorough description of models, but also a theoretical presentation of this new research path without losing sight of the underlying application to population dynamics. It will serve as a handbook of methodology for demographers and other social scientists interested in the study of the timing of events in individual lives.

This chapter aims to contribute to the debate on the role of model-based approaches in the future demography. First we call attention to the developments of the discipline since the XVIIth century, and we describe its four successive paradigms related to the period, cohort, event-history and mutilevel perspectives. We argue that these paradigms are complementary and that demography, since its beginnings, has suscribed to the classical scientific research programme launched by the promotors of modern science. Next, we examine how simulation modelling developing in population science recently, may help to respond to three main challenges: how to overcome complexity in social research, how to reduce the uncertainty, and how to reinforce its theoretical fundations. We sketch a model-based research programme for demography, looking specifically at interactions between various population systems. We then show how this approach might conform to the classical scientific research programme, in order to take advantages of its benefits.

This book addresses the challenge of understanding human life. It compares our life experience with the attempts to grasp it. The main opposition among these specialities lies between understanding and misunderstanding. The book also addersses the central methodological difficulty of capturing a human life. It is first examined how certain appraoches may lead to a misunderstanding. The book contrast the example of astrology-an accepted practice in ancient civilizations, but now classified among the pseudosciences-with astronomy, a full fledged science since Galileo's time. Another more recent approach regards human life as predetermined by genes: the methods used by eugenicists, and later by political regimes under the name of hereditarianism, came to compete with genetics. A broader analysis shows how astrology and eugenicism are not truly scientific approaches. Next, the book looks at the ways of capturing an imaginary or real life story. A comprehensive approachwill try to fully understand their complexity, while a more explanatory approach considers only certain specific phenomena. For example demography studies only births, deaths, and migration. Another crucial factor in the collection of life histories is memory and its transmission. Psychology and psychoanalysis have developed different methods to try to explain them. It concludes with a detailed discussion of the concepts and tools that have been propsed in more recent times for understanding the various aspects of life stories: mechanisms, systems, hermeneutics and autonomy.

This book presents a historical panorama of the evolution of demographic thought from its eighteenth century origins up to the present day, and uses it to demonstrate how the multilevel approach can resolve some of the contradictions that have become apparent and achieve a synthesis of the different approaches employed. Part one guides the reader from period analysis, examining longitudinal and event history analysis on the way. Part two is a detailed account of multilevel analysis, its methods, and the relevant mathematical models notably as regards the type of varaibles being used. Numerous examples, examined accross succesive sections, make the book clear and easy to follow. The theoretical and epistemological treatment of these problems, during which the fundations of sociology and demography are revisited, and the logical development that leads to the most recent approaches, are handled sufficiently rigorously to satisfy social science specialists while remaining accessible for readers new to the field. The whole adds up to a comprehensive account of progress in sociological and demographic savoir faire, as well as being both a textbook and an assesment of the multilevel analysis that tackles one of the major problems of empirical sociology: that of integrating analysis at the individual and group level.

This work examines in depth the methodological relationships that probability and statistics have maintained with the social sciences from their emergence. It covers both the history of thought and currrent methods. First it examines in detail the history of the different paradigms and axioms for probability, from their emergence in the seventeenth century up to the most recent developments of three main concepts: objective, subjective and logical probability. It shows the statistical inference they present, different applications to social sciences and the main problems they encounter. On the other side, from social sciences-particularly population sciences-to probability, it shows the different uses they made of probabilistic concepts during their history, from the seventeenth century, according to their paradigms: cross-sectional, longitudinal, biographical, hierarchical contextual and multilevel approaches. While the ties may have seemed loose at times, they have more often been very close: some advances in probability were driven by the search of anwers to questions raised by the social sciences, conversly the latter have made progress thanks to advances in probability. This dual approach sheds new light on the historical development of the social sciences and probability, and the enduring relevance of their links. It permits also to solve a number of methodological problems encountered all along their history.