Risk

A Sociological Theory

Niklas Luhmann

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781351492904
• Edition: 1
• Topic: Psicología
• Subtopic: Psicología social

A great deal of attention has been devoted to risk research. Sociologists in general have limited themselves to varying recognitions of a society at risk and have traced out the paths to disaster. The detailed research has yet to be undertaken. In Risk, now available in paperback, Niklas Luhmann develops a theoretical program for such research. His premise is that the concept of risk projects essential aspects of our description of the future onto the present. Risk is conceived as the possibility of triggering unexpected, unlikely, and detrimental consequences by means of a decision attributable to a decision maker.

Chapter 1
The Concept of Risk

I.

Risk is addressed nowadays by a wide variety of special research areas and even by different scientific disciplines. The traditional statistical treatment of risk calculation has been joined by economic research. Instrumental in this development has been the brilliant approach taken by Frank Knight.¹ His original aim was to explain entrepreneurial profit in terms of the function of uncertainty absorption. This was no new idea: Fichte had already introduced it in relation to the ownership of land and class differentiation. In the modern context of economics, however, it has permitted the astute linking up of macro and micro-economic theory. Knight’s distinction between risk and uncertainty has, however, meanwhile petrified into a sort of dogma — so that conceptual innovation earns the reproach of not having applied the concept correctly. But other disciplines do not face the problem of explaining company profits, nor are they concerned with the differences and connections between theories of the market and the business enterprise. Why should they then draw the concept from this source?

Statistical theories have been joined by applications in the fields of decision and games theory interested in their own controversies — such as the degree of meaningful subjectivization of expectations and preferences. As a sort of countermove, psychologists and social psychologists have established that in reality people do not calculate in the way they should if they put store by earning the attribution ‘rational’ from the statistician. They commit ‘errors’, some would say. Others would claim that they act in a manner adapted to the requirements of everyday life. In any case it is striking that such deviance displays both structure and direction. The gap is growing ever wider and deeper. As in continental drift, the disciplines are moving farther and farther apart. We now know that housewives in the supermarket and street children in Brazil can calculate highly successfully — but not the way they learned to do so, or did not learn to do so, at school.² We know that values can be quantified — with the result that what was really meant can no longer be recognized.³ And not only private persons cannot do so or do not make the effort. In positions where rationality is among the duties attributed to the role, where particular care and responsibility in dealing with risks are expected, even in the management of organization — risks are not calculated quantitatively; or at least not in the way conventional decision theory proposes.⁴ But if this is the case, what use are theories of risk that determine their conceptual approach in terms of quantitative calculation? Is the aim, as in certain moral theories, only to set up an ideal to permit everyone to establish that he cannot live up to it — luckily no more than others can? Handling quantity and its practical relevance are at stake — at any rate for specialized areas of research and the academic disciplines.

Still within these models of quantitative risk calculation, which are generally guided by the subjective expectation of advantage, we now realize that an important correction must be made. We shall refer to it as the disaster threshold. One accepts the results of such a calculation, if at all, only when it does not touch the threshold beyond which a (however unlikely) misfortune would be experienced as a disaster. For this reason subsistence farmers are highly averse to risk because they are under the constant threat of hunger, of losing their seed, of being unable to continue production.⁵ Under money economy circumstances we find corresponding results: entrepreneurs facing liquidity problems are less willing to take risks than those who are not plagued by this problem when the risk is of a given magnitude.⁶ It will probably be necessary to take into account that the disaster threshold will have to be located at very different positions, depending on whether one is involved in risk as a decision maker or as someone affected by risky decisions.⁷ This makes it difficult to hope for consensus on such calculation even when dealing with specific situations.

But that is not all. In the meantime the social sciences have discovered the problem of risk as well; but not so to speak in their own front yard, but because it has not been nurtured and watered with enough care in neighbouring plots. Cultural anthropologists, social anthropologists, and political scientists point out — and rightly — that the evaluation of risk and the willingness to accept risk are not only psychological problems, but above all social problems. In this regard one behaves as the pertinent reference group expects one to, or — either in conformity with or in breach of prevailing opinion — in terms of one’s socialization.⁸ The background to this position, although initially only postulated as a countertheory, is a better understanding of the extent of the problem, inspired above all by the technological and ecological problems confronting modern society. This brings to the foreground the question of who or what decides whether (and within which material and temporal contexts) a risk is to be taken into account or not. The already familiar discussions on risk calculation, risk perception, risk assessment and risk acceptance are now joined by the issue of selecting the risks to be considered or ignored. And once again, discipline-specific research can reveal that this is not a matter of chance but that demonstrable social factors control the selection process.

However, these efforts still presuppose an individualistic point of departure. They modify the results of psychological research. If, for example, such research demonstrates that individuals in everyday contexts typically underestimate risks — perhaps because everything has gone well to date and because one overestimates one’s capacity for controlling events and underestimates the extent of loss or damage that can be suffered in situations one has yet to experience, — then we can pose the question of how communication that seeks to raise the level of risk awareness must be constituted.⁹ There is no doubt that by including social contexts and operations, a necessary complementation of psychological insights is provided as well as a convincing explanation of why individuals react differently in differing social situations. As we learn more and more in this respect, however, we finally reach a point where we have to ask ourselves whether attribution to individual decision making (whether rational, intuitive, habitual etc.) can still be regarded as tenable at all. Or whether, leaving this aside, we should not attempt a strictly sociological approach, tackling the phenomenon of risk only in the sense of communication — naturally including communication of decisions made by individuals.

Without taking such a radical stance, sociology has finally also turned its attention to the problem of risk; or it has at least laid claim to the term risk. Following the ebbing of anticapitalist prejudice, it now finds a new opportunity to fill its old role with new content, namely to warn society.¹⁰ At present this function is, however, being performed completely without reflection; and by this we mean that sociology is not reflecting on its own role. For even if the sociologist knows that risks are selected: why and how does he do this himself! Sufficient theoretical reflection would have to recognize at least the ‘autological’ component that always intervenes when observers observe observers. The social determination of all experience and action recognized by sociology also applies mutatis mutandis with regard to the discipline itself. It cannot observe society from without, it operates from within society; and of all observers, it should be the first to realize the fact. It may all very well adopt the topics of the moment, may support protest movements, may describe the dangerous nature of modern technology or warn against irreparable environmental damage. But others do the same. What ought to go beyond this is a theory of the selectivity of all societal operations, including the observation of these operations; indeed, even including the structures determining these operations. For sociology, the topic of risk ought thus to be subsumed under a theory of modern society, and should be shaped by the conceptual apparatus thereof. But there is no such theory, and the classical traditions that continue to guide the majority of theoreticians in the field of sociology provide few openings for topics such as ecology, technology, and risk, not to speak of the problems of self-reference.

We cannot at this point discuss the general difficulties of interdisciplinary research. There is cooperation at project level, and there are areas of research that could be referred to as ‘transdisciplinary’ fields, for example, cybernetics and systems theory. Risk research could represent a further possibility. For the moment, however, the negative consequences of participation by numerous disciplines and special research areas are most apparent. There is no definition of risk that could meet the requirements of science. It appears that each area of research concerned is satisfied with the guidance provided by its own particular theoretical context. We must therefore question whether, in individual research areas, and even more so more in interdisciplinary cooperation, science knows what it is talking about. If only for episte-mological reasons we may not assume that such a thing as risk exists, and that it is only a matter of discovering and investigating it. The conceptual approach constitutes what is being dealt with.¹¹ The out-side world itself knows no risks, for it knows neither distinctions, nor expectations, nor evaluations, nor probabilities — unless self-produced by observer systems in the environment of other systems.

When we seek definitions of the concept of risk, we immediately find ourselves befogged, with an impression of being unable to see beyond our own front bumper. Even contributions addressing the topic directly fail adequately to apprehend the problem.¹² The concept of risk is frequently defined as a ‘measure’¹³; but if it is only a problem of measurement, it is not quite clear what all the fuss is about. Problems of measurement are problems of convention, and in any case the risks of measurement (thus of measurement errors) are not the same as what is being measured as a risk. Such examples could be multiplied ad infinitum, paradoxically in the exact sciences in particular; for they seem to assume that exactitude has to be expressed in the form of a calculus and that the use of everyday language accordingly requires no precision.

It is, however, generally agreed that not too much attention needs to be paid to questions of definition, for definitions serve only to delimit, not adequately to describe (let alone explain) the object under investigation. After all, if it is not at all clear what one is supposed to be dealing with, it is quite impossible to start investigating. And, rightly or wrongly, the sociologist will be permitted to assume that this imprecision offers the opportunity to switch topics in accordance with fashion and opinion, with changing sponsors, and shifts in public attention. We thus have good reason to concern ourselves initially with delimiting the object of risk research.

II.

Older civilizations had developed quite different techniques for dealing with analogous problems, and thus had no need for a word covering what we now understand by the term risk. Mankind had naturally always been preoccupied by uncertainty about the future. For the most part, however, one trusted in divinatory practices, which — although unable to provide reliable security — nevertheless ensured that a personal decision did not arouse the ire of the gods or of other awesome powers, but was safeguarded by contact with the mysterious forces of fate.¹⁴ In many respects the semantic complex of sin (conduct contra-vening religious instruction) also represents a functional equivalent, inasmuch as it can serve to explain how misfortune comes about.¹⁵ In ancient oriental maritime trade there was already what could be described objectively as risk awareness accompanied by the corresponding legal institutions,¹⁶ which to begin with were scarcely to be distin guished from divinatory programmes, appeals to tutelary gods, etc., but which from a legal point of view — particularly as far as the distribution of roles between the suppliers of capital and the seafarers was concerned — clearly performed insurance functions, and which with relative continuity right up to the Middle Ages was thus to influence the law of maritime trade and maritime insurance. Even in non-Christian antiquity there was, however, still no fully developed decision awareness. Thus the term ‘risk’ first appears in the transitional period between the late Middle Ages and the early modern era.

The etymology of the word is unknown. Some suspect it to be Arabic in origin. In Europe the word is to be found already in medieval documents, but it spread only with the advent of the printing press, in the initial phase apparently in Italy and in Spain.¹⁷ There are no comprehensive studies on the etymology and conceptual history of the term,¹⁸ and this is understandable, since the word at first occurs relatively rarely and is used in a great variety of contexts. It finds significant application in the fields of navigation and trade. Maritime insurance is an early instance of planned risk control,¹⁹ but elsewhere we also find formulations such as ‘ad risicum et fortunam’ or ‘pro securi-tate et risico,’ or ‘ad omnem risicum, periculum et fortunam Dei’ in contracts regulating who is to bear a loss in the event of its occurrence.²⁰ The term risk does not, however, remain limited to this field, but spreads from about 1500 on, probably with the expansion of printing. Scipio Ammirato writes, for example, that whoever propagates rumour runs a risk (rischio) of being asked where he obtained his information.²¹ Giovanni Botero writes: ‘Chi non risica non guadagna,’ and following an old tradition, distinguishes this maxim from vain, foolhardy projects.²² Annibale Romei reproaches whoever ‘non voler arrischiar la sua vita per la sua religione’ ,²³ In a letter addressed to Claudio Tolomei by Luca Contile on 15th September 1545,²⁴ we find the formulation: ‘vivere in risico di mettersi in mano di gente forestiere e forse barbare.’ Since the existing language has words for danger, venture, chance, luck, courage, fear, adventure (aventuyre) etc. at its disposal,²⁵ we may assume that a new term comes into use to indicate a problem situation that cannot be expressed precisely enough with the vocabulary available. On the other hand, the word goes beyond the original context (for instance in the quotation ‘non voler arrischiar la sua vita per la sua religione’), so that it is not easy to reconstruct the reasons for the new concept coming into existence on the basis of these random occurrences of the term.

...