Aim and Programme of the Congress It is common knowledge that almost every technological application brings with it not only benefits, but also harmful side effects. Avoiding these side effects, or at least going a long way towards minimizing them, is a great challenge to all those in a position of responsibility within the world of technology. This requires action for safety technology and environmental protection based on a self-contained, comprehensive safety science. Even though safety science has now achieved academic recognition in the USA, it cannot be said that it has made enough of a mark as a discipline in its own right.

The first world congress set itself the task of attempting to take stock of efforts up until the current time. It narrowed this task down in so far as it focused its initial efforts on the safety science problems of industrial society, since this is where changes to the world as a result of technology are most advanced. Obviously, it must be ensured that the ways of thinking and working that safety science entails can be applied to all countries. However, this task can be saved for future world congresses.

The initiating force behind this congress was the chairman of the board of management of TUV Rheinland, Professor Albert Kuhlmann. He secured the scientific backing of Professors Birkhofer (Munich), Compes (Wuppertal), Henschler (Wurzburg) and Salzwedel (Bonn). Due to the fact that this was the 1st World Congress for Safety Science, it was considered that participants would generally not know what expectations to have of the content and structure of the congress. For this reason a call for papers was dispensed with. Instead, a scientific congress programme was drafted from systematic standpoints. It included the plenary session on the opening day, with speeches from politicians, several introductory papers presenting the fundamental concepts of safety science, and problem outlines. These problem outlines served as the basis for the development of the congress's issues for the "energy", "substances", "transport" and "production" sessions. In parallel with these four sessions, further sessions were held providing contributions on the interaction of safety science with technology and natural sciences, law, medicine and other human sciences. The sessions were arranged in such a way as to allow participants interested in, for example, the legal issues to move from session to session in order to hear all the papers on legal aspects for each of the four sessions. In the plenary session on the final day the session leaders presented reports describing the conclusions of their session's work. Two final papers reached beyond the subject theme to focus on social and ethical conclusions.

The congress attracted around 60 speakers of international acclaim. In addition to this, around 30 experts from around the world took advantage of the opportunity for short contributions and poster sessions in a programme of presentations in brief. This ran partly in conjunction with the trade exhibition on safety and environmental protection, which accompanied the Congress. The wide range of nationalities of the speakers underlined the international character of the conference. The 90 papers in total were presented by scientists from Austria, Switzerland, the USAA France, the United Kingdom, The Netherlands, Belgium, Japan, China, Denmark, Egypt, the (former) Soviet Union, Indonesia and Germany.

At the congress opening on 24 September 1990 Professor Kuhlmann welcomed around 1,500 participants in the main congress hall of the Maritim Hotel in Cologne. The participants included experts in the field of safety science, natural scientists and engineers, medical experts, psychologists and other human scientists, legal experts, economists, sociologists and insurance experts representing 40 different countries.

OPENING PLENARY SESSION: OUTLINE OF THE PROBLEMS

After a welcoming address from the Lord Mayor of Cologne, Herr Burger, and speeches on environmental and safety policies by the Federal Environment Minister Topfer and the Vice-President of the European Parliament Alber, Kuhlmann opened the subject proceedings with the question "What must be provided by safety science?". He reminded his audience that technical risks and attempts to conquer them with "safety technology" have accompanied technology ever since people have made use of it. Any particular piece of technology and the safety technology belonging to it were therefore always a unified whole, which could be described by craftsmanship, rules of good technical practice etc. Drawing on a multitude of different techniques was characteristic of "safety technology". In fact, in this respect it was more appropriate to speak of "safety technologies" in the plural. The bond between a piece of safety technology and the piece of technology with which it was associated was deepened by the further development of the safety technology. The impetus for this was provided by incidents of damage occurring during the implementation of the technology in question. This bond was achieved within a limited knowledge schema: instead of taking into account the multiplicity of causal relationships between phenomena, the development of safety technology was geared towards monocausal relationships between damage and cause. Damage was viewed as being determined by its cause instead of being viewed according to the laws of probability.

This approach to mastering the risks of technology was satisfactory in the past. However, with the availability of new forms of technology with a high risk potential, i.e. with a higher potential for causing damage, it was becoming clear that it was no longer sufficient to be "wise after the event", because the sheer scope of possible damage no longer permitted this. This meant that prognostic methods had had to be introduced. The risk potential had, however, also increased because the multitude of different forms of technology used had brought the interrelationships between technology, man and the environment closer and closer together. Generally applicable methods and a systems approach were required to analyse such complex relationships, understand how hazardous situations arise and, right from the planning phase, derive targeted measures for the formation of a system. Safety technology was needed that was anchored in a science with a comprehensive content and consistent methods, something which in fact added up precisely to a "safety science". This science therefore not only had to be prospective, probabilistic and technologically comprehensive but, above all, systems-oriented. The science needed to break away from a merely component-oriented and monocausal approach by treating safety technology, industrial safety and environmental protection issues within one system, consisting of the components man, machine and environment, interacting with each other in all directions.

Even safety science could not be expected to bring about absolute safety in handling technology. What it had to achieve was the "best possible protection". But protection from what? It was here that we encountered philosophical and, in particular, ethical questions which could not be answered by safety science alone. Safety science had to adopt the things to be protected as "axioms". This adoption of a list of things to be protected and the description of the characteristics that embody their hazards were a starting point and were something that safety science will achieve in the future.

An essential concept that must be grasped before any discussion of safety science can begin is that of risk, and Kuhlmann saw a comprehensive theory of risk as a central task of safety science. The starting point of an objective description of risk was a statement of scope and frequency of damage, as well as a statement of the likelihood of the damage occurring. From this information standard figures could be derived, e.g. the product of the scope and frequency of damage, which the insurance industry could successfully use in its work. Kuhlmann answered the sceptical question as to whether risk could be quantified at all in the affirmative. He pointed out that it was otherwise completely senseless to set up quantitative scales for minimizing risks. But these scales were essential in the long term for decision makers and creative scientists. Quantitative risk analysis was also a precondition for extending environmental liability law to include delimiting the facts of cases of damage to the environment and attributing this damage to each of several responsible parties. It was also not so much the individual risk value itself that was of interest, but rather this value's dependency on adjustable parameters. This dependency made the comparison of alternative technical solutions transparent and allowed a rational application of resources.

With regard to the methods of prognostic risk assessment available today Kuhlmann noted with regret that it has not yet been possible to replace the qualitative procedures that have been accomplished with methods that are characterized more strongly by the logic of the issue rather than by the chance nature of the user. Data were still lacking to a large degree as far as the application of quantitative methods of risk assessment was concerned. Setting priorities for building up databases and demonstrating optimal data structures were another urgent requirement for safety science.

According to Kuhlmann, the "machine" component of the "man-machine-environment" system lacked data for evaluating sources of hazards according to their possible effects. Above all, however, the analytical handling of the "man" component presented the most difficulties. The problem of adapting technology to correspond better to human capabilities was far from being solved. At the same time human scientists were in possession of a vast store of knowledge, which had to be opened up and presented in such a way as to make it accessible to the engineer.

A series of speakers replied to the points made by Kuhlmann, initially providing a response that extended across all four sessions of the Congress. These speakers were: Geysen (Louvain), dealing with scientific theory and philosophical aspects; Eysenck (London), dealing with the human role, looking in particular at genetically determined aspects; Nicklisch (Heidelberg), dealing with the legal framework for man-machine-environment systems, and Zebroski (Los Altos), dealing with the theory of decisions.

CONTRIBUTIONS FOR THE RESOLUTION OF THESE PROBLEMS IN FOUR PARALLEL SESSIONS

The development of the basic theme "Tasks and Applications of Safety Science" for the work of the sessions on the following day brought the first plenary session to a close.

In the course of this Hafele explained. at the "conversion and utilisation of energy" session, that for some time now it had no longer been simply a question of supply and efficiency improvements, but also one of the analysis and comparison of risks to the environment, as well as one of the communication and acceptance of risks. In order that safety could be achieved strategically, institutional contexts as well as legal limits and standards needed to be considered within the various legal circles. Above all, however, it concerned people, who had to move more and more towards taking a broad view of a situation and giving it order, rather than adopting a serving role.

On the subject of "risks inherent in handling substances and in substances in the environment" Goldstein referred to the dose-effect-assessment methods of recording the health risks of a substance. These needed to leave room for progressive scientific knowledge. Meanwhile, they had been applied to the even more complex field of ecology. Topics such as agrochemicals, transport and changes in the soil were of the utmost importance. The same was true for risk assessment when setting legal regulations for industrial processes, including the flow of waste materials from these processes, as well as the questions of insurability and liability.

Koshi pointed out that of all the four sessions of the Congress, "Transport" was the area that claimed the highest number of fatalities as a result of accidents. Technology and the natural sciences offered far-reaching possibilities for remedying this, especially through progress in information technology, communications and microelectronics. Quality improvement and control for the "man" component warranted particular care. The state and direction of developments in the field of transport were largely determined by legal standards and regulatory frameworks, which needed to be critically assessed in respect of their suitability.

Roland warned that under certain circumstances highly sophisticated "production" systems and exotic materials may hold greater risks, which did not exist in simpler, older systems. Correct safety management and perfect risk limitation and control posed challenges both to government and industry. The basis for legal regulations should be technical information from the field of production on the one hand and social science information about risks that society is expected to bear on the other.

After a day of intensive work in each of the four sessions, the session leaders presented their reports in the plenary session:

In the "energy" session it had become clear that the risk problems of fossil, nuclear and regenerative energy production were each subject to a completely different degree of differentiated and formal analysis (Voss, Fritzsche, Kroger). For each of the three primary energies it would be necessary to carry out a risk analysis of comparable depth. This needed to be extended beyond energy transformation to include other industries (Teague). The comparison of the German, French, American and Japanese legal situations (Breuer, Roisman, Fujiwara) showed that, faced with a progressive globalization of the energy problem, massive legal differences had to be dealt with. Particular characteristics of the German approval procedure were the participation of the public and the safety technology requirements of the state of technology or the state of science and technology. The American Atomic Energy Act had not led to precise safety standards; the application of a direct and indirect regulation of risk had failed. The discussion of human factors covered the recording of human achievement behaviour and the system interactions between man and machine. Data concerning human errors and the factors influencing them were collected in databases such as NUCLARR (Swain). Something which appeared to be user error was often a problem of the system itself: inadequate information triggered errors in behaviour (Carnino). Rasmussen called for a new type of probabilistic risk analysis that supported risk management. Lee called for a better system of incentives and disincentives in order to make legal instruments for environmental protection more effective.

In the field of "substances", stochastic risks (cancer, genetic defects) were at the forefront particularly as many dubious substances accumulated in the food chain because of their persistence and fat-solubility. Particular care was needed in the application of epidemiology as the safest method of recording such risks (Wichmann). Progress could be expected in molecular biology towards a "dose-based chemical epidemiology" (Henschler) with the recording of biological markers. Mathematical models of toxico-genesis were a fortunate supplement (Schulte-Heimann). Animal experiments were irreplacable, although the application of the results to people was problematical. Cancer was essentially caused by chemical substances, above all in food and tobacco; the risk from ionized radiation was minor (Lindackers). A better-founded risk assessment for small doses was necessary. The description of the formation of cancer as a process of several stages opened up the possibility of determining no-effect-doses with safety actors that had been cautiously applied with success by extrapolating the results of animal research in respect of assessments of non-stochastic risks. Bosselmann accused international chemical laws of neglecting the protection of the environment itself in favour of the protection of human health. The principle of market freedom with the rights reserved to ban chemical products needed to give way to a market ban with rights reserved to permit chemical products. De Haan recommended proceeding with soil protection in a source-oriented rather than an effect-oriented way because of the lack of knowledge of the reaction of the soil to the introduction of harmful substances. Bayer reported on process of setting priorities in the testing of contaminated substances according to accepted criteria. With respect to the international transport of hazardous substances, Smets pointed to the principle under international law of "non-discrimination", which had to be transgressed under certain circumstances in favour of the obligation to use the best available technology for the task. Schilling made it clear that only risks were insurable, and not predictable damages that were bound to happen. Insurance could not take the place of precautions.

The "transport" session confirmed that man was the least reliable component of the system. Measures for improving human behaviour were described: education, selection, licensing, examinations for vehicle drivers, restricting drivers' working hours, compulsory redundancy (Rumar et al.). In the most important area, road traffic, all this had had little effect, because the requirements of professional drivers could not be imposed on all drivers. In this area the technical and organizational environment had to be influenced in such a way that people could not make as many mistakes and did not underestimate the danger, e.g. through technical safety systems, which prevent driving mistakes from leading to serious or fatal accidents (Kobayashi). These ranged from antilock brakes to fully automatic driving operation. With regard to the theory of risk compensation reference was made to experience, which in the EC is represented essentially by official accident statistics. These showed that it is highly likely that accident rates could be reduced by improvements in the technical environment (Rompe). The transport of hazardous materials was seen as being even more dependent on human behaviour than the transport of ordinary materials (Kolstad). The discussion of the environmental problems connected with transport showed that these can be addressed largely in parallel with the safety problems (for air transport: Koplin). Different limits of risk acceptance in various countries were presented.

As far as methods of recording risks in "production" were concerned (Krishnan, Spickett, Blokker) the ever-important themes were the identification of hazardous processes, emissions of hazardous substances and the collection of possible accident scenarios. The risk of fatality was actually low. But increased safety requirements forced prospective risk controls to go over and above the reassuring statistics. Process-technical strategy had the aim of achieving inherent safety, for which Griepentrog presented typical constructive solutions. Pierce called for a periodic certification of the ability of safety experts to apply safety science principles. The legal situation (Seillan, Pierce, Salzwedel) was characterized by progress to an integrated safety legislation, but also by the danger of regulations that web not flexible enough. Standards of international organizations and health and safety criteria that were valid throughout the world would achieve recognition. Countries needed to set permissible risk levels. Limitations determined by age (Blankenburg, Rehtanz) and genetic disposition (Rudiger, Ruppe) were dealt with individually. Early diagnosis of these could prevent occupational diseases. In closing, Trondle warned against threatening officials from licensing and supervisory authorities with stricter criminal responsibility.

In two papers that went beyond the boundaries of safety science Langenbucher first underlined the responsibility of the media in presenting the subject of risk to the public, which he described as "communicative power of definition". Cardinal Hengsbach followed this with a fervent address with reference to his religious mission on the priority of what must be done over what can be done and how this was particularly important within the field of technology.

At the closing session of the congress Kuhlmann cast a backward glance over the subject range of the scientific programme: the expectations of industrial society, introduced by the representatives of the political arena; the experience and the need for action expressed by the scientists. He referred to the fact that the analytical instruments, technical and organizational developments were too fragmented in their structures; that they were available, but still geared too greatly towards their users. For their further development he maintained that there were three approaches:

* Young engineers should be taught the appropriate patterns of thought early on in their education.

* The database drawn on for the application of the methods should be expanded.

* Risk analysis should be developed into an instrument of risk management.

The 1st World Congress on Safety Science was declared a success by its participants.