You are here: Biosecurity > Legal Aspects > Consensus Conferences > Technology Assessment

Table of contents: A review of the Danish, Dutch and UK approaches to this special form of technology assessment
Last Document: 2. Conference
Next Document: 4. Organization

3. The Consensus conference as a form of technology assessment

3.1 The Danish consensus conferences

The consensus conference as described above was developed by the Danish Board of Technology (DBT), which was established by the Danish Parliament in 1985 as an independent institution in order to undertake '[technology] assessments and to initiate public debate on technology' (T. Hansen, 1992). The establishment of the DBT was a result of the development of the Danish technology assessment programme. In the 1980's, this programme evolved from a narrowly research-oriented assessment system to a broader dialogue-oriented assessment system involving negotiation about new or developing technologies among interest groups from diverse backgrounds.

The history of this development may be summarized as follows. In 1980, a Danish Parliamentary committee that had been appointed two years earlier by the Council of Technology within the Ministry of Industry published a report about "partial technology assessment". The method of technology assessment recommended in this report was retrospective, in the sense that new technologies were to be assessed through the analysis of existing technologies; but of even more significance in the present context was the fact that the system of partial technology assessment provided for the involvement of only a single independent party such as a trade union or public interest representative.

In 1984, the committee produced a second and radically different report. In place of partial technology assessment, this report recommended "comprehensive technology assessment"; and in place of retrospective assessment, it urged the need for prospective and pro-active assessment. Comprehensive technology assessment adds to the technical considerations of expert evaluation the wider ethical, social and political aspects of the technology under consideration; and in addition, it incorporates all interest groups with significant points of view on the technology. Proactive technol ogy assessment takes account of the novel features of developing technologies, and it attempts to estimate the costs and benefits that will accrue in the future from the introduction of new technologies.

The Danish Parliament discussed both of these reports and the result was a significant shift in its view of technology assessment. Instead of looking upon technology as essentially self-regulating, the Parliament came to regard it as subject to a potentially wide range of cultural and social influences. As a result, the Parliament encouraged the use of a far wider range of methods of technology assessment. Alongside the more traditional consultations of experts and closely related interest groups, a wider variety of interest groups and members of the general public were invited to participate in the process of technology assessment.

It is in this context of changing views of technology and technology assessment that the initiative of the DBT in establishing consensus conferences is properly to be understood. Essentially, the Danish consensus conference is 'a method of involving ordinary citizens with their different backgrounds in the assessment of technology' (L. Hansen, 1992). The aim is simultaneously to stimulate informed public debate about new technologies and to provide politicians with direct information about lay people's perceptions of and attitudes towards these technologies.

The first Danish-style consensus conference was held in 1987 on the subject of "Genetic Technology in Industry and Agriculture". Since then the DBT has organized consensus conferences on a regular basis. Topics considered include: food irradiation (1989); mapping the human genome (1989); air pollution (1990); technological animals (1992); traffic and the environment (1993); childlessness (1993); and electronic identity cards (1994). Suitable topics for consensus conferences are ones which involve a partly or wholly unresolved scientific andlor technological issue of current social and political interest (L. Hansen 1992). It ought to be mentioned here that the DBT is still in the process of developing the consensus conference model. For example, one consensus conference was held over a period of four days, allowing the lay panel an extra day to write the final document; another consensus conference was held during the week rather than during the weekend; and most recently, a consensus conference was held over two days with a team of journalists helping the lay panel to draft their report.

The selection of the subjects for consensus conferences is done as part of the annual planning process of the DBT. Each year, approximately 1000 proposal forms are sent out to individuals and institutions, inviting suggestions for topics for future TA projects. Proposed topics are then assessed by the DBT and the 20 best subjects are short-listed through an internal voting system. Subsequently, the DBT works out a description for each of the 20 subjects, including the choice of methodology. The catalogue of short-listed topics and descriptions forms the basis of the annual work programme of the DBT. Usually, one or two out of the 20 subjects are consensus conferences.

A major feature of the Danish system is- the close relationship between the Parliament and the consensus conferences. This is made possible through the Parliament's relationship with the DBT. The DBT tries to maximize the political impact of its consensus conferences both by holding them in or nearby the Parliament building and by scheduling them shortly before decisions on particular topics are due to be made by the Parliament. In practise, however, scheduling of this sort is difficult to achieve. Often, the time-table for parliamentary debates is produced shortly before the parliamentary session; and even after the time-table has been made available, it is always possible that external events (for example, decisions or actions taken by the European Council of Ministers) may necessitate last-minute changes in it.

Thus far, the impact of the Danish consensus conference final reports on the parliamentary decision-making process appears to have been remarkable. Members of the Danish Parliament regularly consider the findings of the lay panels, and on several occasions the Parliament has debated issues raised by consensus conference final reports. For example, in the case of the consensus conference on mapping the human genome, the Danish Parliament discussed the lay panel's negative view of the use of the information gained from mapping the human genome for purposes of personal employment and insurance, and it requested that action be taken to prevent such uses (Jacobs, 1993).

Apart from the direct use of the information gained from the consensus conferences in parliamentary debates, the lay panel's final document is regularly mentioned in the Danish media. Over the years, successful dissemination has made the consensus conference a part of Danish social and political life; and this, in turn, has served to give the consensus conference and its organizers credibility. The media strategy has therefore always been a very important part of the organization of consensus conferences.

The considerable experience of the DBT has enabled it to reduce the time required to organize a consensus conference to around 6 months (Jacobs, 1993). Each consensus conference costs between DKr300-500,000 (= c. £ 30-50,000), excluding the staff costs of the DBT itself. To date, only the consensus conference on genetic technology in industry and agriculture (1987) has been evaluated (Agersnap, 1989b). However, an evaluation on the group process and opinion formation of the lay panel which worked on the mapping of the human genome was published in 1990 (Rienecker, 1990).

3.2 The Dutch public debate

In 1993 the Dutch followed the Danish initiative and held their first consensus conference on the subject of genetic modification of animals in the science museum in The Hague. This "public debate" - as it was termed in Dutch - was organized by the private Dutch Institute for Consumer Research (SWOKA) under the auspices of the Dutch Organization for Technology Assessment Studies (NOTA) and the Foundation for Public Information on Science, Technology and the Humanities (PWT). The costs for the conference were approximately FI330,000 (= c. £ 114,000), and these were met by NOTA, PWT, the Ministry of Agriculture, Nature Conservation and Fisheries (LNV), and the Ministry of Science and Education.

Although the organizers of the Dutch public debate basically followed the Danish model of consensus conferences, there are some conceptual as well as organizational differences between the two initiatives; and these would appear to have contributed to the rather different outcome of the Dutch as compared with the earlier Danish initiative. First, the Dutch organizers deliberately called their initiative a "public debate" and not a "consensus conference". For the Dutch organizers, the stimulation of public debate rather than the development of public consensus was the primary goal of their initiative. They argued that however advanced and well established biotechnology might be within the scientific community, it was still a new and unfamiliar technology from the point of view of most ordinary citizens. For this reason, they saw the dialogue between the lay panel and experts as a useful way of fostering public debate; but they took the view that only at a later stage, when public understanding and public attitudes towards biotechnology were better developed, would it be realistic to seek a social and political consensus on this subject (Hamstra, 1993).

It should be noted here that it was never the intention of the Danish consensus conference organizers to seek overall social and political consensus about the topics chosen for consensus conferences. Rather, their aim was to use lay panels as a way of involving the public directly in the process of clarifying the ethical, social and political aspects of socially sensitive scientific subjects. Additionally, they sought to 'provoke interest and convey information so that the subject could be more widely debated' (Kodahl, 1992). The Danes call their dialogues "consensus conferences" not because they aim to achieve an overall social consensus but rather they aim to achieve a consensus amongst the members of the lay panel (see section 2.2, above); and even here, they acknowledge that on occasions the lay panel may write a final consensus report which states 'on which points agreement can be reached, and on which there are conflicts' (Ravn, 1989).

It may have been as a consequence of these fundamental conceptual differences between the Danish and the Dutch organizers that the Dutch lay panel found itself unable to agree on the wording of a final report. Instead, it presented a majority assessment (9 out of a total of 15 lay members) and a minority assessment. The majority point of view was that there should be a moratorium on the genetic modification of animals except in cases where such modification was undertaken for purposes of socially accepted scientific research. The minority point of view was in favour of the retention of the present Dutch regulatory system, under which each new proposal for the genetic modification of animals is assessed on the basis of costbenefit analysis. Of course, the fact that the Dutch lay panel was not fully representative of the wider population (see section 3.4, below) prevents us from drawing any firm conclusions concerning the balance of Dutch public opinion on this issue. However, the lay panel's majority view dominated the discussion of the two reports at the final conference of the public debate.

A second difference between the Danish and the Dutch initiatives is that the Dutch public debate took place shortly after the Dutch Parliament had debated and passed a law on the issue of transgenic animals. This fact may help to explain the relatively low interest of the mass media in the conference. Another factor that may have played a part in the rather disappointing press coverage was the lack of a sufficiently strong media strategy. The choice of a regional venue for the public debate probably led to the low number of ordinary citizens among the audience at the consensus conference. Furthermore, one member of Parliament criticized the lack of active involvement of parliamentarians in the Dutch public. debate. In fact, MPs had been invited to be part of the audience; but only a few took up the invitation. Presumably, this helps to explain the fact that most members of the lay panel were sceptical about the impact that the public debate would have on policy- and decision-makers (Hamstra, 1993).

Finally, there are organizational differences between the Danish and the Dutch initiatives. As we have seen, the Danish consensus conferences are organized by the DBT, which is the officially recognized office for technology assessment. In contrast, the Dutch shared the organization of their public debate between three independent institutions: NOTA, PWT and SWOKA. NOTA was founded in 1987 as an independent organization funded by the Ministry of Science and Education. NOTA's main task is 'to contribute to social debate and political judgement concerning issues resulting from or connected with scientific and technological developments' (NOTA 1994). Its main audience is Parliament itself. NOTA engages in both traditional technology assessment and so-called "constructive technology assessment", the latter involving the incorporation of institutional, social and political aspects of proposed technologies into the early stages of their development (Dierkes, 1989).

PWT was founded in 1986 as an independent institution to coordinate and support the public understanding of science and technology in the Netherlands. It is funded by the Dutch Ministry of Science and Education and the Ministry of Economic Affairs. PWT develops projects for the promotion of the public understanding of science and coordinates the different initiatives of the various organizations involved in this subject area. Its activities include science-by-phone, help desks for journalists and regular newsletters. Its interest in the organization of the public debate was obvious since one of its major campaigns, which was launched in 1992, involved informing the public about biotechnology in agriculture, with special reference to the genetic modification of animals.

Finally, SWOKA is a private institute for consumer research and has carried out many studies dealing with consumer acceptance of novel biotechnological products. It is involved in technology assessment programmes, in which it especially focuses on the consumer's point of view.

The fact that NOTA has expertise in TA whereas PWT specializes in public information about science and technology was the main reason why NOTA and PWT took joint responsibility for the organization of the first Dutch consensus conference. NOTA and PWT defined the goals of the public debate and dealt with the press. NOTA was also responsible for contacts with the Parliament and wrote a briefing paper based on the lay panel's final document, and PWT was responsible for writing down the questions and answers put at the conference as well as the final report and a summary of the final discussion. SWOKA was commissioned to administer the debate and also acted as a source of information (Hamstra, 1993). On the whole the organization and management was well regarded by most of the participating parties. Still it is questionable whether a division of labour of this kind is as efficient as the Danish method, in which a single institution is responsible for project organization and management.

There are, of course, different ways of measuring the success of a consensus conference. Not only should one evaluate the most visible effects of a consensus conference, namely the response(s) of the media and of the policy- and decisionmakers to the outcome of the conference, but also one should measure the success of a consensus conference in its own terms, i.e. through monitoring of internal factors such as organization and management. So far, three evaluation reports of the Dutch public debate have been published (all in Dutch): one on the process and effects of the organization (Hamstra, 1993); one on the method applied (Mayer, 1993); and one on the ethics of the consensus conference (van Voorthuisen, 1993). The overall comment on the effectiveness of the Dutch public debate was fairly positive, even though a few aspects of the organization and management were criticized and some adjustments for future initiatives were recommended (de Rooij, 1994; Hamstra, 1993; de Grooth, 1993).

3.3 The UK national consensus conference

Great Britain will be the third European country to organize a Danish-style consensus conference. Last November, the organization of the first UK national consensus conference of this kind on the subject of plant biotechnology was announced at the press launch of the 1992/3 Annual Report of the Agricultural and Food Research Council (AFRC)¹. The AFRC has provided financial support (c. £ 80,000) for the running of the consensus conference, but has delegated the organization and administration of the conference to the Science Museum, London.

The AFRC is funded by the Office of Science and Technology (OST) and by the Ministry of Agriculture, Fisheries and Food (MAFF), and its aims are to 'advance scientific knowledge for agriculture, food and other biology- based industries and for sustainable use of land and biological resources' (AFRC Annual Report 1992/3). The AFRC is aware of public anxieties about some applications of new biotechnology (including genetic manipulation), and argues that, 'we have the responsibility to communicate the new science in a way that is understandable to the public.... Ultimately it must be the public that makes decisions about biotechnology on the basis of its social, legal, economic and other repercussions for the future' (Blundell, 1993).

The Science Museum London, on the other hand, has a strong and professional commitment towards the public understanding of science. In 1989 the Museum and Imperial College of Science, Technology and Medicine appointed the first Professor in the Public Understanding of Science. The Museum's Science Communication Division is conducting several national and international research projects, and its journal "Public Understanding of Science" publishes contributions in this new interdisciplinary field from all over the world. The Science Museum contributes actively to the growing UK public understanding of science movement of the past decade (see Introduction, above); and it regards its involvement in the first UK national consensus conference as particularly appropriate, given both its commitment to the public understanding of science and its status as an independent and nonaligned agency.

¹ As form April 1994 the AFRC is incorporated in the newly established Biotechnology and Biological Sciences Research Council (BBSRC).

The aims of this initiative are: first, to contribute to public policy-making by providing useful information about public perceptions of agricultural and food biotechnology in Britain; second, to contribute to informed public debate about agricultural and food biotechnology in the UK; and third, to evaluate the potential usefulness of the consensus conference as a contribution to public policy-making and public debate in socially sensitive areas of science.

The topic of plant biotechnology was chosen for this conference on the grounds that genetic engineering of plants increasingly plays an important role in modern agriculture and food production and hence is an object of both public interest and public concern. The spectrum of research and development in this area is wide and includes the development of pest- and herbicide resistant plants, the development of saline- and drought tolerant plants, and the use of genetically modified plants for the production of food, drugs and chemicals.

Genetical modification of animals was excluded from the topic on the grounds: first, that this is a distinct area; second, that this distinct area raises a number of sharply different and (in the UK, at least) extremely contentious questions concerning animal welfare and "animal rights"; and third, that to introduce the added complications and controversies associated with animal biotechnology would be to "overload" the lay panel with too many and too diverse issues to give them any realistic prospect of arriving at a meaningful and useful consensus on their designated topic.

The UK national consensus conference on plant biotechnology will take place in the autumn of 1994. It will be evaluated and compared to the Danish and Dutch consensus conferences. Although planning is still at a relatively early stage, initial media responses to the press conference which announced the launch of the consensus conference has up to now been encouraging (see e.g.New Scientist of 4.12.1993, p.9, and of 12.2.1994, pp.3 and 13; The Economist of 18.12.1993, p.99).

3.4 National and cultural contexts

In Denmark the consensus conferences of the DBT are seen as useful ways of actively involving lay people in technology assessment. Up to now they have been a success and they have become an integral part of the Danish technology assessment programme. This remarkable development has been noticed by several other offices of technology assessment; and the Netherlands and most recently the UK have taken up this initiative within their own science communication programmes. Other countries are currently considering whether to run consensus conferences.

At first sight, it might appear that the Danish success could easily be repeated in other countries by the simple expedient of copying exactly their model of the consensus conference. This would be the case if the success of the Danish consensus conference were independent of national and cultural context. This, however, is almost certainly not the case. For there are many external factors that appear to contribute to both the conduct and the outcome of consensus conferences. These include, for example, the size and structure of the population, the nature of the political system and the cultural and social position of science and technology. Unless the influence of these factors is taken into account, attempts to repeat the success of the Danish initiative may meet with disappointment.

The most striking features of the Danish population are its small size (c. 5 million) and its great homogeneity (one language, one religion, one cultural background). Another important cultural characteristic of Denmark is its long tradition of intensive public discussion and public participation in the policy- and decision-making process. In Denmark, striving for consensus is widely regarded as a virtue; and this is an obviously important enabling context so far as the consensus conference initiative is concerned.

In countries where different languages are spoken, where people with different social backgrounds live together, or where the political culture is combative, the cultural and political climate may not be so conducive to the success of a consensus conference. At the very least, it will be important for those who are considering the possibility of launching a consensus conference initiative to take into account the distinctive cultural and political context in which they are proposing to work. This context may necessitate modifications to the Danish model in order to suit altered circumstances. At the same time, however, it will be important to ensure that any modifications do not undermine the principal rationale of the consensus conference.

3.5 Some additional remarks

Hitherto, consensus conferences have at least partly been funded andlor organized by offices of technology assessment (DBT in Denmark and NOTA in the Netherlands), and therefore it seems correct to add them to the long list of technology assessment (TA) initiatives. However, a consensus conference is an unusual form of TA, for it emphasizes the central role of a small group of non-experts in raising, discussing and finally assessing issues about a scientific or technological subject of political and social relevance. Such a conception has little in common with the original concept of TA. It lacks the pretensions to completeness, to thoroughness and to technical precision of that original conception; for by its nature it is not 'a method of analysis that systematically appraises the nature, significance, status and merit of a technological program' (Daddario, 1968).

As a matter of fact, most practical TA initiatives also fail to live up to these ambitious criteria -experience has shown that they cannot be matched within reasonable time and cost (Dierkes, 1989). At the same time, in recent decades much TA has come to incorporate at least some form of public participation (Cambrosio, 1991). Nevertheless, by any standards, a consensus conference is an extraordinary form of TA. The reason for this is the exclusiveness of the consensus conference's reliance upon a lay panel to undertake the assessment exercise. In conventional TA, professional assessors are responsible for setting up the assessment programme, inviting well recognized experts to contribute evidence, arranging (perhaps) for nonexpert inputs, and publishing a final report; but in a consensus conference, all of these vital tasks are assigned to lay people.

In a consensus conference, the members of the lay panel decide which issues will be raised and considered at the conference on the basis of what seems to be relevant to them. It is probable, but by no means certain or inevitable, that in making this decision the lay panel will follow expert judgments concerning the relative importance of particular questions. For example, in the Dutch public debate on genetic engineering of animals, the lay panel wanted to be informed about possible impacts of this kind of research on medical treatment in humans (Hamstra, 1993). This, incidentally, can cause difficulty for the organizers of a consensus conference when it comes to deciding how far the lay panel should be allowed to change the definition or the scope of the topic under consideration.

The lay panel not only chooses questions for the conference but also assesses the resulting issues on the basis of its subjective perception of the subject area. Of course, the organizers provide the lay panel with competent and impartial information and a list of recognized experts; but it is the lay panel that decides which experts it will hear, and what it will do with their advice. The element of subjectivity is ever present in a consensus conference. For the lay panel spends a relatively short albeit relatively intensive - period of time informing itself about the subject area; and it cannot possibly be expected to obtain a comprehensive grasp of the subject-area before coming to its conclusions and producing its final report.

At this point it ought to be made clear that the lay panel of the consensus conference is not fully representative of the population. For one thing, between ten and twenty people cannot be said to represent an entire national population in a statistically meaningful sense; and for another, the process of selection of the lay panel by means of public advertisement and written application necessarily biases the sample of lay people from which the panel is selected. The consensus conferences in Denmark and the Netherlands suggest that the selected members of lay panels tend to have a greater than average interest in the social dimensions of science and technology. This is probably inevitable; indeed, it was part of the original conception of the consensus conference that it would give a small and effective group of citizens, who care about social applications of science and technology, the opportunity to state their considered views on socially sensitive scientific subjects. It has yet to be shown whether or not self-selected and randomly selected lay panels would arrive at similar conclusions about the same issues.

For these reasons, at least, the final consensus conference report should not be understood as the product of a systematic and complete assessment of the subject under consideration. Its conclusions are not as final as those to be found in conventional TA documents such as environmental impact assessments. This point leads us to make the extremely important observation that consensus conference reports do not provide definitive verdicts on the subjects with which they deal. Rather, they provide one (we believe valid and useful, but inevitably partial and incomplete) contribution among others to the processes of public debate and public policyinformation. This may seem an obvious point; but given the nature of the lay panel, it is tempting to compare it with a jury. The analogy between lay panel and jury is flawed. In a consensus conference, there is no accused and no accuser. Above all, no legal or political authority is vested in the consensus conference analogous to that which is invested in a court of law.