Responsible Innovation as a source of inspiration for Technology Assessment, and vice versa: the common challenge of responsibility, representation, issue identification, and orientation

ABSTRACT

Clarifying the historic relationship between Responsible Innovation (RI) and Technology Assessment (TA) can benefit both fields. To achieve this, a basic distinction is made between TA directed toward the promotion side of technology (Constructive Technology Assessment [CTA]) and the control side (Parliamentary TA). In particular, CTA can be seen as a major source of the idea of framing innovation policy in terms of transformative innovation and RI. Originally, the practice of CTA was directed toward democratizing the promotion side of technology. The practice of RI has the broader ambition of encompassing both the promotion and the control side of technology. Parliamentary TA’s familiarity with the control side of technology could inspire RI. At the same time, RI challenges Parliamentary TA to pay more attention to the promotion side of technology and to stimulating politically desired effects of innovation. Accordingly, the paper describes three challenges faced by both TA and RI: representation, issue identification, and directionality. Since Parliamentary TA and RI can inspire each other in many ways, a more intensive interaction between CTA/RI practitioners and Parliamentary/RI practitioners is both desirable and necessary.

KEYWORDS:

• Technology Assessment
• Responsible Innovation
• parliamentary TA
• constructive TA
• orientation

1. Introduction

According to van Lente, Swierstra, and Joly (2017), the rise of Responsible Innovation (RI) ‘shows considerable overlap with the aims, philosophies and practices of Technology Assessment’. Indeed, the field of Technology Assessment (TA) can be seen as ‘a major origin of the Responsible Innovation movement, including already some of the main ideas behind Responsible Innovation’ (Grunwald 2011, 10).

But instead of interpreting RI as a sequel to TA ambitions, van Lente, Swierstra, and Joly (2017) choose to consider RI as a critique of TA. This is a valid position, since RI ‘not only builds on established forms of TA, but also challenges TA approaches’ (Stemerding, forthcoming). And because criticism stimulates intellectual progress, it is also very fruitful to use RI as a conceptual mirror to reflect on the current status, strengths, weaknesses, risks, and opportunities of TA. In particular, RI’s focus on achieving grand social challenges such as sustainability and social equality and cohesion, and thus its attention to the ends of innovation, should give every TA practitioner pause for thought.

It follows that clarifying the relationship between RI and TA can benefit both fields. Such an exercise would make clear what is new about RI, how it can elaborate on the tradition of TA, and how it challenges the current practice of TA in various settings. To make such a comparison productive, it is crucial to have a proper understanding of the current status and history of TA, of RI, and of its relationship to innovation policy and the politics of innovation.

Unfortunately, the paper by van Lente, Swierstra, and Joly (2017) gives an outdated account of the history of TA and an empty description of the history of RI. Their history of TA ends with the introduction of parliamentary TA and Constructive Technology Assessment (CTA) in Europe in the 1980s. Obviously, a lot of developments have taken place since then that should be taken into account when using RI to critically reflect on the current status of TA, and vice versa. van Lente et al. even state that RI ‘does not yet have a history with strands, approaches, institutional forms and experiences’. Their acknowledgement that it is plausible to see RI as a sequel to TA and its ambitions proves otherwise. More importantly, the political history of innovation turns around the notion of responsibility, in other words, the question of who is responsible for the benefits and risks of innovation (cf. Beck 1992).

What does all this mean for the central claim of van Lente et al. that the frame of RI shows that TA should pay more attention to moral ambiguity and the limitations of stakeholder involvement? To reflect on this statement, I will first update the history of TA by placing it in the context of the political debate about innovation and responsibility. Then I will describe three challenges faced by both TA and RI: representation, issue identification, and directionality. Finally, I will address the statements of van Lente et al. and draw some conclusions.

2. Innovation politics and the problem of responsibility

When one describes the history of TA, one first has to acknowledge that TA has been institutionalized in a wide variety of social settings (cf. van Est and Brom 2012). To keep things manageable, I only make a basic distinction between TA directed toward the promotion side of technology (technology developers and promoters; so-called enactors or insiders) and the control side (parliamentarians, regulators, users, citizens; so-called selectors or outsiders) (cf. Rip and Robinson 2013). Parliamentary TA, which aims to stimulate political and public debate on social and ethical issues around technology, is an important way in which TA directed toward the control side is institutionalized (cf. Klüver, Nielsen, and Jørgensen 2015). With regard to the promotion side, CTA (Schot and Rip 1997), which aims to address social issues around technology by influencing design practices, plays a critical role in the history of TA. In the following, I will briefly sketch the development of TA on the promotion and control side against the background of the history of innovation and the core political issue of responsibility. For this, I will use Schot and Steinmueller (2016), who delineate three framings of innovation policy, each of which involves a model of innovation which defines the roles and responsibilities of actors. I will also draw on Daimer, Hufnagl, and Warnke (2012), whose diagnosis of the evolution of innovation policy resonates with the work of Schot and Steinmueller, and according to van Lente et al. also with their analysis.

2.1. Framing 1: R&D for growth

The first framing of innovation policy is guided by the linear model of innovation for growth. After World War II, a consensus developed that scientific discoveries would trickle down to the market and self-evidently lead to societal benefits. Innovation policy was directed toward stimulating basic research, which was seen as a public good. From a political point of view, technology development is positioned as a value-free process that needs to be protected and nurtured. The core of this so-called modernist practice of managing technology lies in separating the promotion from the control of technology (Schot 2003). Regulators are not involved in steering innovation, but only in dealing with its potentially negative societal ‘side effects’. TA emerged in the 1960s in America as a mild critique of this way of distributing responsibilities (van Est and Brom 2012), and was developed to inform actors on the control side of technology. The US Office of Technology Assessment (OTA) was the first organization in the world solely devoted to that task. Its remit was to inform Congress, and its establishment in 1972 aimed to redress the imbalance between Congress and the executive with regard to technological change.

2.2. Framing 2: techno-economic innovation for competitiveness

The second framing is guided by a techno-economic innovation system perspective. In reaction to globalization, innovation policy in the 1980s became aimed at maintaining national competitiveness. It was recognized that knowledge does not simply flow from science to commercialization. Besides stimulating R&D, promoting innovation also required stimulating technology take up and entrepreneurship. From a political point of view, innovation policy is guided by the belief in technological progress, just like in the first framing.

In this model foresight, technology assessment and regulation are add-ons to the core activity of promoting innovation on the assumption that any innovation is to be encouraged since innovation is seen as the motor for producing economic growth and competitiveness. (Schot and Steinmueller 2016, 21)

In other words, the modernist division of responsibilities between the promotion and control of technology is upheld. On the control side, OTA inspired the establishment of parliamentary TA in Europe in the second half of the 1980s. In some countries its focus broadened. For example, in reaction to broad societal protest against nuclear power in the 1970s, parliamentary TA in Denmark and the Netherlands aimed at informing the (political) decision-making arena, but also at dealing with its interface with society. This led to the development of participatory TA, including expert-stakeholder and public participation (cf. Joss and Bellucci 2002).

2.3. Framing 3: sociotechnical innovation for transformative change

Daimer, Hufnagl, and Warnke (2012) and Schot and Steinmueller (2016) signal a new paradigm shift in innovation policy which has been called ‘transformative innovation’ (Steward 2008). In this new framing, the political goals of innovation are broadened and have become more specific and thus more political. In addition to competitiveness, innovation policy is expected to address (grand) societal challenges like health and well-being, sustainable agriculture and energy. In this respect, Daimer, Hufnagl, and Warnke (2012) speak of ‘the normative turn of innovation policy’. Here, the modernist innovation practice of separating the promotion from the control of technology is left behind. In reaction to global issues like climate change, it is acknowledged ‘that the negative impacts or externalities of innovation can be greater than the positive contributions’ (Schot and Steinmueller 2016, 21). At the same time, it is assumed that innovation can be managed in socially desirable directions. This starting point leads to a new distribution of responsibilities. RI is also seen as the responsibility of the enactors of innovation (Rip and Robinson 2013). And in addition to dealing with negative effects, the political system also has the responsibility to make governance arrangements compatible with the politically chosen aims of transformative innovation.

These types of insights find their roots in CTA, which builds on the understanding that science, technology, and society co-evolve. CTA presents an intellectual and political critique of the modernist practice of managing technology, which is based on the political neutrality of science and technology, and the ensuing separation of the promotion of technology and regulating its social effects (Schot 2003). Originally, CTA was not directed toward influencing regulatory practicing, but wanted to address social issues around technology by influencing design practices. CTA proposed ‘to shift the focus away from assessing impacts of new technologies to broadening design, development, and implementation processes’ (Schot and Rip 1997, 251). van Est and Brom (2012, 317) claim that in this way, CTA wants ‘to create a new democratic negotiating space between the actors involved in the design process and actors who are affected by technology’. CTA thus originally aimed to democratize science and technology. During the 1980s, CTA influenced parliamentary TA in the Netherlands but was mainly directed toward the promotion side of technology. For example, CTA was at the origin of strategic niche management, which was driven by the idea that many sustainable technologies remain on the shelves of laboratories as prototypes (Schot and Geels 2008). The CTA perspective also formed an important source of inspiration for setting up Ethical, Legal and Societal Aspects (ELSA) research programs in the field of nanotechnology in America and Europe during the 2000s.

By that time and under the banner of transition studies, the CTA perspective had also come to encompass the control side of technology. In particular, historical research led to a new model of ‘sociotechnical’ transition (Geels 2002), which understands transition as a complex interplay between specific innovations in technologies, business models, culture, human behavior, regulatory arrangements, and so on. These insights lie at the basis of notions like ‘transition management’ (cf. Loorbach 2010) and transformative change and innovation (cf. Steward 2008). Instead of big science and technology push projects like the Manhattan Project, epic sociotechnical shifts like the transition from horse and carriage to gasoline cars provided ways to study and think about innovation and find ways to stimulate sustainable development (cf. Grin et al. 2010). During the 2000s, the study of long-term transformative change led to the new paradigm of innovation policy for transformative change (Schot and Steinmueller 2016). The European Horizon 2020 strategy, of which RI is a core component, is seen as a prominent example of this shift (Daimer, Hufnagl, and Warnke 2012, 219).

2.3.1. CTA as a major origin of RI

Now I have come full circle. I have briefly sketched how the history of framing innovation policy and the politics of the distribution of responsibilities concerning the benefits, risks and directionality of innovation goes hand in hand with the development of TA. Parliamentary TA can be seen as a mild critique of the modernist practice of managing technology in which the promotion and control of technology are strictly separated. Parliamentary TA is focused on the control side and aims to help the political system and society deal with the negative ‘side effects’ of innovation. CTA is a radical critique because it promotes the political view that actors on the promotion and control side should assume responsibility for the negative and desired effects of innovation. I have described how, over time, its ambition has broadened from socially shaping the design process toward ‘directing sociotechnical systems in socially desirable directions and embedding processes of change in society’ (Schot and Steinmueller 2016, 21). CTA thus lies at the origin of transformative innovation as a new way of shaping innovation policy and RI. In extremis, one could even argue that transformative innovation policy is in essence a form of CTA.

3. The challenge of representation, issue identification and directionality

Let us return to the central claim of van Lente et al. that the frame of RI shows that TA should pay more attention to moral ambiguity and the limitations of stakeholder involvement. I explained above that CTA, and the way in which it developed into transition management, frames RI to a large extent. It is important to acknowledge that transition management and RI are both under construction. Maybe this is what van Lente et al. mean when they argue that RI ‘does not yet have a history with strands, approaches, institutional forms and experiences’. CTA has been practiced by academics within ELSA research programs in the field of nanotechnology in Europe (cf. the TA program in the Dutch R&D consortium NanoNextNL, of which van Lente has been program director since 2011). Academic CTA practitioners, however, do not have much experience with the political and societal side of innovation. Parliamentary TA practitioners do. The following describes to what extent Parliamentary TA practitioners take seriously the problem of representation, issue identification, and orientation.

3.1. The participatory turn and the problem of representation

van Lente et al. incorrectly state that the US Office of Technology Assessment did not include stakeholder participation. A few years after its establishment, OTA had learned that it needed to install various procedures to assure the political and scientific credibility of its work, including control of the agenda by Congress and stakeholder involvement in defining key social issues to counter accusations of bias (Van Eijndhoven 1997). OTA inspired the establishment of parliamentary TA in Europe in the second half of the 1980s. These institutions used the lessons learned by OTA and some started to develop participatory TA, including expert-stakeholder and public participation (Joss and Bellucci 2002). It is widely acknowledged within the parliamentary TA community that the most sensitive political framing issue related to participatory TA is the problem of representation (van Est and Brom 2012). For example, in the context of emerging technologies like nanotechnology, it was argued that public participation needed to be moved ‘upstream’ (Wilsdon and Willis 2004). This exposed new sorts of problems of representation. It was found that moving public engagement upstream is by no means straightforward, and that it is not only a matter of involving stakeholders and citizens upstream; experts also need to move upstream (cf. van Est 2011a). To conclude, the role of participation, whether of stakeholders, experts or lay persons, is indeed a core problem in TA. In fact, the issues around (stakeholder) involvement that van Lente et al. raise – limited scale and scope of stakeholder interactions, the issue of representation, power issues – are indeed real. However, my experience is that by playing its role in the political arena and within controversial societal conflicts, the parliamentary TA community has learned by trial and error to be sensitive to them.

3.2. The argumentative turn and the problem of issue identification

van Lente et al. claim that TA is almost exclusively directed to so-called hard impacts like safety risks, which are quantifiable, perceived by many as undesirable, and causally linked to a technology and an identifiable agent (Swierstra and te Molder 2012). Examples of ‘soft’ impacts might be consequences for social order, identity, good life, friendship, and so on. I agree wholeheartedly that TA should pay proper attention to these important types of social issues, and I think the Rathenau Instituut does pay attention to moral controversies and ambiguities (cf. Swierstra et al. 2009). For example, in Intimate Technology: The Battle for Our Body and Behavior (van Est 2014), the Rathenau Instituut argues that the digital age invades our privacy and leads to an economic, social and political struggle for our intimacy, including our attention. This has to do with the fact that since the mid-1990s, ethics has become an integral part of doing TA within the Rathenau Instituut. At the time, (parliamentary) TA experienced an argumentative turn with the aim of deepening the political and normative debate about innovation (van Est and Brom 2012). The argumentative turn in policy analysis inspired the development of new TA methods such as interactive TA (Grin, van de Graaf, and Hoppe 1997) and vision assessment (Grin and Grunwald 2000).

I really hesitate to use the term ‘soft’ impacts. From a political perspective, ‘soft’ might refer to the fact that these issues have not yet been put on the public and political agenda. This brings to our attention the problem of issue identification, which is inherent in both TA and RI. The Rathenau Instituut sees it as its explicit task to expose to public and political scrutiny the societal meaning of innovation. History shows that current ‘hard’ impacts, like environmental risks, have long been ignored as relevant societal risks. For example, it took and actually still takes a lot of scientific and political effort to create consensus and willingness to act on climate change. Meanwhile, the industrialization of our bodies, minds, and social lives is at stake, with all the ethical and political issues that ensue. Indeed, the information revolution has great need for a debate in which human values and visions of the good life play a role. The Rathenau Instituut aims to stimulate such a debate. But it seems to me that it is not TA that stands in the way of such a debate, but a still confident belief in technological progress and our inability to identify and discuss the often unintentional and sometimes paradoxical effects of innovation (van Est 2013).

3.3. The normative turn and the problem of orientation

Related to the above argument, the normative turn of innovation policy reveals the challenge of directing innovation in a socially desired and politically legitimate direction. Steward (2008, 5) makes it clear that: ‘Transformative innovations need not be aimed at social benefit, or indeed be “aimed” at all.’ The current digital revolution seems to be a prime example of that. Moreover, orientation is not a new thing. The dominant political goal of innovation was techno-economic progress, and the assumption or myth was that this would automatically lead to societal progress in general. As the new kid in town, RI is oriented toward new types of political goals, like ecological sustainability. It is important to acknowledge the morally ambiguous and politically contested nature of concepts or objectives like ‘progress’ or ‘ecological sustainability’. In addition, it should be realized that these types of goals relate to the negative ‘side effects’, like social inequality and unsustainability, of innovation that have (gradually) become clear since the beginning of the industrial revolution some two centuries ago. Huge political struggles were needed to put these issues on the political agenda. Timely goal-setting thus can be quite problematic. Reflecting on the current debate, Stiegler (1998) is deeply worried about the total absence of criteria for orientating ourselves in the techno-scientific future that will inevitably be thrust upon us. Stiegler means: we do not know what we want, and in addition, we do not know what we do not want from innovation (van Est 2013). Reflecting on the future of human–computer interaction, Harper et al. (2008, 8) therefore astutely argue that:

Human–computer interaction can no longer be solely the scientific investigation of what role the technology might have – it will need to be part of the empirical, philosophical and moral investigation of why technology has a role. It will entail new questions about how we ought to interact with technology in this new world and it will even entail asking what the use of computing implies about our conceptions of society.

4. Conclusion

TA, in particular CTA, can be seen as a major source of the idea of framing innovation policy in terms of transformative innovation and RI. CTA put forward a radical critique of the modernist practice of managing innovation, which is apparent in the first (R&D for growth) and second framing of innovation policy (techno-economic innovation for competitiveness). CTA formerly acted as a counter-framing, but now forms an important conceptual driver of the third framing of innovation policy: transformative innovation. So, in some respects at least, RI can be seen as a new generation of CTA. Originally, the practice of CTA was directed toward democratizing the promotion side of technology. The practice of RI is also founded on democratic values and principles (cf. Wong 2016), and has the ambition of encompassing both the promotion and the control side of technology. Breaking new ground implies new challenges. van Lente et al. rightly identify the need to pay proper attention to moral ambiguity and the limitations of stakeholder involvement. But they fail to recognize that its involvement with the control side of technology confronts CTA/RI directly with these and other challenges, like finding new ways to distribute responsibilities, finding legitimation, and addressing orientation failure.

Because the control side of technology is familiar ground for Parliamentary TA practitioners, RI practitioners could use Parliamentary TA as an important source of inspiration (cf. van Est 2011b). The primacy of politics and democracy forms the starting point of Parliamentary TA, which is practiced in the political sphere and is often confronted with socially contested developments. This deeply political context forces Parliamentary TA to acknowledge the political and morally ambiguous nature of notions like ‘relevant social and ethical issues’, ‘responsibility’, ‘relevant stakeholders’, ‘grand societal challenges’, ‘RI’, ‘transformative innovation’,¹ and so on. Problems of representation, issue identification, and orientation present wicked power problems that cannot easily be solved. But through trial and error, Parliamentary TA has developed ways to constructively address and debate such issues. RI could benefit from this experience.

But Parliamentary TA should also use transformative innovation and RI as sources of inspiration. So far, Parliamentary TA’s focus has mainly been on the control side and on negative effects of technology. CTA argues that actors on the promotion and the control side should assume responsibility for the negative and the desired effects of innovation. This challenges Parliamentary TA to pay more attention to the promotion side of technology and to stimulating politically desired effects of innovation. First of all, Parliamentary TA could stimulate political and public debate about the proper framing of innovation policy and the need for transformative change and RI. Secondly, Parliamentary TA as the study of the societal and ethical aspects of innovation could be combined with assessing the functioning of the innovation system.² Thirdly, Parliamentary TA should assume responsibility not only for mitigating the negative effects but also for achieving the desired effects of innovation. Parliamentary TA could assess and stimulate debate on how governance arrangements can be made compatible with the politically desired goals of innovation. This would lead to a form of constructive Parliamentary TA that assesses and discusses the role of governance in innovation. Finally, in the wake of climate change, the notion of transformative change has come to play a central role in the discussion about innovation policy. The challenge is to attain sustainable sociotechnical systems within the next three decades. This transition perspective could also be used to reflect on the rather unguided transformative power of the current digital revolution. Parliamentary TA should play a role in clarifying common moral principles that would give a dignified political direction to these changes.

To conclude, since Parliamentary TA and RI can inspire each other in many ways, a more intensive interaction between CTA/RI practitioners and Parliamentary/RI practitioners is both desirable and necessary.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Rinie van Est works at the Rathenau Instituut, where he is primarily concerned with emerging technologies. He also lectures at the Department of Industrial Engineering and Innovation Sciences of the Eindhoven University of Technology. Some recent publications: Human Rights in the Robot Age (2017), Just Ordinary Robots: Automation from Love to War (2016), Intimate Technology: The Battle for Our Body and Behaviour (2014).

Notes

1. This implies, for example, that the three framings of innovation policy described above should be regarded as political framings. Framing innovation policy as transformative innovation thus has political repercussions, and needs broad political support if it is to become the most significant approach. I have shown how various types of framing both enable and constrain the practice of TA. The first framing enabled the establishment of OTA. The second enabled Parliamentary TA in Europe and the development of participatory and argumentative TA, plus the institutionalization of CTA in the form of ELSA research linked to large scale R&D programs. The third framing enables RI. The promotion and the practical elaboration of RI, therefore, in itself is a political endeavor. In fact, current innovation policies, on the European level too, are still dominated by the first and second types of framing of innovation policy, and in my opinion support for transformative innovation and RI is still rather fragile (cf. Rip 2016).

2. At the Rathenau Instituut, we aim to develop such an integrated form of innovation assessment.