Reconfiguring organizations to generate richer innovations
About 8 years ago, when I was still working at the Royal Tropical Institute in Amsterdam, I produced – with some expert help – an online dossier on ‘rural innovation systems’ (RIS), which is still available via the Bibliotheca Alexandrina website (here) as are some of its associated documents (see references below). Soft or social systems thinking underlies the main methodology on which RIS is based. The methodology in question was named RAAKS (Rapid Assessment of Agricultural Knowledge Systems). RAAKS applies a knowledge systems perspective to innovation networks. There is hardly a difference between the knowledge systems perspective and the institutional innovation perspective, so it could be applied to many other areas of inter-institutional or inter-organizational activity, including health care, traffic and transport, education (Salomon 1997, 19) and – why not – business (Deloitte, 2013). In 2010, I knew very little about soft systems, so I lacked the reference framework to make good sense of RAAKS. Since then I acquainted myself with the systems field, co-wrote a practical book on the subject, and am currently in the process of writing another, more theoretical one. I am exploring an innovation focus of Churchman’s dialectical systems approach, so this seemed a good time for revisiting RAAKS and examine once more what it has to offer.
AKIS concept The AKIS concept was developed by Professor Niels Röling at Wageningen University in the 1980s. It adopts an integrated view of all organizations and businesses that play a role in the effectiveness of agricultural research and extension. The main reason for doing so is that it is one thing to develop some new agricultural technology – say, fertilizers use -, but quite another to get it adopted by farmers. This example may look simple or silly to you, but a 2015 World Bank report asks “Is increasing inorganic fertilizer use in Sub-Saharan Africa a profitable proposition?” In fact, I have personally been involved in several smallholder fertilizer programs in Zambia in the early 1980s (settlement schemes and the Lima programme) so I know first-hand how hard it is to answer that question in the affirmative. Other concepts that have influenced the development of RAAKS at Wageningen include Rapid/Participatory Rural Appraisal (developed by Robert Chambers. I note in his biography that he spent some time at ‘Pitt’, the Alma Mater of Churchman and Ackoff, so Chambers may have been influenced by one or the other, directly or indirectly) and Farming Systems Research. The term ‘rapid’ means that no use is made of large-scale surveys with tedious questionnaires that produce boring, misleading reports (Chambers 1994, 956). Instead, stakeholders are called upon to participate directly in the production of useful (self-)knowledge.
RAAKS theory … was developed by Paul Engel from the mid-1980s to the early 1990s under the supervision of Niels Röling. It is available online as a PhD thesis (1997, e.g. here). In my experience, it is difficult to explain RAAKS in a concise, comprehensible manner without the use of a concept map, see below. I will explain the concept map starting from the top, working my way downwards. The central question is “how to design more innovative knowledge and information systems?” An example of such systems is AKIS, Agricultural Knowledge and Information Systems. The main players in such AKIS fall in a number of domains or sectors, including those of policy-making, industry, research, commerce, extension, education and agriculture (farmers are the so-called ‘end-users’). The problem is that seemingly good ideas or inventions or knowledge do not get applied: there is an implementation problem and this implementation problem is a systemic one in the sense that the AKIS system is not well enough co-ordinated to learn of its failures. It may not even conceive itself as a learning system, muddling on as it does. Previously it would be said that extension isn’t working, or research isn’t working: in other words the problem was conceived as a technological problem, rather than a ‘social’ one. Now we say that there is something wrong in the management of the AKIS that prevents good ideas from being implemented. Finding out what is wrong requires multi-stakeholder learning, i.e. looking at the problem of implementation through the eyes of those who have relevant roles to play. This in turn requires the application of principles from soft or social systems thinking. The confusing part is where RAAKS takes one more step back by considering how the AKIS should be managed and configured as an innovation system for addressing a whole bunch of ‘systemic’ problems. The solution, or at least a major part of it, is conceiving AKIS as networks of stakeholders, whose missions must more or less correspond with the problems to be dealt with. Multi-stakeholder learning of the network must then design strategies for co-ordination of the network, communication within it, and the development of commitment to apply these strategies for actually producing real results.
RAAKS practice … is really well explained by Salomon and Engel in their manual “Networking for innovation: A participatory actor-oriented methodology”, which is available here. Paul Engel’s PhD thesis is mostly about showing the methodological validity of RAAKS. In practice there are three main phases, which I indicated with pink, green and a thick line. Phase A is about problem definition and system identification, phase B deals with the main constraints and opportunity analysis, while phase C sums it all up in a practical sense by articulating a common policy and/or plan a joint intervention. Each phase has a number of so-called tools and windows. At this point it is important to re-emphasize that ‘intervention’ here means a change in the ways the AKIS is co-ordinated and conducts its communication.
Tools and windows Windows are lenses to look at particular facets of the problematic situation of an innovation network. Each phase has a number of facets that need to be looked at to create a full picture. Each window has one or more tools to examine or inquire into that part of the picture that is covered by the window’s facet. So in phase A the system is defined by the stakeholders that are considered relevant to the purpose of the RAAKS inquiry. A tool can be used to identify the stakeholders. There is another tool to look at their missions and how these could fit in the network mission. The tools and windows are presented in a sequence. The general sequence is logical, as explained above, but it may not be necessary to follow the sequence to the letter. People are natural systems thinkers and cannot be constrained to one facet, because facets are – well – facets of a whole.
RAAKS rationale Engel’s thesis contains a section (p. 86-90) entitled ‘configurations as emergent joint management structures’. In it he uses cases to explains how and why certain patterns of institutional relationships evolve into configurations that are less productive of innovation than if their activities would be co-ordinated more rationally. He concludes “that social actors who are responsible for designing and implementing interventions in agricultural innovation theatres differ in their appreciation of the relevance to innovation of particular social actors, including their functional interdependence.” The study, awareness raising, and realignment of actor perspectives of each other’s potential contribution to innovation is what RAAKS is all about.
Horse husbandry Nothing is more convincing of a methodology than a couple of good cases. Two are described in Engel’s PhD thesis. A particularly good one is that of the Dutch horse husbandry sector (1997, p. 176-182). Another two are described in Salomon’s RAAKS manual (1997, p. 34-40), but they seem a bit less convincing. There is no point in me trying to summarize these cases in this post, but I will make a reiterate a few conclusions from the horse husbandry case. It was found that not all three horse husbandry segments (professional riding and racing, breeding and export, recreational riding) were equally well represented in the national coordination committee. Some of the segments were not well organized. Data collection was sketchy, which hampered policy making and setting the right priorities. With horse racing on the decline and recreational riding having half a million practitioners, there was a clear need for restructuring. A comprehensive list of general information areas was formulated, as were the information priorities of each segment.
RAAKS users It is obvious that the main intended users of RAAKS are facilitators (or their trainers) that guide stakeholders through the entire process. But RAAKS can also be used by individual researchers or field workers to better understand the context (i.e. the knowledge and information networks) for innovations they try to develop or implement. In this sense RAAKS could be considered a complementary method for the dialectical systems approach, in which ‘implementation’ is a key category (see previous post). Other important users are trainers, managers and consultants. As a manager, one “can encourage teamwork, self-monitoring and the generation of ideas on how to improve collective performance related to innovation, with built-in feedback and follow-up.”
Historical notes In the past it was thought that research and extension were sufficient to innovate in agriculture. In fact, before WWII there was hardly any extension at all anywhere in the world. An agricultural university like that of Wageningen was only created after WWI (although there were predecessors of some kind). These days, things are much more complex, not just in agriculture, but in society as a whole. Systems thinking developed and kicked in at Wageningen in the late 1970s, particularly in the form of Checkland’s soft systems methodology, which has a very practical and easily understandable format that appealed to many academics, especially in the UK and the Netherlands. The more generalizing, theoretical work of Churchman, which was well known to Checkland since the early 1970s, was by and large ignored, probably because it was conceived by Churchman himself as a fairly loose set of principles and insights. I am convinced that Churchman’s work should not be ignored, because its generalizing nature helps understand and learn or train a very broad range of systemic or wicked problems as well as the more specific techniques (or methodologies or methods) that have been developed to deal with particular categories or types of systemic problems such as network organization, knowledge management and institutional innovation.
Some final remarks There is a lot that can be said about RAAKS in relation to Churchman’s dialectical systems approach. Here is my pick: (1) RAAKS deals with a real problem: in a complex world multiple actors are involved in the successful design and/or implementation of systemic interventions. A better co-ordination of the knowledge and information system (KIS) is definitely part of that. By concentrating on the KIS part RAAKS leaves a lot of freedom to stakeholders for making autonomous contributions. The level and detail of co-ordination of the innovations required is left entirely to the RAAKS participants, it is not dictated by the methodology. (2) RAAKS is incomplete in the sense that it does not offer systems methods for designing or assessing end-user innovations. This seems to carry the risk that RAAKS will turn into a participation tool for the sake of participation. Churchman’s dialectical systems approach offers a framework for assessment (and even design) that provides room for unfolding the decision-maker and implementation categories across other elements of the framework to examine whether there are any critical oversights. This includes the category of measures of performance, which is needed for decision-makers (with their missions) to know whether system components achieve their objectives. (3). RAAKS lacks a method for learning how it works. The generic social systems thinking method of the Dialectical Systems Approach of Churchman could be of considerable use, although it would have to be learned itself. Once Churchman’s approach is mastered it could be applied to or in any other systems situation or method. (4) The applicability of RAAKS may be limited to situations where systems intermesh. In a developed country such as The Netherlands, we have many, many systems, even to the point that the systems become the problem, they turn into a mess (in the words of Ackoff), whereas in developing countries there is rather a lack of systems, which is why international agencies call most of the shots. And many international agencies make very poor stakeholders since they lack real insight and commitment. Their commitment is to their mission and the avoidance of risk. (5) RAAKS makes me wonder about the inter-relationship between the potential for end-user innovations and the redesign of the innovation system.
Conclusion To paraphrase Churchman (1968, p. 232) when he stated the fourth principle of deception-perception of the systems approach: RAAKS is not a bad idea! In fact, I think it is a brilliant example of systemic agency, because it offers a dearly needed way of increasing the probability of the right things getting innovated without decreasing the autonomy of the stakeholders involved.
- Chambers, R. (1994). The origins and practice of participatory rural appraisal. World Development, 22(7), 953–969. Retrieved from https://www.ircwash.org/sites/default/files/125-94OR-16929.pdf
- Engel, P. G. H. (1997). The social organization of innovation: a focus on stakeholder interaction. Amsterdam, The Netherlands: Royal Tropical Institute. Retrieved from https://www.researchgate.net/publication/40195776_The_Social_Organization_of_Innovation_A_Focus_on_Stakeholder_Interaction
- Salomon, M. L., & Engel, P. G. H. (1997). Networking for innovation: a participatory actor-oriented methodology. Amsterdam, The Netherlands: Royal Tropical Inst. Retrieved from https://www.researchgate.net/publication/40204028_Networking_for_innovation_A_participatory_actor-oriented_methodology [contains overviews of RAAKS’s tools and windows]
- Salomon, M. L., & Engel, P. G. H. (1997). RAAKS’ tools and windows [detailed descriptions]. In RAAKS Resource Box. Amsterdam, The Netherlands: KIT Publishers. Retrieved from https://www.researchgate.net/publication/254762043_Facilitating_innovation_for_development_a_RAAKS_resource_box