 So thank you very much for the opportunity. It's a great honour and a privilege being asked to address you during the opening session of your conference, of what we'll know that be a very exciting and fruitful conference. And a lot of important issues have already been raised by your opening speakers today. I hope that through my presentation, I will be able to contribute to your future success, particularly at a time of great transformational changes that have taken place within the CGR. COVID-19 and the one CGR change agenda have already created a considerable amount of anxiety and fear of the unknown. And this is absolutely understandable. So it is very important that we react to this and that we do this in a productive and considered way. Change and uncertainty also provide opportunities for future success. I decided my talk today to facilitate such success. Given that I only have 15 minutes, I want to jump right into it and skip a lot of the preliminary issues. I want to talk to you today about how we can create a culture of high performance under those environments. I want to share with you some strategies that we might be able to use in order to make a boundary organisation such as the CGR even more successful. So what is it that actually drives success within the CGR? I could think of hundreds of initiatives over the last 49 years that would demonstrate that. But what all of those initiatives have in common is that they all created what is called social capital. Social capital is really the cornerstone of real innovation and creating social capital requires high performance. And I can think of four key conditions that need to be satisfied for such high performance to occur. Firstly, the organisational culture needs to support the basic needs of his people. Secondly, the scientific knowledge that the organisation generates needs to match the problem domain. And Alexander actually touched on that in his opening remarks. Thirdly, the work needs to compel action. So something needs to follow directly as a consequence of that work. And fourthly, the scientific knowledge needs to be evidently of very high quality. Now let me skip through those quickly. The first one, the organisational culture needs to support the basic needs of his people. And I want to remind you of Maslow's hierarchy of needs. Most of you would have heard about that. And what I'm really talking about here is the top end of that pyramid. Obviously physiological needs, safety, emotional needs need to be met for people to be functional. But in order to perform at their best and do the best science possible, they also need to need esteem and they need self-actualisation. Under esteem, we have respect, we have status, we have recognition, strength, freedom. And under self-actualisation, we have the desire to become the best that one can be. And that requires effort on the individual as well as on the organisation. And as we are going through that transformational change within the CG, I would appeal to everybody who is there and helps to shape that future that they keep that in mind because we need to work together, we need the partnerships that Alexander has already referred to. The second point is matching the scientific knowledge with the nature of the problem. LMB and CERVITS have defined a three-level category, a simple model that allows you to look at the different types of science that we can do. And traditionally, the CG has been very good at level one science. Level one technologies that have a very clear and deterministic cause and effect that are strongly disciplinary based, that provide solutions to very clearly defined goals and where the decision-making is about the control of the system, where it solves well-defined problems with obvious outcomes. However, there are many issues that we as scientists need to work on that are not quite as clearly defined, not quite as simple. So as the systems become more complex, we move to a level two where cause and effect becomes a bit more blurred. A lot of the properties are emergent. The research goals are a lot more interdisciplinary, and we are dealing with systems to create options for addressing often contended goals. The decision-making becomes more adaptive, and we need to find ways to manage reasonably well-understood situations. What gets really complicated is when we move to level three. And at level three problems are characterized by often unforeseeable shifts with no clearly definable boundaries. The research goals have become very transdisciplinary, and it's often much more about designing governance systems that are capable of consensus on ill-defined and contended goals in support of a common vision. The decision-making is much more about consensus-seeking and developing robust institutional responses that anticipate future conditions despite persistent deep uncertainty. Let me give you a few concrete examples. Under level one, we could imagine a problem like improving border use efficiency on farms through the introduction of a high-yielding, more efficient variety. Very much sort of the bread-and-butter stuff that the CG's been doing for a long time been doing very well. Under level two, we could imagine that we want to optimize the use of limited water resources across an enterprise of farming operations. And here it becomes already clear that we are actually dealing with something that is a little bit more morphe where the actual structure of the farm becomes critically important, but also where the skills of the operators become important in terms of which optimization pathway you might want to take. It gets really problematic when we move up to level three. And here we could think of introducing irrigation on a large scale as a transformational technology that will alter not only the farming operations but the entire fabric of a rural community or of a region. Examples that we can think of at that level three are, for instance, the issues that we're dealing with with climate change. Now, the problem that we often create, and that is why we need to match our scientific knowledge with the nature of the problem, we're dealing with level three type issues, but we are trying to solve them using level one technologies. And once we have these sorts of level confusions, then the problems become often even more enhanced. And everybody gets dissatisfied because we're actually talking cross purposes. So when can science actually compel action? And that's the third point. We can think of that as two axes. On the x-axis, we have value consensus. On the y-axis, we have uncertainty. Now, ideally, we want to have high value consensus and low uncertainty. In those cases, science can be used to select the preferred options. There's clear agreement about what the options are, why action should be taken, and what the best cause of action is. And this is typically the space in which we operate when we have level one type problems and level one type technologies. There are quite a few examples for these sorts of things. On the other end of the extreme, we have low value consensus and high uncertainty. And under those situations, science is often used to justify divergent approaches. And more science can lead to an increased number of contented options and paralyze the decision-making process. It doesn't have to be like that, but it is often the case, particularly when they're not careful enough and think about the nature of the problem and the type of science that is actually needed to solve it. I don't have enough time to talk about the other two quadrants in that diagram, but I encourage you to have a look at the presentation afterwards. So let me get to the fourth point, ensuring high quality research. And a lot of you are probably familiar with the word that the ISPC and subsequently the ISDC has done with many of the scientists throughout the CGAR on developing that framework. I believe that some of you might have even participated in the workshops that we ran towards three years ago when we first started talking about it. This framework is now available for use right across the CGAR. And in fact, it's used for many other purposes as well because it lends itself to be adapted to a whole range of purposes. And there are really four lenses that characterize that quality of research for development. The first one is relevance. The second one, scientific credibility. The third one, legitimacy. And the fourth one, effectiveness. And I provided you with a web link where you can download some of those documents. But I quickly want to go through these four lenses. So relevance really refers to the importance, the significance and the usefulness of the research objectives and the processes and the findings to the problem context and to society. And that incorporates a strategic stakeholder engagement along the entire agricultural research for development continuum. It also demands that original and societally relevant research is well aligned to international, national and regional priorities. It does recognize the importance of public goods. And again, that's an issue that's already come up earlier today. Scientific credibility. Very important component. In the past, that was often the only component that people considered when they looked at research quality. Scientific credibility requires that the research findings are robust and that the sources of the knowledge be dependable and sound. It includes a very clear demonstration that the data used are accurate, that the methods used are fit for purpose and that findings are clearly presented and logically interpreted. Legitimacy recognizes the importance of good scientific practice such as peer review, for instance. It also means that we have to look very carefully at how we conduct the research process and ensure that it is absolutely transparent and that the people need to be involved in the process are appropriately involved in it. This touches then on the next point which is legitimacy. And legitimacy means that the research process must be fair and ethical and it must be perceived as such. And that encompasses the ethical and the fair representation of everybody involved as well as the consideration of the interests and perspectives of the intended users. It requires transparency, sound management of potential conflicts of interest, recognition of the responsibilities that go with public funding and a very genuine involvement of partners in the co-design and the recognition of the partner's contribution. And finally, we need the effectiveness. And effectiveness means that the research generates knowledge, products and or services that actually lead to innovations and provide the solutions. And it implies that the research is designed, implemented and positioned for use within a dynamic theory of change, that it has competent leadership including clear elements of capacity building as the research progresses and a supportive and enabling environment to translate knowledge into action and to help to generate the desired outcomes. I touched on quite a few principles that I believe are very important for the future success of the whole of the CG but particularly of a program like FDA that is by its very nature addressing a lot of the level three problems that I alluded to early on today. I hope that you can use some of those principles to your advantage and to create a very important scientific knowledge that we need in order to solve many of those problems that were already raised in the opening session today. Thank you very much.