What we know and do not know
Jorge Perciado - Asset Management Specialist at Work Management Solutions
Human knowledge can be simplistically categorised into “knowns” and “unknowns” as a powerful tool to address complex problems in our work and everyday life. Donald Rumsfeld famously expressed this idea in one of his most memorable speeches as US secretary of state :
This classical philosophical concept, a keystone to our understanding of bias and research, is today used in psychology, business, engineering and science [2-7]. Figure 1 depicts a schematic representation of this idea with four areas of knowledge: (1) what we know we know, (2) what we know we do not know, (3) what we do not know we know, and (4) what we do not know we do not know.
Known Knowns (what we know we know)
The first zone is determined by our fundamental knowledge and the basis of deterministic facts, and it makes up the smallest part of our individual knowledge. As humans, we are limited by the breadth of knowledge we can store within our brains. While it is difficult for us to conceptualise, the fact is that no single person can know everything. Nevertheless, we overcome this challenge by storing and sharing our collective knowledge — through scientific methods, publications, repositories, and technology. This knowledge is largely gained during our formative years at school and later at university. We should always strive to increase this type of knowledge through books, scientific papers and professional experience, regardless of where or at what point of life we find ourselves. It is important to highlight, however, that we should never take this knowledge for granted. We should always inquisitively doubt, evaluate and criticise what we know (especially new knowledge).
Known Unknowns (what we know we don’t know)
The second zone is possibly the most important and ultimately the most easily recognised. When we know what we do not know something, we are recognising our own limitations. In fact, several studies have demonstrated that the more a person DOES know, the more they recognise that there is wealth of knowledge that they do not. In part, this occurs because knowledge and our understanding of the world are dynamic. As time passes, often many “facts” that we held as true are superseded and become obsolete – eventually fall into this category. While daunting, “what we know we do not know” is the most imperative knowledge we can have. Understanding this, we recognise that there are facts and knowledge that exist beyond the current extents of our experience. Since we still have the knowledge to say we do not know it, we can prepare for it and investigate or uncover this knowledge using different strategies. Asking or hiring subject matter experts with adequate knowledge and skills on the subject is one of them. People and organisations dedicated to a particular subject are able to provide specialist knowledge effectively to turn the “unknown” into a “known”. Known unknowns can be assumptions that we can validate through the scientific method and convert into knowns.
Alternatively, when the unknown knowledge is too difficult to know (due to lack of resources or physically impossible), probabilistic approaches can be used. For instance, in maintenance and reliability, we know that machines will eventually fail due to ageing and degradation. However, we cannot know when or how the failure will occur. This is when probability is a powerful tool to help make decisions in asset management strategies. This is why we use preventive and predictive maintenance strategies like inspections and condition monitoring to know if a component is failing or deviating from expected operating conditions. However, probability must be based on “known” facts and the assumptions taken must be validated.
Unknown unknowns (what we don’t know we don’t know)
The third zone is possibly the most dangerous and difficult to recognise. Just as the more we know illuminates the areas we do not yet know, a lack of understanding of our own limitations often obscures what we don’t know from ourselves. I like to refer to this as the “dark zone”. The unknown unknowns often lead us to wrong conclusions and decisions and when this zone is overlooked, the probability of failure increases. Another of the dark “spawns” of this zone is the prevalence of ignorance that ties into our own sense of self and self-worth that leads to further entrenching of misguided beliefs. Take for example a “flat-earther” who, despite a wealth of evidence, firmly believes that the earth is, in fact, flat. When presented with evidence to the contrary, and ignorant of the guiding principles that support it, the person is unmoved and likely to become more entrenched in the belief as they seek confirmation bias. In the world of asset management, an organisation that operates in this zone would be called a reactive organisation, as they only react retrospectively to inevitable failure instead of preventing or preparing for it.
The best way to deal with the chaos and uncertainty in the dark zone is to be prepared to respond to eventualities with flexibility and agility. However, detailed planning and forecasting in a chaotic environment can be wasteful and detrimental. Instead, it is often more practical to adopt adaptable holistic approaches when preparing for the unknown (e.g. exploration techniques, trial and error). In fact, new technologies (i.e. machine learning) are currently being developed to perform better and more effective exploratory processes, cutting out the danger posed by “confirmation bias” [8, 9].
Unknowns knowns (what we don’t know we can know)
The final zone deals with the knowledge that is in front of our noses, but of which we are unaware of or cannot interpret. A good example is the vast databases that modern workplaces generate every day, sometimes of seemingly mundane information. However, these databases are full of information that, if analysed properly, could lead to significant and useful insight for businesses – including understanding workplace trends, demographics, areas of growth or areas of improvement. However, due to lack of personnel, skills, computational power or aids to analyse these complex and large quantities of data, this knowledge remains ultimately unknown. Other examples of this area of knowledge are not recording information, lack of organisational policies and leadership.
To address the limitations of this zone, increasing the current knowledge about the topic is the most effective approach, but this can be slow and resource-intensive. Another way is to get an external perspective from subject matter experts or consultants, who can help to identify, mine and analyse valuable data that may be otherwise overlooked.
When facing a problem, professionally or in our everyday life, it is important to identify which of these knowledge zones we find ourselves facing and know how to deal with each circumstance. As the known knowns are based on historically or scientifically tested facts, access to this knowledge is relatively fast and simple thanks to modern databases and repositories. Similarly, identifying and solving the known unknowns can be relatively simple using interrogative methods. On the other hand, intuition or instincts will play a major role when facing unknown knowns, or alternatively, bringing new perspectives on how to uncover such knowledge if approached with a keen understanding of the data’s value. Finally, for the complex unknown unknowns, an exploratory approach is usually the best way to tackle problems in this zone. This can be arduous and time-intensive, but eventually, a better understanding of the problem, as well as defined boundaries, can be identified. After all, even though not all possibilities can be thought of or planned, by identifying these knowledge zones, you will be better equipped to detect, predict and manage the risks in any of your professional or personal endeavours.
1. Rumsfeld D. The knowns and unknowns framework. 2002.
2. Fortis Z., Maon F., Frooman J. & Reiner G. Unknown knowns and known unknowns: Framing the role of organizational learning in corporate social responsibility development. International Journal of Management Reviews, (2018). 20: 277-300.
3. Hunt R. W. The challenge of our unknown knowns. Color and Imaging Conference (280-84). Society for Imaging Science and Technology:2010.
4. Žižek S. Philosophy, the “unknown knowns,” and the public use of reason. Topoi, (2006). 25: 137-42.
5. Pawson R., Wong G. & Owen L. Known knowns, known unknowns, unknown unknowns: The predicament of evidence-based policy. American Journal of Evaluation, (2011). 32: 518-46.
6. Gervasi V., Sawyer P. & Nuseibeh B. Unknown knowns: Tacit knowledge in requirements engineering. 2011 IEEE 19th International Requirements Engineering Conference (329-29). IEEE Computer Society:2011.
7. Logan D. C. Known knowns, known unknowns, unknown unknowns and the propagation of scientific enquiry. Journal of experimental botany, (2009). 60: 712-14.
8. Paltrinieri N., Comfort L. & Reniers G. Learning about risk: Machine learning for risk assessment. Safety science, (2019). 118: 475-86.
9. Zhang Y.-J., Zhao P. & Zhou Z.-H. Exploratory machine learning with unknown unknowns. arXiv preprint arXiv:2002.01605, (2020).