THE QUALITIES OF TRUTH AND SAFETY

It is possible to draw an analogy between the scientist's search for truth and the engineer's search for safety. Both truth and safety are qualities of what is being created; for the scientist truth is a quality of knowledge, for the engineer safety is a quality of an artefact. At the most fundamental level of description, the differences between the nature of the work of engineers and scientists are not substantial; both are problem solvers. The actual differences are not due to the different nature of the methods each uses, rather they are due to the qualities of the objectives they pursue.

The qualities of an engineered artefact will include, for example, function, safety, economy, reliability and environmental friendliness. The qualities of a scientific theory will include predictive power, explanation, truth, precision, simplicity and abstraction. It would be possible to write extensively on the analogies between these two sets of qualities. For example, it is at least arguable that one of the functions of a theory is to predict. However, we are concerned here with engineering safety and we will restrict the discussion to an analogy between it and truth. The scientist predicts a result based on theory and then sets about testing that prediction in as precise a way as possible in order to progress towards true knowledge. As discussed previously, it is an unfortunate fact that, in strict logical terms, it will never be known whether it has been attained or not. Similarly, the engineer wants to progress towards a safe artefact and therefore pictures the likely scenarios for its behaviour.

Engineering knowledge is used to make predictions and then the consequences of those predictions are interpreted in the light of the uncertainty known to be present. It does not follow that because the artefact has not yet failed it is therefore safe since again in strict logical terms the engineer can never know.

An important difference between science and engineering is that the consequences of error in the predictions made by the scientist and by the engineer are dramatically different. If during an experiment a scientific theory is falsified then the logical result is new knowledge. If an engineer's CPM is falsified, then an artefact has failed. Engineers are therefore interested in safe cautious theories that produce safe artefacts; scientists are interested in detailed accurate theories that produce true knowledge. Both are interested in solving problems.

Engineering scientists tend to be dominated by the scientific interest in accuracy and, as a result, often frown on many of the necessary heuristics and rules in the CPM as being intellectually inferior. Designers rely on rules when science lets them down. Many misunderstandings arise because of a failure to appreciate this distinction. The discussion so far presents us with a strange antithesis : it is the very success of engineering that holds back the growth of engineering knowledge and it is its failures that provide the seeds for its future development.

Not only is it necessary therefore to identify the particular causes of an accident and the dominant causes of groups of accidents, but it is also important to identify the important changes in the current paradigm that have followed. The changes occur at many different levels with widely varying scope, as has been mentioned, and are often difficult to define in any precise sense. Many of the changes are well known and have been discussed individually at length.

One of the important consequences of a heightened awareness of the changes is that new developments in research and practice can be seen as part of the continuing process of the growth of engineering knowledge. This aids an understanding and appreciation of the rules of those who contribute to that process and the uncertainty, which is inevitably part of the process, has to be suitably managed.