Systematic and Conditional Anomalies
we discover anomalous behaviours, we need to determine whether the
behaviour occurs regularly or only occasionally. For example, if we
expect the car to start when we turn the key, and it doesn’t, but
then after a couple tries it does, then this is a conditional
anomaly. We find no established
pattern; sometimes the car starts when we turn the key, and sometimes
not. If we tried turning the key again and the car still didn’t
start, then again, and again, and so forth, then the pattern suggests
a systematic anomaly.
We call a “one-off” anomaly “conditional” because we see the behaviour only under some conditions, some of the time. In the case of starting the car, it might be that the car doesn’t start on the first try in cold environments. This suggests that the problem isn’t within the car itself, rather outside the car, in this case, in the physical environment. Again, context and contrasts have done much of the work here, and how we portray the context in which anomalous events occur depends on our characterisation of our expectations.
Continuing the car starting example, perhaps the car doesn’t start when the headlamp switch is turned on. Turn the headlamp switch off and the car starts. In this case, we call the anomaly “systematic”, as the behaviour suggests a problem with the car itself.
Note that with sufficient refinement, any conditional anomaly might be seen as systematic; it could be that we simply haven’t grasped the systematic nature of the problem just yet.
In either case, the characterisation of the anomaly guides our explanations of the unexpected behaviour. This depends heavily on our experiences in the world, which taken together are our understanding of the world. We lean heavily on this general understanding when we characterise our reasoning. Enhancing our ability to express our understanding, then, comes with practice, practice, and more practice.