Historically... particularly before the scientific era... the concept of 'cause' was multifaceted and metaphysical. It was used variously in the sense of 'materials', 'design', 'purpose' or 'agent', and might even be restricted to refer only to consistent event sequences. It was a concept that enjoyed much speculation and manipulation in the minds of armchair philosophers.

As the power and utility of the scientific method became apparent, the concept of cause became more pragmatically linked to the idea that there was an active identifiable agency that was responsible for all change events. Eventually, when it was realized that the multifarious manifestations of the cosmos could be modelled by various transformations and interactions of a single 'substance' called 'energy', it became a practical scientific reality to be able to suppose that all changes could be assumed to be the result of 'energy' transfer events. For the purposes of science therefore, we assume that a cause is an event during which a transfer of energy event alters one physical configuration to another.

In simple cases, such a transfer is often direct and seemingly obvious. Thus a hammer-blow event is the cause of the nail penetrating the wood because the energy of the hammer was transferred directly via the nail to the wood. Everyday life is full of incidents wherein the cause of one situation is most probably an obviously observable energy transfer. The victim is dead because of a bullet hole thru the heart. The fire is alight because the match was applied to the kindling. It may even be acceptable to suggest the rotation of the earth 'caused' the sun to 'rise'. After all, the sunrise event is a direct consequence of the mass rotation relocation event of the earth, even though energy does not actually 'flow' from one event to the other.

For a cascade of supposedly causal events it could be a matter of opinion as to what caused what. The cause of all the fireworks might be supposed to be the gunpowder, or the fuse igniting the explosives, or the message which came down the optical fibre, or the resentment of the mouse which got its nose whacked, or the attraction of the diode bait, or the energy stored in the coiled spring of the trap, or.... In many situations it becomes a matter of how far back in time it is sensible or pragmatically possible to go. If the fireworks display portrayed in the image is a harmless celebration finale, then it would be academic as to which event in the sequence suggested is taken to be the cause... but if someone was to be seriously injured or killed, then the cause of that event would probably be sought further back in time, in terms of what was the cause of such dangerous materials getting into the hands of irresponsible individuals. It all depends upon who thinks what is important.

For complex systems, establishing credible causal linkages is scientifically demanding. It is very frequently the case that a set of one or more events or circumstances are influential on the outcome of one or more other events. Each cause can be viewed as a trigger event within a cascade of energy-change transfer events, or as an umbrella event stimulating many disparate event-phenomena.

A consistent event sequence does not necessarily indicate a causal relationship, even though a so-called "effect" always follows the so-called "cause". Many phenomena are cyclic and to suppose that one phase "causes" the next phase is quite unhelpful. Thus night does not cause day, nor low-tide cause high-tide, nor recession cause growth, even though the former always precedes the latter... or vice-versa depending upon your starting point. One O'clock certainly doesn't cause 2 O'clock unless you are touring Wonderland. The setting of the sun however, could possibly be thought of as causing night, because the arrival of local light energy has changed to a much smaller value, and the consequent visible environment is what we call night.

Correlation is not a defining characteristic of cause either. The statistical population increase/decrease of a biological species A that correlates to the statistical changes of a species B does not indicate that either species "causes" the changes in the other. Both could be responding independently to something else... like climate change for example. The statistical changes in carbon dioxide concentrations that are correlated to the statistical changes in average global temperatures does not indicate that either causes the other. Both are more probably responding to changing solar and cosmic energy inputs.

Survival depends upon acquiring an extensive data base of highly probable causal relationships and yet in the chaotic environment of our existence, the smallest of energy transfer events can occasionally cascade catastrophically to precipitate a totally unexpected outcome. Nature will always surprise us thus and forever keep analytic predictability an impossible aspiration.