The world is a very confusing unpredictable place. Humans have, from the very beginning tried to create some sort of order and means of reducing its uncertainty.
They began by trying to find ways of ensuring the continuation of the regularities of the daily appearance of the sun, and of the annual cycle of the seasons. The responsible Gods had to be propitiated by sacrifice and ritual. The beginnings of this process are described in “The Golden Bough”1 .
The rituals that developed are still in existence in the form of daily prayers and annual holidays for Easter and Christmas, absorbed by more recent religions.
By the time of the Middle Ages it was believed that the world was basically static and unchanging, apart from any deviation which could be blamed on the Gods and solved by prayer or sacrifice.
The beginnings of modern science came with Kepler, Galileo and Newton who found that the behaviour of solid bodies could be predicted successfully by a fairly simple mathematical model. In order to do this Newton had to invent friction and gravity. It became possible to predict the movements of the planets as well as solid objects on the earth. The principle of mathematical models took hold to the extent that it was believed that the entire universe resembled a huge clock operated by mathematical formulae.
The complacency was disturbed by Einstein’s theory of relativity and by Planck’s quantum theory, but the idea has survived even the discovery of evolution, and the structure of atoms, ..
Science depends on measurement and all measurement inevitably involves inaccuracy. It was only in early 20th century that inaccuracy itself became subject to mathematical models. It is unfortunate that so many people who make use of them do not take care that their measurements comply with the assumptions of the model used.
The scientific study of the climate began, as with other disciplines, by the measurement of its properties. The science of meteorology is today amongst the most successful of all scientific institutions in its ability to measure and forecast local climate anywhere on the globe.
By comparison with other scientific disciplines it is faced with serious handicaps. Science demands that an experimental determination cannot be accepted unless it can be repeated, to an agreed level of accuracy, by an independent observer. Climate observations cannot be verified in this way. In addition, instruments, procedures, supervision, location are not standardised and the qualifications or the identity of the observer is often unknown. Recently the observer may just be automatic.