ABSTRACT
When we come to civilized times, Man's increasing ability to modify the landscape — at first only by clearing limited areas of forest by fire, but ultimately turning over most of the temperate deciduous forest zone to cultivated grasses (food grains), draining or irrigating huge areas of swamp and near-desert, creating vast lakes by building great dams and regulating to some extent the levels of nearly all great lakes and rivers — means that, one by one, the environmental indicators which mainly registered climatic fluctuations in earlier times cease to do so. For that reason — as well as for more obvious ones — an attempt must be made to use the original reports of the weather prevailing, and related aspects of the landscape and effects on the human economy, which have come down to us from the observers who lived in the times concerned and of which numerous collections now exist (see Chapters 12 and 13 and the corresponding lists of references). Methods of analysis must take account of, and preferably be designed to eliminate the effect of, differences in the frequency of reporting or survival of the reports until our own times. Allowance must be made for the shortness of human memory in records such as ‘the highest flood that ever was known’, which unless measurements are given can be accepted only as meaning that there was a very high flood. Standards of comparison were simply not available to the reporter, when no long records of past observations were to hand for consultation. It is not surprising therefore that archaeologists, botanists, meteorologists and others, who have used this material, have sometimes been accused of arguing in a circle and suspected of calling in entirely hypothetical climatic changes — as a sort of deus ex machina — to explain the facts of history and shifts of vegetation limits which might more properly be ascribed to the works of Man, clearing and burning forests, grazing with goats, building and later neglecting irrigation works, and so on. In recent years, however, new techniques have provided new types of evidence — particularly measurements of the stable isotopes of oxygen, carbon, etc., in ice sheets, tree rings, lake sediments and other materials, and multivariate analysis of the responses to climate registered in such items — which, whatever the difficulties and uncertainties of interpreting such proxy data, are independent of the works of Man and of human bias in reporting. The use of such ‘proxy’ data offers the hope of verifying, or correcting, the analysis of the climatic sequence derived from original reports, as well as extending it to most parts of the Earth.
