Times of Feast, Times of Famine: A History of Climate Since the Year 1000 - Emmanuel Le Roy Ladurie, Barbara Bray
First published in 1967 and only slightly updated since, this is a pioneering work on the study of climate by historians. Le Roy Ladurie's main concern is to make a case for the careful and systematic historical investigation of climate evidence; to lay out the kind of sources that a historian could use; and to demonstrate the promise of the method with a few detailed case studies.

Throughout, the author firmly rejects any temptation to speculatively attribute human events to climatic causes, pointing out that his predecessors who did so had inadequate information to base such speculations on. Instead, Le Roy Ladurie is concerned to discover which years really were unusually cold, unusually wet, and so forth -- a research program that was just in progress, leaving huge amounts of work yet to be done. Dendrochronological data is one source, but it had (at least at that time) mostly been studied in marginal areas like the US southwest. Historians can add data like records of the date of grape harvest (strongly influenced by the weather in the preceding season), date of the yearly first freezing of a lake or port, evidence of the advance or retreat of glaciers, and so on. This kind of systematic year-by-year series is probably more reliable than documentary comments on striking weather events, although the latter can be valuable too.

The longest chapter in the book examines in detail the evidence for the extent of glaciers in the Alps, from the 16th century to the present. The author establishes to his satisfaction that glacial advances corresponded with overall cooling of temperatures by 1 degree C or so, it having been warmer in the late middle ages, but cooling down again starting about 1560 -- this lasted until about 1860, a period known as the "Little Ice Age". Of course, the warming temperatures of the 19th and 20th centuries have been all over the news lately, but what did Le Roy Ladurie think of them in 1967? Not much. They were obvious, but seemed no different than previous warm periods, such as that of the Middle Ages. In fact, noting a slight cooling in the 1960s, he suggested that the trend might be ending. The contribution of industrial carbon dioxide was noted as but one, doubtless not very important, contributing factor.

There is very little discussion of the causes of meteorological and climatic fluctuations in this book (after all, this is not the province of historians). The few accounts that Le Roy Ladurie summarizes indicate that climatology was in a state of infancy at that time.

I hope that this book has now been superseded by more developed work (that's why I rated it only 3 stars); nonetheless, it is valuable to have the goals and methods of research laid out.