A highly debated question over the past few decades has been whether or not the study of history can be counted as a science or not. A lot of the books I have read recently have discussed this idea, and I have decided to collect a few of the notes I have made and ideas I have thought of whilst reading together in one (rather jumbled) blog post.
Trevelyan has said history is a mixture of the scientific (research), the imaginative or speculative (interpretation) and the literary (presentation). Historians' facts will always be incomplete; no one is ever going to unravel scientifically the metal processes of twenty million Frenchmen during the French Revolution of 1789. But interpretations of this event cannot be arrived at by a mere process of induction.
Elton thought history was neither an art or a science. 'History', he declared proudly, 'is a study different from any other and governed by rules peculiar to itself'. Elton argued that historical knowledge is cumulative; he said that through the steady accumulation of empirical knowledge, professional historians were advancing 'ever nearer to the fortress of truth', making history somewhat scientific. He also distinguished between a historian and a poet or novelist by arguing that historians have to be trained; i.e. they have to learn technical details of the documents they use. However Zeldin argued that the history one writes is an expression of their individuality, and that one cannot be trained or taught how to write history.
Carr said there is no clear separation between the researcher and the object of research. However he did say that historians should not judge the past in moral terms; their purpose was rather to understand how the past had contributed to human progress. It is pointless, for example, to condemn slavery in the ancient world as immoral; the point is to understand how it came about, how it functioned, and why it declined, opening the way to another form of social organisation. There is also, surely, a moral element in research in the natural sciences; moral concerns may drive scientific research, or they may emerge from it.
Carr also argued that whilst no two historical events are identical, no two atoms are either, no two stars, no two living organisms, yet this does not stop scientists from framing their laws. Similarly, he said 'the very use of language commits the historian, like the scientist, to generalisation'. He pointed out that history is not merely confined to the establishment of particular, isolated facts: it teaches lessons, such as the constant reference made by delegates at the Versailles Conference in 1919 to the Vienna Congress of 1815. Carr also said history can still be a science even though it cannot predict the future. The law of gravity, he observed, could not predict that a particular apple would fall as a particular time in a particular place. Scientific laws operated only under certain specific conditions, and history seems to be able to predict events too in some ways, such as the necessary conditions and time and place where a revolution might take place - thus history was just like any other science in its generation of laws and predictive capacities.
One reason to suppose that history is not a science is that historians' conclusions cannot be experimentally replicated. But a number of important aspects of sciences are based, like history, on observation rather than experimentation, such as, for example, astronomy. Such observation may not be directly visual; it may be indirect, for instance, as when the existence of Pluto was postulated through the observation of irregularities in the orbital path of the next-outermost planet in the solar system. Likewise we infer the existence of some objects in outer space from radio waves, but we cannot recreate them in experiments, like we can in a chemical reaction, Yet nobody has claimed that astronomy is not a science.
Froude claimed that history had to confine itself to a presentation of facts, and not be used to 'spell out' theories of a 'scientific nature'. Indeed, in every case where historians have tried to prove generally applicable laws of history, they have been easily refuted by critics who have demonstrated instances where they do not apply. For example in A Study of History Toynbee tried to draw a series of general laws as to why civilisations rose, developed and collapsed. Yet this has later been criticised as some historians have suggested Toynbee simply selected the evidence he wanted. Buckle suggested that history could be scientifically reduced to a series of mathematical formulae.
While many people, especially politicians, try to learn lessons from history, history itself shows that in retrospect very few of these lessons have been the right ones. Time and time again, history has proved a very bad predictor of future events. This is because history never repeats itself; nothing in human society, the main concern of the historian, ever happens twice under exactly the same conditions or in exactly the same way. And when people try to use history they often do so not in order to accommodate themselves to the inevitable, but in order to avoid it. For example, British politicians in the post-war era based much of their conduct of foreign policy on the belief that the 'appeasement' of dictators, such as was carried out by Prime Minister Neville Chamberlain in his relations to Hitler in 1938-39, could only lead to disaster. In practice however, this very belief only too often led to disaster itself, most notoriously in 1956, when the British Prime Minister Sir Anthony Eden, obsessed with the desire to avoid the 'appeasement' of dictators which he himself had criticised in the later 1930s, launched an ill-advised and unsuccessful military strike against the Egyptian government under Colonel Nasser when it nationalised the Anglo-French owned Suez Canal.
Nor does history enable one to predict revolutions, however they are defined. Historians notoriously failed to predict, for instance, the fall of the Berlin Wall and the collapse of the Soviet Union in 1989-91. And although Carr argued that the historian's role was to use an understanding of the past in order to gain control of the present, very few historians have shared this concept of using the past as the basis for concrete predictions. Whilst a chemist knows, in advance, the result of mixing two chemicals together in a lab, a historian has no such advance knowledge.
History, however, can produce generalisations. It can identify, or posit with a high degree of plausibility, patterns, trends and structures in the human past. In these respects it can legitimately be regarded as scientific. But history cannot create laws with predictive power. An understanding of the past might help in the present as it can broaden our knowledge of human nature, provide us with inspiration or warning and suggest plausible (although fallible) arguments about the likely possibilities of things happening under certain conditions. None of this, however, comes anywhere near the immutable predictive certainty of a scientific laws. Marx, Engels, Toynbee and Buckle all claimed to have discovered laws in history; all were wrong.
Paul Kennedy's The Rise and Fall of the Great Powers, argued that there is a pattern in history where wealthy states created empires, but eventually overstretched their resources and decline. The book used the failure of the Habsburg Empire to achieve European domination in the 16th and 17th centuries as a prediction that the United States would be unable to sustain its global hegemony far into the 21st century. Written in 1987, the book also argued that the Soviet Union was not close to collapse. Within a few years the Soviet Union had indeed collapsed and the world hegemony of the United States appeared more assured than ever. Even in the economic boom of the 1990s the US showed few signs of suffering from the 'imperial overstretch' which Kennedy had prophised. As soon as Kennedy tried to turn his generalisations into laws, he ran into trouble. It is always a mistake for a historian to try to predict the future.
So, history is a science in the weak sense of the word, but it is also an art. Marc Bloch claimed it is also a craft because historians learn on the job how to handle their materials and wield the tools of their trade. History is a varied and protean discipline, and historians approach what they do in different ways That means history can be both scientific, linguistic, literary and a craft.