Julian Coy (African / Asian History MA) 14 December 2001
Discuss the Methodological Issues Raised in the Study of Archives
From the time when Leopold von Ranke wrote that the role of professional historians was to “simply show how it really was” (wie es eigentlich gewesen)1, they have concentrated their research on archival material. Archives are popularly regarded as the primary source for the “raw facts” of history and, despite the wide range of sources now available to the modern historian, from archaeological artefacts to oral testimonies, they remain a fundamental part of historical research. “For the vast majority of historians,” wrote John Tosh, “research is confined to libraries and archives.”2
Why should this be so? Documentary material is likely to find favour with historians because of several distinct advantages. The nature of a document is usually known, from its title page or the circumstances in which it was found, allowing the historian to assess the context in which it was produced and its relative importance to the subject of study. The information it provides can be easily extracted and assessed and referred to again at a later date (assuming the document has not been lost or stolen in the meantime). It can be extremely versatile, appearing in different formats such as a government record, a private letter or diary, a newspaper report, a bill, or a work of fiction. It can also be assessed in many different ways. This can be done directly, analysing the evidence in the way its author intended, as a historical record kept for posterity or as a document with a role within a historical event or process. Or it can be done indirectly, looking at the author’s use of language, or apparent bias and objectivity, or incidental details considered unimportant to the main purpose of the text, or even the material on which the document was written, to reveal “hidden” details of life and attitudes of the period. In this way, according to Marc Bloch, we are able to know “far more of the past than the past itself had thought good to tell us.”3 Finally, a document is created in a moment in time that cannot be revisited. Whatever its purpose (including documents that are plainly fraudulent or forged) it is a witness of its period. E.H.Carr famously wrote that history was “a continuous process of interaction between the historian and his facts, an unending dialogue between the present and the past.”4 Historical documents are vital for continuing that dialogue, revealing the “secrets” of the past to the modern researcher.
“Historical method,” wrote G.R.Elton, “is no more than a recognised and tested way of extracting from what the past has left the true facts and events of that past, and so far as possible their true meaning and interrelation.”5 There are two general methods of approach to the research of archive material. It can be done by either selecting a specific body of material and letting the sources guide you towards a particular topic or conclusion, or by first choosing a task or problem and selecting the sources that seem most appropriate. These “problem-based” and “source-based” approaches both have their respective advantages and disadvantages.
The “source-based” approach is helpful where the archive material is limited, because few documents exist from the period or region and it is important that no detail is left out to present the widest possible picture. Alternatively the amount of material available might be so vast in a particular area that the historian has little idea what will be found until the research begins. The hope is then that a “key” document can be found which is then used to unlock a particular aspect of the research area. The danger is that the research is very much at the mercy of the documents available. The historian may find insufficient evidence with which to draw any conclusions, or simply find that the mass of information is simply incoherent and unusable. In that case, much time will have been wasted, and the historian might leave still convinced that the key documents, like the proverbial needles in the haystack, are present somewhere in the archives, but have proven impossible to find.
As the name suggests, one use for the “problem-based” approach is the solution of particular historical problems and controversies, using the evidence to try and approach these subjects from previously unexplored areas. It is also common to use it to examine previously ignored topics, to create a body of work that is unique and which might bring greater insight onto more general topics. For example, the SOAS historian Richard Sims decided to explore the relationship between France and Japan after the Meiji Restoration, as it had been largely shunned up till then. As well as uncovering specific details of the diplomatic activities between the two countries, he believed his study more generally revealed “the superficiality of some of the assumptions and generalisations about Western policies towards nineteenth-century Japan.”6
A disadvantage of the approach is that is often impossible to know beforehand whether any of the material will be relevant to the research. The historian should also avoid the danger of “Mining the source”, becoming too closely attached to the first evidence that appears to confirm their theories, or being too accepting of the author’s bias and preconceptions, because a proper balance has not been sought. Efforts should be made to find as large and as varied a body of evidence as possible that supports the historian’s case, and they should search actively for anything that appears to refute or contradict it. These exceptions can then be examined to see if they are either merely anomalies, or are signs that the initial hypothesis is flawed. “The deeper the research,” wrote Bloch, “the more the light of the evidence must converge from sources of many different kinds.”7
In practice, therefore, the two approaches tend to overlap. It is helpful to have a general idea or subject area, if only so the sources can be narrowed down due to the constraints of time. Few historians have the luxury of an unlimited amount of time in which they can carry out their research, so will set boundaries in order to ensure prompt results to please their supervisor or publisher. However, it remains necessary to let the sources “speak for themselves”. “Preconceived notions,” Elton warned, “are a much greater danger to historical truth than either deficiency of evidence or error in detail.”8 It is vital that the historian should prepared to change their theories and assumptions if the evidence finds them wanting or even halt their research altogether if it proves insufficient for the task in hand. “Without this flexibility,” wrote Tosh, “historians risk imposing on their evidence and failing to tap its full potential.”9 Often much research is done by returning to familiar sources and assessing them in the light of new theories. The same rules apply about rejecting the theory if it is found to be insufficient, but the process of analysing the evidence from a new angle may still bring new insights to light.
Carlo Ginzburg’s “The Cheese and the Worms” is an example of a historian who approached an archive with a specific task in mind, but was prepared to follow his evidence in whatever direction it took him.10 Researching the is records of the Inquisition in Sixteenth Century Italy, he came across the trial of a miller from north-eastern Italy, Domenico Scandella (nicknamed Mennochio). He used the details from Mennochio’s interrogations to piece together his beliefs and views of the world (which were sufficiently extreme that he was eventually executed) and explore how these might have come about through the man’s reading and his experiences within that society and culture. The source may have been a “lucky find” but Ginzburg, having already built up a substantial background knowledge on the subject area, was able to analyse it in sufficient depth to demonstrate that, as Jim Sharpe wrote, “even meagre, scattered and obscure documents can be put to good use.”11
Having decided upon a general “strategy”, the historian must consider the basic “tactics” of research, beginning with the basic preparations to be made before visiting the archive for the first time. “To use unpublished sources fruitfully,” wrote Philip C. Brookes, “the investigator…should know their nature and background, their physical characteristics, and the problems they present.”12 Clearly it is crucial that historians are well versed in the background material and latest scholarship of their subject area, so that the context and content of their sources can be more easily understood, and any obvious errors involving fraudulent or misplaced material can be quickly spotted.
Good background knowledge is essential for choosing what and where to study, the most appropriate records and the archive in which they can be found. A historian of British government procedure is likely to begin with the Public Records Office in Kew, but may also need to hunt for evidence in one of the County Record Offices, or search through a private collection. They might even need to travel abroad to compare the approaches of other countries during the period. “There are practically no topics of major research…” commented Brookes, “that can be studied adequately in one repository of papers.”13 However, it is certainly sensible to make enquiries beforehand to avoid travelling a great distance only to find that the relevant material is unavailable.
Further preparation might be required to examine the nature of each archive – how and why the documents were gathered together - and whether there is a chance that any material has been lost. Documents are of greatest value to the historian if they have been retained within their original collection, allowing their interrelationship to be recorded more easily. Describing the collection of the British Public Records Office, V.H.Galbraith wrote that it formed an “organic structure, systematically recording the growth of the State, from the simple times when all government was a function of the king’s household to the multifarious departments charged with contemporary administration.”14 A collection as comprehensive as this allows the historian to compare documents from different departments and periods, seeing how institutions differed from each other, and how they evolved over time. Other records may exist in an artificial collection, gathered together, perhaps originally by a private collector, under a common format or around a specific theme. Comparisons can still be made between the documents, but the historian must be aware that the order has been imposed upon them by an external source. Finally there are those documents that stand on their own, or have been gathered together at random. These will require plenty of effort to define the context within which they originally stood.
It is important to bear in mind that the conditions in which the original documents were written is likely to affect the quality of information available to us today. “We are, for instance,” wrote Galbraith, “much better informed throughout the Middle Ages about the affairs of the Church than of the State, simply because literacy was a clerical monopoly.”15 The criteria by which the documents were selected and kept for records will also have a bearing on what remains. The Letters and Papers collection of the Tudor government, for example, is mainly restricted to the letters that were received. Few copies were made of those that were sent, so the historian receives only half of the picture of government affairs during the period.16 Other records may have been collected but then subsequently lost, either misplaced or accidentally destroyed. Many Privy Council papers were burned during a fire in Whitehall in 1619, and nearly all of the House of Commons’ records went up in flames in the Palace of Westminster in 1834. Others may have been stored in such poor conditions that they have rotted away, or been eaten by vermin. The archives of the early Chinese Communist Party had to cope with being carried along on the Long March in 1934 and 1935 across China’s harshest terrain. Inevitably many documents were lost along the way, but it is a miracle that any have survived at all.
Other documents have been deliberately “lost”, due to their potential to cause embarrassment for some individuals or regimes. An individual may ask for certain articles to be kept from public view when donating records to an archive. Access may also be restricted if the government feels that the documents cover a sensitive topic, such as the Abdication Crisis of 1936, for example, which understandably remains a personal issue for the Royal Family. Historians of Eastern Europe probably enjoyed greater access to the archives after the collapse of Communism than they do at present, as new governments have become aware of the damage candid records of the past could do to their credibility.
The United Kingdom operates a “Thirty Year Rule” for cabinet papers and sensitive material, preventing current members of the government from being publicly compromised, and the United States’ Freedom of Information Act contains exemptions for any matters of defence, foreign policy and other areas that it considers to be “secret”. So, historians of most recent history may have to cope with a more restricted body of material than their colleagues studying an earlier period.
The other important preparation the historian must make before beginning research is to foster a good relationship with the archivist. Ideally, according to Brookes, “A competent archivist is to be looked upon as a scholarly colleague of the researcher, far more than solely a researcher or caretaker.”17 It may well be that the archivist’s knowledge can direct the historian to the most useful sources, and warn of any potential pitfalls. More generally, the archivist’s advice concerning the accuracy of the catalogues and most efficient search procedures will also save plenty of time and effort. Historians working in some parts of the world might well add that there is one more important consideration to make about the archivists – the size of the bribe required before they are prepared to reveal where the most useful documents in the archive are located.
Once the historian has chosen the archive, established a relationship with the archivist, and selected the documents, the research can finally begin. The first task should be “external criticism”, testing the validity of the evidence. If a historian’s research is to be accepted, the evidence on which it is based must be proven to be authentic. Its age can be checked by analysing the language used, the handwriting or typeface, the composition of the paper itself, aided by chemical analysis and carbon dating. The text can be checked for consistency of the facts and whether the subject matter and incidental detail match the period. Some correspondence of Abraham Lincoln was proven to be fraudulent, when several small errors were discovered, including references to “Kansas” – this name did not come into public use until several years later.18 The provenance of a document can be checked by tracing its history up to its insertion in the archives, looking for the office from which it originated, and the likelihood of its author being there at the time. These are all “basic checks” which are likely to prove that nothing is amiss as the chance of finding a forgery is rare. However, the potential for embarrassment is so great, as Hugh Trevor-Roper discovered with the false Hitler diaries, that it is vital that such checks are carried out as extensively as possible.
The historian’s “internal criticism”, assessing the content of the sources, often begins by attempting to divide them into suitable categories. This could either be between primary and secondary sources, contemporary records and those written at one remove, or between published and unpublished sources, printed material and manuscripts. Primary sources should not be considered to be free from bias or inaccuracy simply because their author was present at or initiated a historical event. The writer’s opinions and access to information will certainly affect their content. Secondary sources may have the advantage of a broader viewpoint. A historian writing an account of the battle of Waterloo, for example, is likely to have had a greater understanding of what took place that day than a soldier writing in his diary, but the soldier still had the advantage (if one can call it that) of understanding how it felt to be in battle – the sounds, the smells, and the emotions he experienced.
The second category is easier to define from the physical appearance of the documents, but a more important consideration is whether the documents were meant to be left for posterity. An author who is aware that their work might be consulted in the future is likely to be more selective with their information, leaving out what they consider to be trivial and unimportant. Certainly a historian will try to get “behind” such published material if possible, examining the letters, diaries and papers that led up to its publication to try and get the broader view. An important technique when examining primary sources is to look for the incidental detail, the descriptions and passing comments, to build up a picture of the “everyday” events and attitudes that would otherwise be forgotten, encouraging them to be, as Bloch famously said, “witnesses in spite of themselves”19. Natalie Zemon Davis used this approach in her work “Fiction in the Archives: Pardon Tales and their Tellers in Sixteenth Century France”, using legal documents, letters written to the king asking for a pardon for a murder conviction, to examine the status of women and contemporary attitudes towards them. The letters provide interesting insights into the relationships of married couples, some acknowledging that the husband had the right to beat their wives and that those who murdered their wives having caught them committing adultery “achieved grace easily”.
Davis’ work utilises another technique to draw more information from the letters, analysing the style of the writing and comparing it to fictional narratives popular at the time, presenting the supplicant’s ordeals in a well-known form. The historian must be aware that the language of the document may well carry special meanings unique to that period which have now been lost. “The language of historical documents is never transparent,” wrote Richard J Evans, “and historians have long been aware that they cannot simply gaze through it to the historical reality behind.”20 It is important to bear in mind the author’s original intentions, the tone they wished to convey to the reader, and the desired result that the document might have. Clearly historians are at a disadvantage if they cannot speak the original language of the document. John Iliffe, in his history of poverty in Africa, had to restrict his study to published sources in Europe for this reason, and so was unable to read the few unpublished African sources that remained, running the risk of portraying the continent from only the colonial point of view. He apologised that he could only provide a “rough outline” of the subject.21
Davis and Iliffe’s work are both examples of attempts to use archives, assumed to be the preserve of governments and institutions, to tell the “history from below”, that of minority groups - French women, the African poor, the Chinese peasantry. Their contributions to the development of modern society have largely been ignored by traditional historical accounts. E.P.Thompson, in his influential work, “The Making of the English Working Class” set out to use archives such as the Home Office papers, municipal records, contemporary newspaper accounts, the minutes of local societies, letters and memoirs, to examine the development of the class as a social and cultural formation, during the early nineteenth century. “I am seeking to rescue the poor stockinger,” he wrote, “the Luddite cropper, the “obsolete” hand-loom weaver, the “utopian” artisan…from the enormous condescension of posterity.”22 Roger Chartier attempted to examine a specific aspect of the life of the French peasantry by studying an ecclesiastical survey of their reading habits. Although responses to the survey were limited, some that were received were not encouraging for the clergy, portraying “a world of superstitious beliefs, trivial fables and long-standing prejudices.”23 This undoubtedly encouraged some priests to write on behalf of their communities, presenting a more idealised view of peasant culture, of families gathering round the fire to read aloud to each other from the scriptures. So the survey, according to Chartier, “is a better indication of the nostalgia or the expectations of literate people at the end of the Eighteenth Century than it is of peasant practices.”24
Carlo Ginzburg’s analysis of the worldview of Mennochio is an important example of how “Microhistory” can be brought out of the archives, using close analysis of a very specific subject to make broader conclusions and generalisations about the society in which it was set. “Phenomena previously considered to be sufficiently described and understood,” wrote Giovanni Levy, “assume completely new meanings by altering the scale of observation.”25 Emmanuel Le Roy Ladurie attempted a similar project with his study of the village of Montaillou, using inquisition records during an investigation of heresy between 1318 and 1325. He used the peasants’ own words during their interrogations to build up a more general picture of the way of life and world view of rural society during that period.26
Alain Corbin had similar ambitions of studying the experiences and views of the masses, in his case the urban population of nineteenth-century France, but subscribed more to the view of “Total History”, popularised by Fernand Braudel, Marc Bloch and other members of the French “Annales” school. This proposed the use of archive material from every possible source (and any other historical evidence that could be found, including architecture, archaeology and oral testimony) to create as detailed a picture as was possible. His work “The Foul and the Fragrant: Odour and the French Imagination” attempted to describe the smells one would come across in France in the early nineteenth century, and how the general attitudes to odour and hygiene were fundamentally different from our own. He used a wide variety of contemporary material, including medical journals and popular literature, to show how these attitudes changed, from a general acceptance of the stink of excrement to the point where tolerance of smell and attitude to hygiene defined one’s social class. “The attachment of the great mass of the population to strong, foul odours,” he wrote, “despite the injunctions of the privileged classes, might provide a mode of access to the history of social psychology.”27
Corbin, Ginzburg and Thompson’s work all demonstrate how archive material can be used to go further than simply studying the policies and activities of the prominent individuals and institutions of each country, although clearly it is important that research into these areas must also continue. The archives may also hold other material beyond the traditional documents. There is a vast amount of statistical information - census records, parish registers, account books and electoral returns – that can be examined either individually or more generally, using the techniques of quantitative analysis to discover patterns and trends. Bloch’s “witnesses in spite of themselves” technique can be used on such records to draw out information the original compilers might have considered incidental. Army recruitment records were examined to chart the average height of British men during the nineteenth century, discovering that the long term upward trend went into reverse in the 1850s, despite improvements in the British public’s health and diet.28 Many archives now contain transcripts of oral material, recorded as part of a concerted effort to preserve a nation’s popular memory. The records of the Australian War Memorial include interviews with Anzacs of their experiences at Gallipoli, an important part of the country’s national identity. The American national archives include interviews from the 1930s with the last remaining citizens who had experienced slavery. In each case the questioning was suitably broad, so that future historians could approach the transcript material from different perspectives.
A challenge for the historian of the future will be the increasing impact of the computer age on historical records. Already computers have greatly enhanced archive catalogues and search techniques, have contributed to the developments within quantitative analysis, and have made it possible for vast amounts of material to be stored and retrieved easily. However, the explosion in the amount of information available also has its drawbacks. Many records are now being lost through ignorance or simply through lack of storage space. The “incidental detail” so important for social and cultural study can disappear as redundant documents are thrown out and computers are replaced. It is impossible to retain everything but a greater understanding is required of an article’s potential as a historical document. “One of the great problems is that even historically minded people usually think of only older materials as of value,” said Brookes.29 His comments predate the age of mass computing and the Internet, but they remain valid today.
Despite the introduction of many new techniques and perspectives in the historical profession, archives continue to play a key role within historical research. Although it is essential that alternative forms of evidence are studied alongside, such as artefacts, oral testimonies and, for more recent periods, audio-visual material, the examination of documentary evidence will remain a fundamental skill for the historian. The basic techniques of preparation and authentication remain essential if the historian’s work is to stand up to external criticism, but now there is a greater freedom to pursue subjects and issues that until recently might have been considered trivial and irrelevant. The archives certainly contain much about the histories of great figures and regimes, but they also can provide insights into the experiences of the tin-miners in Cornwall, the mentality of a miller in Italy, the treatment of widows in France, or the smell of the Parisian sewers. Although incomplete and fragmented, the archives that remain are so vast that the possibilities for further research and historical discovery are infinite.
General works on historiography and historical research
Bloch, M. The Historian’s Craft (Manchester: Manchester University Press, 1954)
Brooks, P.C. Research in the Archives: The Use of Unpublished Primary Sources (Chicago: University of Chicago Press, 1969)
Burke, P. (ed.) New Perspectives on Historical Writing (London: Polity Press, second edition, 1991)
Carr, E.H. What is History? (London: Penguin Books, second edition, 1987)
Elton, G.R. The Practice of History (London, Fontana, 1987)
Evans, R.J. In Defence of History, (London: Granta Books, 1997)
Galbraith, V.H. An Introduction to the Study of History (London: C.A.Watts & Co., 1964)
Tosh, J. The Pursuit of History: Aims, Methods and New Directions in the Study of Modern History (London: Longman, third edition, 1999)
Examples of historical methodology
Chartier, R. Cultural History: Between Practices and Representations (Oxford: Polity Press, 1988)
Corbin, A. The Foul and the Fragrant: Odour and the French Imagination (Cambridge, MA: Harvard University Press, 1986),
Davis, N.Z. Fiction in the Archives: Pardon Tales and their Tellers in Sixteenth-Century France (Cambridge: Polity Press, 1988)
Ginzburg, C. The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller (Baltimore: John Hopkins University Press, 1980)
Iliffe, J. The African Poor, (Cambridge: Cambridge University Press, 1987)
LeRoy Ladurie,E Montaillou: Cathars and Catholics in a French Village, 1294-1324, (London: Penguin,1980)
Sims, R. French Policy towards the Bakufu and Meiji Japan, 1854-95 (Japan Library, Richmond, Surrey, 1998)
Thompson, E.P. The Making of the English Working Class (London: Penguin Books, revised edition, 1968)