If citing, please kindly acknowledge copyright © Penelope J. Corfield (2016)

‘Evidence, evidence: I hate that word’, a vehement colleague in the English Department once hissed at me, when I had, all unawares, invoked the word in the course of an argument. I was surprised at his vehemence but put it down to a touch of dyspepsia, aggravated by an overdose of (then) ultra-fashionable postmodernist doubt. What on earth was he teaching his students? To disregard evidence and invent things as the passing mood dictated? To apply theory arbitrarily? No need to bother about dates, precision or details. No need to check one’s hunches against any external data or criterion of judgment. And certainly no need to analyse anything unpleasant or inconvenient or complexly difficult about the past.1

But I thought my colleague’s distaste for evidence was no more than a passing fad. (The date was sometime in the later 1990s). And indeed intellectual postmodernism, which was an assertive philosophy of doubt (a bit of a contradiction in terms, since a philosophy of doubt should be suitably doubtful), has faded even faster than the postmodernist style of architectural whimsy has been absorbed into the architectural lexicon.2

Fig. 1 The Rašin Building, Prague, known as the Dancing House, designed by V. Milunić and F. Gehry (completed 1996) – challenging classical symmetry and modernist order yet demanding absolute confidence in the conventional solidity of its building materials. Image by © Paul Seheult/Eye Ubiquitous/Corbis

Fig. 1 The Rašin Building, Prague, known as the Dancing House, designed by V. Milunić and F. Gehry (completed 1996) – challenging classical symmetry and modernist order yet demanding absolute confidence in the conventional solidity of its building materials.
Image by © Paul Seheult/Eye Ubiquitous/Corbis

Then, just a week ago, I was talking to a History postgraduate on the same theme. Again to my surprise, he was, if not quite as hostile, at least as hesitant about the value of evidence. Oh really? Of course, the myriad forms of evidence do not ‘speak for themselves’. They are analysed and interpreted by historians, who often disagree. But that’s the point. The debates are then reviewed and redebated, with reference again to the evidence – including, it may be, new evidence.

These arguments continue not only between historians and students, but across the generations. The stock of human knowledge is constantly being created and endlessly adjusted as it is transmitted through time. And debates are ultimately decided, not by reference to one expert authority (X says this; Y says that) but to the evidence, as collectively shared, debated, pummelled, assessed and reassessed.

So let’s argue the proposition the other way round. Let’s laud to the skies the infinite value of evidence, without which historians would just be sharing our prejudices and comparing our passing moods. But ok, let’s also clarify. What we are seeking is not just ‘evidence’ A, B or C in the cold abstract. That no more resolves anything than does the unsupported testimony of historian X, Y or Z. What we need is critically assessed evidence – and lots of it, so that different forms of evidence can be tested against each other and debated together.

For historians, anything and everything is grist to the mill. If there was a time when we studied nothing but written documents, that era has long gone. Any and every legacy from the past is potential evidence: fragments of pottery, swatches of textiles, collections of bones, DNA records, rubbish tips, ruined or surviving buildings, ground plans, all manufactured objects (whether whole or in parts), paintings from cave to canvas, photos, poems, songs, sayings, myths, fairy tales, jokes … let alone all evidence constructed or reconstructed by historians, including statistics, graphs, databases, interpretative websites … and so forth. Great. That list sounds exhausting but it’s actually exhilarating.

However, the diversity of these potential sources, and the nebulousness of some forms of evidence (jokes, fairy tales), indicate one vital accompaniment. Historians should swear not only by the sources but by a rigorous source critique. After asking: what are your sources? the next question should be: how good are your sources, for whatever purpose you intend to deploy them? (These stock questions or variants upon them, keep many an academic seminar going).

Source auditing: here are three opening questions to pose, with reference to any potential source or set of sources. Firstly: Provenance. Where does the source come from? How has it survived from its original state through to the present day? How well authenticated is it? Has it been amended or changed over time? (There are numerous technical tests that can be used to check datings and internal consistency). No wonder that historians appreciate using sources that have been collected in museums, archives or other repositories, because usually these institutions have already done the work of authenticating. But it’s always well to double-check.

Secondly: Reliability of Sources and/or Methodology. A source or group of sources may be authentic but not necessarily reliable, in the sense of being precise or accurate. Evidence from the past has no duty to be anything other than what it is. A song about ‘happy times’ is no proof that there were past happy times. Only that there was a song to that effect. But that’s fine. That tells historians something about the history of songs – a fruitful field, provided that the lyrics are not taken as written affidavits.3 All sources have their own intrinsic characteristics and special nature, including flaws, biases, and omissions. These need to be understood before the source is deployed in argument. The general rule is that: problems don’t matter too much, as long as they are fully taken into account. (Though it does depend upon the nature of the problem. Fake and forged documents are evidence for the history of fakery and forgery, not for whatever instance or event they purport to illuminate).

One example of valid material that needs to be used with due caution is the case of edited texts whose originals have disappeared, or are no longer available for consultation. That difficulty applies to quite a number of old editions of letters and diaries, which cannot now be checked. For the most part, historians have to take on trust the accuracy of the editorial work. Yet we often don’t know what, if anything, has been omitted. So it is rash to draw conclusions based upon silences in the text – since the original authors may have been quietly censored by later editors.4

When auditing sources, it also follows that a related test should also be addressed to any methodology used in processing sources: is the methodology valid and reliable? Does it augment or diminish the value of the original(s)? Indeed, is the basic evidence solid enough to bear the weight of the analytical superstructure?

Thirdly: Typicality. With every source or group of sources, it’s also helpful to pose the question as to whether it is likely to be commonplace or highly unusual? Again, it doesn’t matter which it is, as long as the historian is fully aware of the implications. Otherwise, there is a danger of generalising from something that is in fact a rarity. Assessing typicality is not always easy, especially in the case of obscure, fragmentary or fugitive sources. Yet it’s always helpful to bear this question in mind.


Overall, the greater the range and variety of sources that can be identified and assessed the better. Everything (to repeat) is grist to the mill. Sources can be compared and contrasted. Different kinds of evidence can be used in a myriad of ways. The potential within every source is thrilling. Evidence is invaluable – not to be dismissed, on the grounds that some evidence is fallible, but to be savoured with full critical engagement, as vital for knowledge. That state of affairs does include knowing what we don’t (currently) know as well as what we do. Scepticism fine. Corrosive, dismissive, and ultimately boring know-nothingism, no way!

*NB: Having found and audited sources, the following stages of source analysis will be considered in next month’s BLOG.

1 BLOG dedicated to all past students on the Core Course of Royal Holloway (London University)’s MA in Modern History: Power, Culture, Society, for fertile discussions, week in, week out.

2 For the fading of philosophical postmodernism, see various studies on After- or Post-Postmodernism, including C.K. Brooks (ed.), Beyond Postmodernism: On to the Post-Contemporary (Newcastle upon Tyne, 2013); and G. Myerson, Ecology and the End of Postmodernism (Cambridge, 2001), p. 74: with prescient comment ‘it [Postmodernism] is slipping into the strange history of those futures that did not materialise’.

3 See e.g. R. Palmer, The Sound of History: Songs and Social Comment (Oxford, 1988).

4 A classic case was the excision of religious fervour from the seventeenth-century Memoirs of Edmund Ludlow by eighteenth-century editors, giving the Memoirs a secular tone which was long, but wrongly, accepted as authentic: see B. Worden, Roundhead Reputations: The English Civil Wars and the Passions of Posterity (2002).

For further discussion, see

To read other discussion-points, please click here

To download Monthly Blog 66 please click here