Hello again everyone. It’s been too long since I’ve posted here. One of the reasons for my absence was that I have been working my way through several new works in other languages that I have had to scan and translate mostly “by machine” as I go. Reading one work led to several more and so it went. One result: I now have many new perspectives and questions relating to the New Testament and related literature, especially (but not only) to the Gospel of Mark, Book of Revelation and the Ascension of Isaiah.
While I was reading I sometimes sought escape by kind of doodling on and off on the Biblical Criticism & History Forum – earlywritings.com. While there, I had occasion to list what historians themselves have explained are the building blocks of historical research. That is, their own explanations of how they determine the facts of what happened long ago. From these raw facts historians reconstruct history itself and develop hypotheses about causes and the nature of the cultures and so forth, but “bedrock facts” come first. (Not that the information I posted on the forum had much impact since certain persons continued to discuss historical questions according to their long-held habits of thought that break all the rules for determining “facts”. They’re having fun and maybe that’s what matters most to some of them. C’est la vie.)
In sum, the methods common to historians and that they themselves have explained are these:
- Look for a “primary source”, generally meaning a source that is contemporary with the person or event to which it testifies.
One example: Accounts of events and persons are found in writings that we have valid reasons to believe were produced by contemporaries of those persons and events.
- Look for reasons to have some level of confidence in those sources — or not: e.g. do we know who wrote them and why?
- Look for independent corroboration of the information in the sources or for general trustworthiness of the source.
An example I like to use is Socrates. How do historians know Socrates existed?
We have writings about Socrates that we can determine were written by his students (e.g. Plato, Xenophon), other writings that confirm that those texts are indeed by whom they say they are (e.g. Aristotle’s references to Plato’s works), and we also have contemporary works that are critical of Socrates, mocking him (e.g. by the playwright Aristophanes) — that is, confirmation of Socrates that appears to be independent of the works of Plato and Xenophon.
Those principles are not always spelled out by historians in their publications but they are generally noticeable to any reader who is looking for “how they know” what they are writing about.
Below is a collation of various quotations by historians and philosophers of history that do make the above principles explicit. It is a revised copy of what I posted on the BC&H forum.
Rules of historical reasoning
The New Testament scholar Richard Bauckham has argued that a historian should give the benefit of the doubt to any testimony. That is a fine starting principle when one needs to get along with neighbours and colleagues, but few nonbiblical historians would agree that it applies to the sources from long ago.
If we are deciding whether to believe a witness or documentary source, the hypothesis is that the source is true. The Bayesian equations would therefore have us consider which makes the testimonial evidence more likely: that the source was true, or that it was not. That decision might be based on knowledge of the particular case. . . . Where a witness has a proven track record and we have no reason to doubt their word, we would conclude the opposite. Finally, the judgement might be based on corroborating testimony. (p. 45)
Day has a section headed “Rules of historical reasoning”. I’ll quote from it:
So what is the methodology that historians typically take to govern their practice?
Before describing the rules of historical reasoning in more detail, a brief note on my methodology. By examining what is recommended, praised and criticized, we can arrive at an approximation of the rules which govern the production of historical writing. For while one can’t infer a norm simply from observing what is and is not done – since people get things wrong ignorantly, negligently, and deliberately – the inference of a norm from others’ recommendations and responses to what is and is not done is more plausible. I have used ‘historiographical manuals’ – those books written for the student of history, and in particular postgraduate or PhD students of departments of history to elucidate the method of source criticism. I have used peer review of professional historiographical monographs to investigate wider rules governing the practice.
From historiographical manuals we gain the appreciation that the historical practice has, at its heart, the Rankean method of source criticism. All historiographical claims should be based on the sources. . . . What follows are five points concerning the use of sources, each of which is consistently emphasized by pedagogical material of the above kind.
“What does Mark Day mean by a primary source?
The ‘primary source’ is . . . the only building block . . . essential for . . critical history. (p. 27)
Day explains that one of the conditions that makes a source “primary” in the literature is that it is contemporary or “temporally proximate” with the event to which it refers. But he gives as one example of a document that is generally considered a primary source for the “daily lives of the Catalan peasants” in the early fourteenth century: the document is the Inquisition Register of the bishop of a region in southern France but it was written “some eighteen years after the events reported”. So does it really qualify as a “primary source” for the daily lives of the peasants eighteen years earlier?
Another historian has said that twenty years is a gap that must arouse suspicion:
It is typical of popular tradition that it is first heard of long after the time when the events it reports are supposed to have occurred. Almost invariably there is a gap, more or less broad, between the events and their first appearance in recorded history. Such a gap occurring in the case of any report is enough to make it suspect from the start. Instances of such reports, found on examination to be unverified, are without number. Thus, unaccountably tardy first mention of them in written record of any kind is a major argument used by critics in discrediting such onetime general beliefs as the False Decretals, the Popess Joan, the authenticity of the reputed works of Denis the Areopagite. Again, no contemporary biographer of St. Thomas of Canterbury records that his mother was a Saracen princess whom his father had married in the Holy Land. John Morris, “Legends about St. Thomas,” The Life and Martyrdom of St. Thomas Archbishop of Canterbury( 2d ed., London, 1885), 52325.
That Luther committed suicide is a story first heard of some twenty years after his death, when it began to be circulated by persons hostile to his memory. H. Grisar, Martin Luther, his Life and Work . . . . (Garraghan, p. 265)
(1) In common with Ranke’s approach, the historian should prioritize primary sources, though should nonetheless be critical of these sources.
(2) Criticism of sources is two-fold; not only with regard to the claims of those sources concerning their intended topic, but with regard to the implicit claims of those sources concerning themselves. The second sort of criticism is the investigation into the document’s authenticity, established by asking whether the author could have written it, whether they could have been where they claimed to be, whether the paper, authorial style and handwriting permit the truth of the self-proclamation of the author. (So far we have not departed from critical SAP.)
(3) Source criticism is extended beyond the establishment of the identity of the author, to so-called ‘internal’ features of the source: the author’s aim, their ideological background and their intended audience. It is assumed that knowledge of these facts will aid the historian’s use of the source. (Exemplification of this point has already been suggested, in the case where the historian would be wise to find out whether the author had reason to lie, and why they might have done so.)
(4) Source criticism should also trace the path connecting the source with the historian, asking why it has survived and in the form that it has. . . .
(5) The historian is warned not to depend too much on a single document, but rather to utilize a wide range of evidence. This warning is to some extent implicit in the demand for source criticism, since it is obvious that no serious source criticism can proceed without employing knowledge gained from other sources.
Historiographical manuals tend to emphasize the rules of source criticism. Peer reviews tend to emphasize different rules that govern historiography. We can suppose that the former are regarded as more basic, in terms of both difficulty and perhaps importance.
The following quotation has special significance for sources of ancient history since they are often much later than the times and persons they describe. All of our written histories of Alexander the Great are from centuries after he lived. But note: those late histories have some use for us today because they cite other sources contemporary with Alexander. See Comparing the sources for Alexander and Jesus for the details.
Historians usually exclude evidence written by authors that were separated by time or space from the events they describe, unless there is evidence for a credible causal chain that could have transmitted information from events to evidence. Historians may assess the fidelity of some types of evidence without evidence for a causal-information chain if other independent sources consistently agree or disagree with the information conveyed by particular authors, sources, or institutions. For example, though Diodorus and Pausanias are both sources for ancient Greek history who were separated by centuries from the events they described, Pausanias is valued as having a higher fidelity than Diodorus because other testimonies and material evidence confirm much of what he wrote (Kosso, 2001, p. 120).
That statement has significance for Christian origins studies, obviously. The “causal chain” most often proposed between the events of Jesus and the gospels is “oral tradition”, and Q. Unfortunately, both of those “causal chains” are hypothetical. There is no evidence for them. (You may be wondering how we could have surviving evidence for a tradition that was passed on entirely by word of mouth. One example: Jewish writings of late antiquity document and explain that debating points were part of the oral teachings of various rabbis generations earlier.)
That’s not to say that oral tradition or now-lost texts behind the gospels did not exist. But if other explanations can be found for the sources that do not need to rely upon the hypothetical links then, all things being equal, one would expect those other explanations to have priority.
Now, never let an author get away with an unchecked citation, I have learned. Tucker cited Kosso and here is Kosso in his own words (p. 20):
These historical accounts [by Diodorus Siculus and Plutarch] might be thought of as direct references in that they mention cleruchies [i.e. types of Athenian settlements outside Attica] explicitly, but it would be misleading to regard them as direct information about events in the past. This is information that has been passed along by the historian, via his sources and with his writing, through time. There is work to be done in securing the credibility of these ancient writers on this particular subject. Diodorus and Plutarch, after all, were describing events that had taken place centuries earlier and their evidence is acceptable only insofar as we can keep track of how they would know about these things. Diodorus, in fact, is often inaccurate in his tales of ancient Greece, underscoring the need for justification of his particular claims about cleruchies. Pausanias, also writing centuries after cleruchs had come and gone, likewise makes explicit reference to the cleruchs sent to Euboea and Naxos, and Pausanias enjoys a good reputation for accuracy. This is due largely to his frequent description of landscapes and monuments, features of the land and culture that are preserved today and can be checked against his account. Here again, archaeology, the study of the material record, is used in accounting for historical evidence.
Back to Tucker, p. 122
Historians search assiduously for primary sources because they have higher fidelity than secondary sources. A major part of the education of historians consists of learning to distinguish primary from secondary sources. Historians usually agree on the classifications of sources. Still, the abstract definition of primary sources is more challenging. Marwick suggested first that “primary sources are sources which came into existence during the actual period of the past which the historian is studying, they are those relics and traces left by the past, while secondary sources are those accounts written later by historians looking back” (1993, p. 199). However, some primary sources come into existence long after the events the historian is studying; for example, later copies of ancient manuscripts such as the Greek classics.
Turning to a later edition of that work by Marwick (p. 26):
Manifestly, one cannot travel to the past by ship or plane, or even through e-mail or the internet. This is a very serious point: the only way we can have knowledge of the past is through studying the relics and traces left by past societies, which I have already mentioned. Primary sources, as it were, form the basic ‘raw material’ of history; they are sources which came into existence within the period being investigated. The articles and books written up later by historians, drawing upon these primary sources, converting the raw material into history, are secondary sources (pedants insist on pointing out that secondary sources may become primary sources for still later historians, but this is a matter of such triviality as scarcely to be worth bothering about). The distinction between primary and secondary sources is a critical one, though no historian has ever pretended that it offers a magic key to the nature of historical study, or that primary sources have a necromantic potency denied to secondary ones. There is always some excitement about being in contact with a genuine primary source, but one will not learn very much from a single source. Reading through an edited selection of excerpts from primary sources will have the salutary effect of bringing one in contact with the thinking and language of past generations, but it will not amount to research. If the ordinary reader, or history student, wants to learn quickly about the role and status of women during the Renaissance, or about the causes of the First World War, they will be well advised to go to the secondary authorities, a knowledge of the principles of history being useful in separating out the more reliable from the less.
But if you are planning to make an original contribution to historical knowledge, you are unlikely to make much of a stir if you stick strictly to other people’s work, that is, the secondary sources – to which, it should be stressed, the research historian will frequently return throughout all stages of research and writing. The difference is critical in that strategy which all historians, in one way or another, devise in embarking on a new research project. It is through the secondary sources that one becomes aware of the gaps in knowledge, problems unsolved, suspect explanations. It is with the aid of these secondary sources, and all the other resources of the profession, that one begins to identify the archives in which one will commence one’s researches.
Primary sources, numbingly copious in some areas, are scarce and fragmentary in others. Much has to be garnered indirectly and by inference. Historians do not rely on single sources, but are always seeking corroboration, qualification, correction; the production of history is very much a matter of accumulating details, refining nuances. The technical skills of the historian lie in sorting these matters out, in understanding how and why a particular source came into existence, how relevant it is to the topic under investigation, and, obviously, the particular codes or language in accordance with which the particular source came into being as a concrete artefact.
Corroboration is what I have referred to in other posts as “external controls”. Obviously, to avoid circularity, a corroborating or supporting source needs to be independent of the source it backs up, as is made clear in the quotations that follow:
Historians cannot transcend the evidence to check its reliability against history. They must bootstrap their evaluations of the fidelity of particular evidence by other evidence. Historians use independent sources to evaluate each other’s fidelity. Independent sources testify to the fidelity of a source or provide information on what the author could and could not have seen or known, as well as information on the information causal chain that may or may not have connected an event or a document with the evidence (Kosso, 1993, pp. 3–4). This is the fundamental practice of historians, which is taught as part of the historiographic guild’s right of passage (Evans, 1999, pp. 16–17). This is also the great success story of modern historiography as its countless exposures of forgeries may attest. (Tucker, p. 123)
Here is Kosso from that 1993 citation in full:
Our knowledge of the past, like much of our knowledge of the present, is based on evidence, the informational link between the observable and the interesting. But evidential claims must themselves be justified, and a thorough epistemology of history or of science must describe both the role of evidence as supplier of justification and the need by evidence to be justified. . . . . The goal is to understand and evaluate how textual information of the distant past comes to be deemed accurate and reliable. . . . .
. . . .The guiding question here is not, as the historian initially asks, Is Thucydides believable? or even the less tendentious, What in Thucydides is believable? It is rather the methodological and epistemological question, What are the available standards for evaluating the credibility of Thucydides? pp. 1f.
Thus, when Pausanius’s written description of architectural monuments and natural terrain of ancient Greece agrees with present observations of the material remains, both archaeological evidence (claims about dates and uses of objects, for example) and the textual evidence in Pausanius receive a measure of justification.
Mutual corroboration can also occur between texts if different authors make references to the same or to related things. To the degree that their accounts are consistent and of explanatory relevance to each other, each gains credibility. They can function, in other words, as accounting claims for each other. But there is epistemic benefit in their agreement only if they are independent sources of information in the sense that neither uses the other as a source. This is a concern, for example, in noting the agreement in accounts of Diodorus Siculus and Thucydides. Much of what Diodorus says may well be lifted from Thucydides, thus building in from the start the consistency between the two, and thereby making the comparison no test for either. Since neither source of evidence is epistemically privileged (neither, that is, brings a special status of justification into the comparison), epistemic benefit derives only from their being independent. Comparison to an independent source is a genuine risk, and hence agreement is a genuine accomplishment.
Another kind of accounting claim occurs when one text refers to another. For example, one text may refer to the author of another, thereby providing a description of how the evidence was produced. Claims about authors could be general claims about their methods, for example, descriptions of their skills as observers and reporters. They could also be more specific accounts of their access to particular events. This kind of information contributes to our understanding of how the textual record was formed, since it provides background information on the transfer of information from historical events to witnesses, from witnesses to the original texts, and from the original texts to the versions of the texts that are available today. Textual claims about authors, their methods, and the preparation and subsequent treatment of texts, are very much like the middle-range theories used to account for the archaeological record. Just as the middle-range theories, mostly archaeological claims themselves and both general and specific in content, are used to describe the formation of the present material record and thereby give it meaning and credibility, so textual accounting claims within history can often be found and used to similar purpose. In both cases, and as before, there is epistemic value in using independent claims in this role of accounting. The middle-range theory must be independent of any theory for which the sponsored evidence is a test. This is to block the circularity of a theory accounting for its own evidence, a circularity that would make the testing vacuous. For the same reason, a textual source of information about some particular textual evidence must be independent of that evidence. The author of the accounting claims, for example, must not be a sycophant of the historian being described, but must have an independent source of credibility. This suggests that the structure of justification of historical knowledge will be one of coherence rather than foundational, since claims that function to account for evidence are themselves evidence in need of justification.
Three kinds of accounting claims for historical evidence have been described so far, beginning with the most external source of support and moving to a more internal. The fourth and final category is the most internal of all, that is, the use of information in textual evidence to account for itself, to aid its own interpretation and credibility. In this case we look to claims made by the author, for example Thucydides, rather than about the author, for indications of justification. An author might, for example, explicitly discuss his methods of observation and recordkeeping. He might mention his own sources and even details of their access to events in the narrative, showing care to distinguish eyewitness reports from hearsay. And there are less direct features of the writing that might also contribute to the case for credibility. Consistency in the account is at least a necessary feature of accuracy. Attention to detail, vivid writing, and explanatory coherence are all features of a text that can be evaluated on a careful reading and that might be used as indicators of accuracy and credibility. A balanced presentation of two sides to the issues might suggest that the author is objective in reporting events. But the ice is clearly getting thinner, and the epistemological question to press at this thoroughly internal level of accounting is whether any of these features of the text itself can reliably distinguish the truth from clever fiction. pp. 3f.
Whatever the means they use, historians still have to engage in the basic Rankean spadework of investigating the provenance of documents, of inquiring about the motives of those who wrote them, the circumstances in which they were written, and the ways in which they relate to other documents on the same subject. pp. 16f
I have addressed this topic several times now as some long-time readers will know: e.g. see similar comments by the historian Eric Hobsbawm and his reviewer:
- How modern historians use myths as historical sources – or, Can Hobsbawm recover the historical Robin Hood?
- Contrasting methods: “nonbiblical” historians vs “Jesus” historians
In principle, the method is no different from how we normally expect to verify information in our common experiences. Or at least it’s how we should seek to verify anything that is important enough financially, legally, and so forth.
Historians of ancient times don’t lower their standards because there is less surviving evidence than for modern times. Rather, they tailor the questions that they can reasonably expect the available evidence to answer. (Some of the distinctive problems arising with ancient sources are illustrated at The evidence of ancient historians.)
There are other principles of historical reasoning that I have not covered here. When and to what extent can different types of testimonies be trusted? Where does Bayesian reasoning fit in with all of this? I’ll return to some of these in future posts. It’s all about the basics. Doing historical research is really not like doing rocket science. The hard part is having the patience and time to sift through all the data available. And being alert to when arguments are and are not based on the bedrock principles set out above.
Day, Mark. The Philosophy of History: An Introduction. London ; New York: Bloomsbury Academic, 2008.
Evans, Richard J. In Defence of History. London: W. W. Norton, 1999.
Garraghan, Gilbert J. A Guide to Historical Method. Edited by Jean Delanglez. Fordham University Press, 1946.
Kosso, Peter. “Historical Evidence and Epistemic Justification: Thucydides as a Case Study.” History and Theory 32, no. 1 (1993): 1–13.
Kosso, Peter. Knowing the Past: Philosophical Issues of History and Archaeology. Amherst, NY: Humanity Books, 2001. p. 120
Marwick, Arthur. New Nature of History, The: Knowledge, Evidence, Language. Basingstoke, Hampshire: Palgrave, 2001. pp. 26f
Tucker, Aviezer. Our Knowledge of the Past: A Philosophy of Historiography. Reissue edition. Cambridge University Press, 2009. p. 121f
Latest posts by Neil Godfrey (see all)
- When God Created Humans, then Retired: Genesis 1 as Science — [Biblical Creation Accounts/Plato’s Timaeus – 5c] - 2022-11-19 08:02:27 GMT+0000
- The “Late” Origins of Judaism – The Archaeological Evidence - 2022-11-17 08:16:43 GMT+0000
- In Six Days: Genesis 1 as Science — [Biblical Creation Accounts/Plato’s Timaeus – 5b] - 2022-11-11 07:33:44 GMT+0000
If you enjoyed this post, please consider donating to Vridar. Thanks!