On the reliability of medieval sources

The debate over the reliability of a primary source is central to its interpretation, and thus to the liberty of the historian to follow or diverge from it.  Primary sources whose reliability is generally accepted are held up as approximating historical truth and narratives can be criticized for ignoring them or disagreeing with them.  On the other hand, primary sources whose reliability is often seen as poor can be safely left out of the story and even attacked when used to support a theory.

A chronicler's reliability is often strengthened by his claims to be an eyewitness to events, or by his inclusion of a level of detail which suggests to the historian that he was an eyewitness or heard it from eyewitnesses.  This assumption, however, may conflate the level of detail a chronicler included in the text with the level of detail a chronicler had available to him.  Another factor increasing estimated reliability is if the chronicler recounts events likely to be well-known to his audience or events involving famous persons to whom he may be accountable were he to misrepresent.  However, this assumption may ignore the propagandist nature of some sources, which seek to retell known events in a light different from that known to its audience, or in which famous persons wish events to be misrepresented.

Another estimate of reliability may be gained by attempting to interpret the bias of the chronicler and the intended rhetorical effect of the narrative in the text.  Reduced reliability can be attributed where the chronicler appears to be relating events in an attempt to convince his audience, but greater reliability can be attributed where the chronicler reveals information tangential to his narrative or despite his intended persuasive effort.  For example, the bloodthirstiness of the legate Arnau Amalric, the Abbot of Citeaux, can be determined with some reliability from his own letter to Pope Innocent III after the sack of Beziers: "To our wonderment, within the space of two or three hours they surmounted ditches and walls and the city of Beziers was captured and these ribalds of ours spared no order of persons (whatever their rank, sex, or age) and put to the sword almost twenty thousand people.  After this great slaughter the whole city was despoiled and burnt, as divine vengeance raged marvellously" (translation from Pegg's A Most Holy War).  In this case, it is not Arnau Amalric's intention to depict himself in a bloodthirsty light -- he does so in passing while fulfilling his intention of assuring the Pope that the Crusade was accomplishing God's work.  If a critic of the crusade, such as the Anonymous, had depicted Arnau Amalric boasting of the number of women and children slaughtered by an army under his command, we might be more hesitant as to his reliability.

But what do we mean by reliability?  At first glance, reliability would be seen as congruence between the events as related by the source and what we know of those events independently.  Where sources agree, then, they are seen to reinforce each other's reliability.  Where they disagree, the reliability of at least one of them will be impugned.  But comparison between sources is not the only way we can know of events.  There is also our rational knowledge of what could and could not have happened in the past.  Yet I find that this basis for the evaluation of a source's reliability is almost never used.

First, an exception.  Medieval chroniclers were, we can now acknowledge, very poor at numerical estimation.  When describing the sizes of armies or the populations of cities, they count more men than we now deem possible and seem, in fact, to be off by a very large margin.  The phenomenon is commented upon by almost every historian who repeats the sources' claims about the size of the crusading army, the number of victims of the sack of Beziers (above), and so on.  Costen, in The Cathars and the Albigensian Crusade, called it "the normal inability of commentators at this period to deal with large numbers", and the problem is so pervasive that while it is acknowledged as a general reliability issue, it is never seen as a problem with the reliability of an individual source.

Another, and perhaps more important, way in which we can consider the reliability of a source should be when the source recounts events which we know could not have occurred as described.  Many historians have the alarming habit of quietly editing out the impossible sections of a tale and then presenting the plausible remainder without further skepticism or comment.  Take the following infamous anecdote, retold by Zoe Oldenbourg in Massacre at Montsegur:

"The case of the young girl from Rheims is very typical in the light it sheds on the mentality of the Cathars' opponents.  Radulph, Abbot of Coggeshall, relates that one day the Archbishop of Rheims was taking a stroll outside the city accompanied by some of his clergy; and that one of these, Gervais Tilbury, noticing a young girl walking on her own through a nearby vineyard, went up and accosted her with amorous intent ('although,' as Radulph says, 'he was a Canon').  His proposals must have been blunt and direct in the extreme, since the girl, 'with modest and solemn mien, scarce daring to look at him', replied that she could not give herself to him; for, said she, 'if I were to lose my virginity, my body would be corrupted on the instant, and I should be damned irremediably for all eternity'.  From these utterances the holy clerk perceived that he had to do with a heretic, and denounced her as such to the Archbishop, who meanwhile had come up with his suite.  The girl herself, together with the woman who had instructed her in the Cathar faith, was condemned to the stake, and died with a courage that won great admiration from those who witnessed her end.  It is hard to know which element in this story is more surprising: the heroism of the anonymous martyr, or the moral callousness displayed by judges and chronicler alike.  It seemed quite natural to them that a cleric should not only try to seduce a young girl, but also that he should utilize the very fact of his shameless conduct as an argument against his victim.  A Church in which such moral decadence flourished was hardly qualified to cast the first stone against anyone else." (pp. 65-66)

If we could not know anything more about this tale than what Oldenbourg tells us, it would appear to be of good reliability.  It is not Radulph of Coggeshall's intention to demonstrate the "moral decadence" of the Church -- he seems to do so in passing while pursuing another rhetorical intention.  The individuals involved, such as the Archbishop of Rheims, would presumably be able to hold Radulph to account should he largely misrepresent their conduct.  The level of detail involved is considerable and, although Radulph was not likely an eyewitness to events, he wrote his chronicle starting in 1187 and the burning occurred only about ten years previously.  The author of his Wikipedia entry says of his chronicle that "the corrections and erasures of the autograph show that he took pains to verify his details; and his informants are sometimes worthy of exceptional confidence."  In addition, the burning is also mentioned by another source, thereby confirming our chronicler's apparent reliability -- at least from the information presented so far.

The problem is that Zoe Oldenbourg has been quite selective about the portion of Radulph's story which she relates.  Consider this account of the same from R. I. Moore's The War on Heresy:

"It [Ralph of Coggeshall's famous story] tells how Gervase of Tilbury, an English clerk in the service of the archbishop of Reims, was attracted by a young girl whom he saw working alone in a vineyard.  When she declined his amorous advances, pleading that the loss of her virginity would bring her to certain damnation, 'Master Gervase realised at once that she belonged to the blasphemous sect of the Publicani, who were being searched out and destroyed all over France.'  The girl was arrested and taken to the archbishop's palace for questioning.  It transpired that she had an instructress in the city, who, she was confident, would be able to answer the arguments that were being advanced against her beliefs.  Found and brought before the court,

the woman was bombarded by the archbishop and his clerks with questions and citations of the holy scriptures to convince her of the greatness of her errors, but she perverted all the authorities which they brought forward with such subtle interpretations that it was obvious to everybody that the spirit of all error spoke through her mouth.

The two women, refusing to recant their errors, were condemned to the stake, but the elder escaped:

When the fire had been lit in the city, and they should have been dragged by the archbishop's servants to the punishment that had been allotted to them, the wicked mistress of error called out, 'Madmen!  Unjust judges!  Do you think that you can burn me on your fire?  I neither respect your judgement nor fear the fire which you have prepared.'  So saying she took a ball of thread from her breast, and threw it through the great window, keeping one end of the thread in her hand, and calling loudly in everyone's hearing, 'Catch!'  At this she was raised from the ground in front of everyone, and flew through the window after the ball of thread.  We believe that she was taken by the same evil spirits who once lifted Simon Magus into the air, and none of the onlookers could ever discover what became of the old witch, or whither she was taken.

The girl, who had not yet achieved such madness in the sect, remained behind.  No reason, no promise of wealth, could persuade her to give up her obstinacy, and she was burned.  Many admired the way in which she let forth no sighs, no wailing, and bore the torment of the flames firmly and eagerly, like the martyrs of Christ who (for such a different reason!) were once slain by the pagans for the sake of the Christian religion." (pp. 3-5)

As Moore then asks, "What is the relation between what really happened and what the sources tell us?  It is easy to accept the burnings and dismiss the ball of thread ...."  Perhaps it is too easy.  That is what Oldenbourg has done, lamenting the "martyrdom" of the young woman as fact while obscuring the escape of the older woman, saying only that she was also condemned.  But when Oldenbourg rhetorically states that "it is hard to know which element in this story is more surprising", this is not the case.  Clearly the flying woman with the ball of string is the really surprising element.  The rest surprises us so little as to seem somewhat plausible.

Now Radulph or Ralph of Coggeshall had an explanation for this surprising element, namely "evil spirits", which he presumed would provide an adequate explanation for his audience.  As modern historians, however, we must find this explanation inadequate and so conclude that nothing even remotely like the event described did, in fact, occur.  Nor can any of the hypothetical attestations to the chronicler's reliability prevail upon us in this situation.  In the case of the ball of thread escape, we know that the chronicler is completely unreliable. 

The question then, is whether we can "accept the burnings and dismiss the ball of thread" and if it is appropriate to edit out the ball of thread from the narrative and to present just the burning as coming from a reliable source.  We can not know if the young woman actually was obstinate in her error and eager for the flames, nor if she was actually given any "promise of wealth".  What large financial incentive could, in fact, credibly be offered to a convict sentenced to be burned alive?  Would we not be rather more surprised to hear that a formerly obstinate young heretic became wealthy by accepting an offer to recant and avoid the stake?  The notion seems so preposterous as to cast doubt on whether any such offer could be made or considered.  The level of distortion between any actual event and the chronicler's retelling of it is so large that we can no longer distinguish fact from fiction unless the fiction becomes so fantastic as to no longer be remotely plausible.

Some historians will argue that medieval authors were ill-suited to judge the credibility of miraculous stories, such as the escape of the witch, above.  To a certain extent, this is no doubt true.  But chroniclers then and now were capable of being reliable or unreliable; of making an effort to discern truth from rumour, or of choosing to uncritically pass on dubious information.  There are accounts which do not challenge our credulity in any places, and there are those which pass seamlessly from obvious falsehood to stories of uncertain truth.  I think it is a worthwhile, but too seldom performed, exercise, to distinguish between these types of accounts.

In future posts, I will compare our primary sources for the Albigensian crusade through this lens, and note the differences, especially between Peter of Les Vaux-de-Cernay and the Anonymous.