Ancient History: A How-To Guide

By Owain Williams

In today’s blog, I want to discuss the ever-exciting topic of historical methodology. Or, more precisely, how the study of history works. This, of course, is not going to be a masterclass – this is a blog, after all – but I hope it will be enlightening, nonetheless.

Online, I regularly see people asking for the ‘truth’ about this or that event or cultural facet. The problem is, though, there is no such thing as objective truth in most cases of historical study. Without a time machine, it is impossible for us to know what precisely happened on a certain day at a certain time. This is even more so for ancient history. Instead, historical study deals with degrees of informed probability.

Before we get into what I mean by ‘informed probability’, I want to go over why, precisely, we cannot know anything with absolute certainty. Ultimately, we rely on unreliable sources. Each and every literary source has biases. Some biases are more hidden than others, and occasionally it can seem like there are no biases at all. The historian’s job is to examine and critique everything about these sources. People will probably have their own names and acronyms to help teach such source criticism, but it ultimately boils down to the five Ws:

• What is the source?

• Who created the source?

• When was the source created?

• Where was the source created?

• Why was the source created?

• A bonus question is: Who was the source made for?

Now, answering these questions is no easy feat, as it involves comparing our source to other sources, about which we must also ask these questions. However, keeping these questions in mind is a very helpful way to determine how reliable a source likely is. 

Once you have asked these questions, the next step is to examine the source from the perspective of informed probability. As the term suggests, this effectively means examining the context of an event from the source in question to determine the probability that the event occurred. I have termed this stage ‘informed probability’, rather than ‘historically informed probability’, because the material with which you compare an event does not necessarily have to be a historical source. It could be an ethnographic source, for example. A famous instance of this is Han van Wees’ comparison of Homeric warfare to warfare in the highlands of Papua New Guinea.

Domitian’s ‘black banquet’ is a useful case study for demonstrating how this methodology works. For those who don’t know, the ‘black banquet’ is an infamous event from Domitian’s reign where the emperor hosted a dinner for several dinner guests in which everything, from the walls to the food, was black, and he only discussed matters of death. While it is quite a strange occurrence, it is certainly not the most outlandish story about a Roman emperor. However, only Cassius Dio, a Roman Senator and historian active in the late second and early third centuries AD, records this event.

This should, immediately, make us pause. After all, there are a lot of surviving primary sources (as an aside: it is debatable to what extent we should consider Cassius Dio a primary source for Domitian’s reign, over 100 years before his time) from Domitian’s reign, such as Martial and Statius, none of which make any mention of such an event. Nor do any of the sources written in the immediate aftermath of Domitian’s reign, such as Tacitus or Pliny the Younger, even though they were overtly hostile towards Domitian. Given this lack of corresponding evidence, it is unlikely Cassius Dio was recording an actual event. However, there is a precedent for such a morbid event. Seneca the Younger notes how one Pacuvius would hold mock funerals for himself after his dinner every day (Letters 12.8–9)! With this in mind, Domitian’s dinner does not seem impossible. However, given that no other sources remark on this banquet, it is likely that Cassius Dio exaggerated its significance (and many of the details). It may have gone unremarked in other sources because it was unremarkable. Domitian may have hosted such a death-orientated dinner, but it may have been more philosophically motivated than driven by the desire to mock Senators, as Cassius Dio suggests.

This isn’t even touching on our own biases and those of the historians we are following. The same process – the five Ws and informed context – can be applied to modern scholarship. With modern scholarship, however, it is somewhat easier, as historians should lay out their own sources, effectively cataloguing how they came up with their conclusions.

This is, in a very simplified form, what the study of history is, regardless of what period or place is being studied. A lot of work goes into determining whether we should or should not, and to what extent we can, trust our sources, and our perspective is constantly being revised based on different authors’ own interpretations of the evidence and new discoveries.