How do we know what we know about antiquity? That’s not a trick question or an invitation to philosophize; it’s actually a key issue of methodology. You can’t add to what we know, or even use and cite what we know, until you know what we know and how (or why) we know it. That’s a long way of saying that this week’s post is about knowledge creation in classics at various levels of study.
In some fields, the question of how knowledge is created has a pretty clear answer. We’ve all seen TV shows or movies that show the scientific method, for example: scientist hangs out in lab, tests hypotheses, and ultimately can create a theory based on multiple attempts to do a similar task with similar results. (Of course, Hollywood’s version of science isn’t the same as real-life science — but the process is at least sort of similar.) The Hollywood version of humanities research, which is
pretty rare anyway apparently not as uncommon as I thought, also tends to focus on the discovery of the new: new manuscripts, new artifacts, new sites. For some classicists (numismatists, epigraphers, papyrologists, and archaeologists), pursuing novelty may be the norm. But for many, it’s not: we work with old texts, many of which have been studied for centuries, some of which (like Homer and Vergil) have been analyzed since antiquity.
Instead of finding new texts, we find new ways to understand (or interpret) those texts. Sometimes that means applying new models or theories (I’ll call it method 1); sometimes, comparing new groupings of evidence (method 2); sometimes, simply looking at the same material in a new way (method 3). Because the first two methods are more straightforward, I’m going to start by explaining them.
Applying a new model or theory means taking another person’s argument (usually) from its context and seeing how well your evidence fits. I should add that I’m using theory to describe abstract arguments, while models have a concrete aspect and/or application. Freud’s theory of psychology centers on sexual repression. When a son’s repressed desire for his mother controls his psychological state, you get the model of the Oedipus Complex — a model that can then be applied to real or fictional men.
When you apply a new model or theory to your evidence, your main concern is about fit. You can think of this application process as sort of similar to trying on clothes. Sure, those pants look great on the mannequin — but the important decision for you is how well they fit your body. Whatever source material you have forms your body of evidence. Your job is to see if the existing argument fits. And like pants, if it mostly (but not entirely) fits, there’s some room for making adjustments. But if you make too many adjustments, you’ve basically abandoned your model (and then you might be into method 3 territory).
A good example of method 1 can be found in the works of Georges Dumézil and his followers. Dumézil postulated that Indo-European societies recognized three male roles: priest, warrior, and farmer. Myths center around this triad of ‘functions’, because gods often specialize in one of them (in Greek myth, for example, Ares is obviously a warrior). Dumézil claimed to find traces of this social structure in ancient evidence from India, Germany, Greece, and (sort of) Rome. But Rome presented a tough fit. Although there was an ancient divine triad, it proved very difficult to separate the gods into the three distinct functions (interested readers can check out Arnaldo Momigliano’s summary of the problems, as well as a recent defense of the theory). The difficult fit could be compared to pants that fit in the legs, but not the waist: they might be useful in some ways, but you could also find something better.
An important question you should consider about your chosen model/theory is why you’ve chosen it. There are many good reasons to test classical evidence against hypotheses from other fields. But it’s generally not a good idea to test your evidence for the sake of doing it, even if the fit is relatively good. You should aim to discover something new about antiquity in the process. In that sense, even a very well-fitting model could be a less-than-ideal choice: if all you learn is that Rome is patriarchal, well, we know that already.
Method 2 is similar to the ‘model’ part of method 1. In method 2, you compare two (or more) similar bodies of evidence to gain insights from their points of intersection and differences. This evidence can be from the same time and/0r place, from the same broad ‘culture‘, or from very different societies. (NB: I apologize for the completely random nature of the books I chose as examples. They represent the 1-2 works I immediately thought of for each heading, and are not meant to imply that certain topics are more suitable for certain types of analysis than others.)
Comparative studies, rather than centering on fit, are more like puzzles. The parts that don’t fit are just as helpful as the parts that do fit. A single society isn’t exactly the same over long spans of time, and societies that separated in time and space aren’t culturally identical. By finding the places that don’t intersect as well as the places that do, you are helping to determine what makes each culture unique.
As was true of theories and models, it’s a good idea to have both a reason for choosing your specific comparanda and an insight that can be drawn from your overall study. For example, a very common topic in the media (and, apparently, American classrooms) is the comparative politics and history of the United States and Rome. This can be done well, but it’s hard: you need to be able to come up with something useful to say about at least one of those two cultures. “Rome fell, so America is doomed” may sell copy, but it’s hardly novel — just ask Herodotus.
The last group, method 3, is sort of a grab bag category. That’s partly due to space and partly by design. Generally speaking, I think the first two methods can be done well by researchers at any level of study (adjusting for expertise): that is to say, an undergraduate is just as capable of comparative analysis as a senior professor, once you account for the fact that we expect the professor to have more experience. The types of analysis I’ve grouped under method 3 are not for beginners. They require at least some knowledge of the field. But they are also useful ways to start thinking about how research comes together in classics.
This group includes many different ways of developing new knowledge based on interior knowledge of the discipline (primary sources and scholarship). For example, you could scan the secondary sources to find gaps in the literature, which means you are looking for questions that you have about the material that no one has asked before (or, often in classics, no one has asked for a really long time). A similar method is locating problems in the literature, which is finding a phenomenon or piece of evidence that is not adequately explained by existing theories and models.
You could also try a thought experiment, which often involves inverting the existing idea(s) on the topic — in historical terms, essentially using a counterfactual narrative to better understand the evidence. As an example, what would have happened if Caesar hadn’t crossed the Rubicon? Maybe he would have been tried; maybe he would have been assassinated earlier; maybe he would have turned back to become king of Gaul. Gathering a list of these possibilities can help better understand the actual decision.
Cataloguing, collecting, and synthesizing information is another way to create new knowledge. The key here increasingly lies in synthesis, or coming up with ways to understand your collection; digital tools make collection of information much easier.
A more difficult (and less predictable) way to come up with new knowledge is the eureka moment — a time when you accidentally stumble upon an idea that turns out to be a viable avenue of research. Although this sounds unlikely, it does happen (I suspect more often than people let on). The initial spark of an idea may lead you towards identifying a problem in the literature, a comparative solution, etc., but I’ve included it as a separate method for generating knowledge because it’s not the same as setting out to read a vast body of literature and identify gaps, or writing a paper on “what if the Trojans had won the war?” Random insights can’t be counted on, obviously, but they shouldn’t be discounted, either: my first article came from the incidental association of passages from Diodorus and the Old Testament. I’ve always thought that reading widely (or deeply) was a good way to foster these sorts of associations, but obviously there’s no guarantee.
The final question of, of course, what to do with these methods now? If you’re considering a research project for the summer, especially if it’s your first major research project, you may find that one of these methods is more helpful or more interesting than others. Even if you’re not, you may want to refer back to it in the fall. For our readers in other fields (and I know there are a few of you!), I hope this helps give you an idea of where classical knowledge comes from and how new knowledge is produced. Please let me know in the comments if you have questions, and I’ll try to answer them!