I’m going to start trying again to write comments on the reading I’ve been doing over the last six months. It hasn’t been quite one-a-day, but there’s a lot of books and articles in my backlog to talk about.
Pathfinders is like other works of global history by Fernandez-Armesto: a readable, pleasant synthesis that doesn’t add much to a historian’s analytic toolkit, but it puts narratives and information that have often been told in a markedly Eurocentric way into a broader comparative perspective.
I’m going to focus on one specific thing I noticed in this book that I think speaks to a wider problem in the writing of global histories of this kind. Fernandez-Armesto works very hard to draw in examples and cases from most regions of the world, particularly when he’s talking about premodern exploration. When he gets to 1500, he serves up a modest amount of Iberiocentrism, but he’s honest about that, and rather charming. Plus it’s hard to argue against the centrality of Spain and Portugal in maritime exploration from 1500 to 1650.
However, this drive to globalize history often draws world historians into a complicated tension with area studies specialists. I’ve been very clear about my dissatisfaction with the tendency towards intellectual parochalism among Africanists, but some of that tendency is rooted in some genuinely important priorities.
Here’s an example drawn from Pathfinders. Early in the book, Fernandez-Armesto is talking about early cartographic or navigational practices in human societies, and reasonably concludes that there must have been some practices employed in some early premodern societies beyond dumb luck. Absolutely: it’s completely fair to infer that premodern human societies may have had all variety of interesting mnemonic and representational techniques for remembering or communicating how to go from here to there, and some of these may be of profound antiquity.
In this discussion, he uses a few African examples that unfortunately illustrate the intellectual predicament of Africanist historians.The way he uses Africa is as many world or comparative historians do: by citing recently observed practices that suggest what was likely done in antiquity by all humans. This is the conventional logic of using 19th or early 20th Century Africa as if it were a window into prehistory, unchanged tradition. The examples Fernandez-Armesto uses specifically in this case are Marcel Griaule’s work on Dogon cosmology with its map-like constructions, MacGaffey’s work on Kongo cosmographs, and scholarship on 20th Century Luba chiefship ceremonies which include geographical knowledge of sacred and ritual sites.
The problem is that all of these examples, Africanist scholars know, are anything but unchanged windows into the distant human past: they’re practices and ideas that have almost certainly changed considerably over time AND they were described by and in some cases actively reinvented by Western officials and scholars with very bounded ideas and preconceptions about African history and society. There’s a considerable mini-literature on Griaule and his intellectual history that makes it impossible to take anything he wrote or said about Dogon cosmology at face value. MacGaffey has noted that cosmographs as he describes them strike him as both recent and highly dynamic, changeable inventions.
These examples tell you about the distant human past about as much as saying, “Contemporary Americans often give directions to important locations by citing commercial landmarks rather than formal maps, e.g., ‘Take a left at the Kentucky Fried Chicken, if you pass the Wal-Mart you’ve gone too far'”. That actually does tell you in a universal, general sense about how people might have navigated in early societies, but it doesn’t tell you any specifics. Fernandez-Armesto regards recent African examples as directly suggestive of practice in antiquity (e.g., not just general illustrations, but “Maybe here’s how the Dogon did it back then” or “The Luba probably valued this kind of geographical knowledge back in prehistory and remembered it through chiefly investiture and oral tradition.”) Africanists know that there weren’t any Luba as such in the early premodern period that Fernandez-Armesto is working with early in his book, any more than there were suburban Long Islanders with lawns and backyard swing sets in Neolithic times.
But when the Africanist gets all snippy about this problem, insisting in almost cliched terms that African societies were also dynamic and historical and changing, the world historian or comparativist says eagerly, “Great, so tell me about what West African societies were like around 200 AD or so, especially any details on their cartographic or geographical practices, so I can give some examples that compare to Rome, China, Arabia, and so on”. And here the Africanist mostly has to say, “Sorry, don’t know. Pretty much can’t know, not at that level of specificity.” Or even more sheepishly, “Well, I can tell you what a handful of Mediterranean, Middle Eastern and South Asian sources say about those places, but a lot of that is little more than thrice-told stories from merchants.” Plus the Africanist can make use of some archaeology and linguistics, but as Fernandez-Armesto points out, there are a great many imaginable geographic or cartographic practices in prehistory or antiquity which may have left no material or artifactual record behind.
So now the world historian has to ask, “So, how about it, guys? Should I just leave Africa out of what I’m talking about unless I have highly specific, properly historicized examples, which means it’s going to be left out of a lot of my account, or should I use whatever I can find, and without a lot of methodological song-and-dance each time I do so, because that’s going to mess with the readability and coherence of my synthesis”.
Africanists often seem to reply, “Either one, you choose, and we’ll complain about it either way”. The problem is that the complaint in either case has some validity to it, but the global historian’s choice either way is also fairly valid. I don’t have an easy answer. On some subjects, there may well be a chronologically appropriate comparison for a global historian to use that comes from Africa or Mesoamerica or Oceania. On many others, the details and specificity in premodern world history are going to come from literate, record-keeping societies even when we can be fairly certain that there were examples of the same practices or institutions or ideas in non-literate societies elsewhere in the world.
I really liked this post – you bring up some points that are important to keep in mind. For many modern/European/American historians, it’s easy to suffer from a kind of embarrassment of riches and fall into the trap of taking sources for granted. In many cases, abundant source material allows for a particular methodological approach that in turn influences an even broader approach to historical scholarship. While this may be harmless or even beneficial for regional/area studies, it can (as you’ve highlighted) become problematic with any kind of wider global analysis.
The premodern Africa chapters of every world history textbook I’ve ever used contain this problem (so do the Latin American ones, to a lesser extent) and the vast majority of them follow Fernandez-Armesto’s lead and project backwards. I even used one which had a recent color photograph of a tribal dance ritual which they used in the pre-800 c.e. chapter!
I like to use those chapters to talk about the historiographical problems, and how some of them are being solved, but also about the history of anthropology (which started from the “Africa-as-preserved-antiquity” position) and the problems of overweening theory. Then I tell them that the Africa chaptes will bet better in a few years…..
Historical linguistics and, when used carefully, oral traditions can get at this type of history. Tim, I’d be interested in your take on Jan Shetler’s book, Imagining Serengeti. Her book is, I think, a very carefully exploration of early history, rooted in what she can glean from linguistics and oral traditions. Turning to Shetler, Schoenbrun, Klieman and others who use historical linguistics may be one way out of the dilemma of Africa and world history. Chris Ehret has taken a stab at placing Africa in world historical context–The Civilizations of Africa–which provides another model.
For me this breaks down into priorities about what I want the students to learn. If I want them to learn about the dynamic diversity of African history, then the book needs to be written that way. So it would be important to talk a bit about how we know that recent cultures are not simply artifacts of antiquity.
If I want them to know about how we historians try and sometimes fail to construct robust knowledge based on spotty and often-problematic sources, these would be excellent case studies.
And if I want them to have a smooth fantasy narrative that makes us all feel good about including everybody everywhen (in a semester or two), I can make up any old just-so shit about mapmaking or camel’s humps just like Kipling. ;-p
I teach three sections of required intro World History every semester to mostly business majors. The kids can handle some historiography, which is an instance of a more valuable cognitive skill than absorbing trivial pursuit answers about where maps come from, and gets Africa back in where highly specific, properly historicized examples are scarce.
So this book doesn’t sound so good to me.