Sorry for the extended hiatus; I’ve been busy with various things. I’ll have more on the Mississippians at some point, but for now I want to discuss a more general issue: the relationship of historical linguistics to archaeology in attempting to reconstruct past events. Both disciplines provide ways to study past events beyond the reach of historical scholarship in the traditional sense, which is based on written documentation. (I would argue that in many cases oral tradition provides an additional line of evidence, similar to written history in many ways, useful for understanding the prehistoric past, but that is a controversial position and I’m not going to defend it in detail right now.) What makes linguistics and archaeology particularly powerful is that they are independent lines of evidence, which means that tentative conclusions drawn from one can be compared to the evidence from the other to see how well they coincide. This provides a much more robust and reliable reconstruction of past events than would be possible based on either line of evidence alone.
And yet, despite the potential interpretive power to be gained from using linguistics and archaeology in tandem, this integration is rare, and both linguists and archaeologists have a tendency to ignore each other’s work most of the time and, when they do acknowledge it, to use it in a very uncritical and superficial manner that doesn’t come close to unlocking the full power of real integration. Back in 1976 the journal World Archaeology published a theme issue with several papers exploring the potential for integrating linguistics and archaeology, and it’s a sign of how little this integration has progressed that some of these papers are still useful summaries of the issues and the state of research.
The introductory paper in the issue, and the one I will primarily focus on in this post, is by Christopher Ehret of UCLA, and it provides a general overview of the types of historical inferences that can come from historical linguistic research. He notes that most research up to that time had been oriented toward just one of these types: those inferences that can come from evidence of genetic relationship between languages, which is to say, the knowledge that a given group of languages descends from a single “proto-language” presumably spoken by a single socio-cultural group at some point in the past. These are certainly useful, in a general way, but there are real limits to how much can be learned just from knowing which languages are related in a given region. Looking at the reconstructible vocabulary of the proto-language can give some important information about the culture that spoke that language that can, at least in theory, be correlated with archaeological evidence to pinpoint which archaeological “culture” corresponds to the speakers of that proto-language. Two other papers in this issue address different aspects of this kind of research in different context and with different language families: Robert Blust’s paper on Proto-Austronesian shows how information gained in this way can supplement the archaeological record by providing evidence for the presence of certain items of material culture and social institutions that are not recoverable by archaeology because of their perishable or intangible nature, while J. P. Mallory’s survey of research along these lines on Proto-Indo-European mostly points out the difficulty of attributing cultural items for which there are reconstructible words to a single culture when the proto-language being reconstructed may not actually represent a single language or its speech community.
Moving beyond these issues, however, Ehret points out that there is more to historical linguistics than determining which languages are related and what words can be reconstructed for various proto-languages. A potentially much more productive line of evidence for culture history, and yet one that has seen remarkably little research, is loanword studies. Languages may adopt words from other languages for a variety of reasons, many of them quite important for understanding political and economic relationships between societies at various points in the past. Furthermore, loanwords are often (though not always) easy to identify either in currently spoken languages or in the reconstructed proto-languages from which they descend. Ehret gives various examples from his own research in Africa to illustrate how loanword studies can give substantial insights into cultural relationships and history.
It’s important to note, however, that while most of what Ehret says in this paper about general methodology is unexceptionable, his own conclusions about African prehistory are quite controversial and one of his faculty pages at UCLA even has this remarkable paragraph:
His linguistic works include A Comparative Reconstruction of Proto-Nilo-Saharan (2002), Reconstructing Proto-Afroasiatic (1995), and The Historical Reconstruction of Southern Cushitic Phonology and Vocabulary (1980). He has also written monographic articles on Bantu subclassification, on internal reconstruction in Semitic, on the reconstruction of proto-Cushitic and proto-Eastern Cushitic, and, with Mohamed Nuuh Ali, on the classification of the Soomaali languages. These reconstructions have not been well received, and are not followed by other linguists.
Evidence for the controversy engendered by Ehret’s interpretations comes in the very same special issue for which he wrote this introductory paper. One of the other papers, by the archaeologist D. W. Phillipson, addresses the Bantu language family and the potential for both linguistics and archaeology to shed light on the issue of when and how Bantu-speakers spread across much of southern and eastern Africa. Phillipson notes Ehret’s interpretations but disputes them in detail on various points. This is obviously a one-sided account, but Phillipson’s arguments seem pretty strong to me. I don’t know much about this issue, of course, and it’s quite likely that research has progressed a bit since 1976 in any case, so I’m not going to draw any conclusions about who was more right.
Africa is actually an interesting case here because it seems that historical linguistics has played a much bigger role here than elsewhere in developing hypotheses about prehistory (probably due at least in part to Ehret’s work). This is in contrast to the Americas, where linguistics and archaeology have mostly operated separately and the latter has been more dominant in developing historical hypotheses. I think the African model offers a potentially productive route for Americanists to take in trying to come up with more detailed reconstructions of culture history, although the many controversies over the proper interpretation of African prehistory show that this more integrated approach is by no means a cure-all.
I’ll have more on the potential implications of all this for North America later. For now I just want to introduce the topic.
Blust, R. (1976). Austronesian culture history: Some linguistic inferences and their relations to the archaeological record
World Archaeology, 8 (1), 19-43 DOI: 10.1080/00438243.1976.9979650
Ehret, C. (1976). Linguistic evidence and its correlation with archaeology World Archaeology, 8 (1), 5-18 DOI: 10.1080/00438243.1976.9979649
Mallory, J. (1976). Time perspective and proto‐indo‐European culture World Archaeology, 8 (1), 44-56 DOI: 10.1080/00438243.1976.9979651
Phillipson, D. (1976). Archaeology and Bantu linguistics World Archaeology, 8 (1), 65-82 DOI: 10.1080/00438243.1976.9979653