Friday, 12 June 2015

The Importance of Being Uncertain?

A little Tweet this morning inspired today's blogging:


You'd think that archaeological scientists would be willing to admit there is always uncertainty. However well we collect our data, however good the sampling strategy is (and it often isn't!), we are almost always dealing with a record that is complex and fragmentary. The best we can offer, to all but the most basic questions, is a range of possibilities. Multiple working hypotheses that we can continue to refine as techniques improve and more data becomes available. I think I read somewhere once that we should present a 'definitive story' of archaeology that can then be changed if needed. But I am not sure if this works - it can be hard to change an idea once it moves outside academia.

As usual I think my perspective as a geoarchaeologist, and a microarchaeology specialist, come into play here. Geoarchaeologists are especially aware of site formation processes and taphonomy, and the palimpsest nature of the archaeological record. We can get closer and closer to a 'truth' of sorts by narrowing down the multiple options, but I am not sure we can ever be 100% certain of anything. I was recently long-listed for a Fellowship where I was planning to look at some of these questions, the way that we collect data and interpret it, and the perspective that microarchaeology can provide. I didn't get it in the end, but the feedback was interesting. The committee liked the idea, but also commented that if we admit uncertainty to the public they take it for ignorance.

What bothers me most is when archaeological research is picked up by the media, or when it is pitched at the 'high impact' journals. In the case of 'ordinary' journals, I think most of the time the language used is more cautionary (though I have seen examples when an author moves from 'probabilities' in the discussion, to definitive statements in the conclusion!). However in journals like Nature and Science, and in press releases, basically anything high profile, the language begins to change. We see possibilities becoming absolutes, one possibility becomes the story, and makes its way into popular knowledge. I guess Nature and Science aren't going to publish 'could be the earliest evidence of animal domestication but actually it could be interpreted in a number of ways depending how you look at it'.

In many areas of research, it doesn't matter much. It's not like it's going to change anyone's life if the earliest domesticated animals were in Turkey, Syria, Iran, wherever. But I do wonder about some research. Studies of ancient migration, peopling of the Americas, anything with the potential to be used politically. These are topics that could have a significant impact on living communities. Surely it is important that we do not create definitive narratives? But as the paper by Crema states, there is little interest in defining and reporting uncertainties, in their case with chronometric data, but I think it applies to the discipline as a whole. Does it matter whether or not we tell our archaeological stories with certainty? Should I remove the 'potentially' and 'possibility' and 'one way of interpreting..' from my writing? I am genuinely interested in hearing from other archaeologists on this one!

No comments:

Post a Comment