Tag: Radiocarbon

Still Ranting About Dating!

Kinloch, Rum excavations
Excavation taking place at Kinloch, Rum in 1986. This was our third season of excavation and we understood that the site was Mesolithic by then. At the time it overturned conventional wisdom about the early settlement of the north of Scotland – it proved to be quite a heavy burden!

Still ranting!

Bayesian dating is a new, and useful tool in the archaeological box of tricks. And like any new toy – we are keen to make use of it. But, it is a specialised piece of kit and we need to be careful that we know exactly what we are doing. I’m amazed by the number of times people say: ‘but why did you not use Bayesian dating’. In most cases the answer would be that it is not appropriate for poorly contexted samples on a site with shallow stratification. But I’m too polite to say that. Too many archaeologists seem to consider Bayesian dating as a simple means of getting more accurate dates.

Even when undertaking the old-fashioned application of radiocarbon assay it was necessary to remember to constrain our data carefully so that we understood the context of the sample we were dating and therefore exactly what the date might relate to. Is the pit-fill contemporary with the digging of the pit? Is the deposit on a house-floor related to the use of the house, or is it some sort of dumping after the house had gone out of use? Is the wood heartwood from an aged tree? Has the wood (or bone for that matter) been curated over a period of years? So many things to take into account.

However we use radiocarbon assay, and whatever method we use, we should never use it unquestioningly.

I know I come over as obsessed with all of this. But I learnt of the dangers of type-fossils and pre-conceived ideas early on in my career as an archaeologist.

When I studied archaeology we were taught that there was no Mesolithic occupation to speak of in highland Scotland. Not long after graduating I had the opportunity to excavate a nice lithic scatter on the island of Rum. The site was defined by the occurrence of a barbed and tanged arrowhead among the lithics, so we drew up a research design to excavate a site that was, clearly (according to the type fossil), Bronze Age. As we sieved our test pits and sorted the residue a lot of flaked stone occurred. It included a lot of small pieces with tiny retouch – microliths. But I knew they should not be there because microliths were Mesolithic and did not occur in the Bronze Age. And I had learnt at university not to expect Mesolithic occupation that far north. Even when I filled in the radiocarbon forms to send our precious carbon samples off for assay I noted that the predicted date range lay within Bronze Age parameters. I remember that well.

Imagine my surprise therefore when the dates came back: a nice little sequence around 8500 years ago. This was long before the Bronze Age. Indeed, it sat within the Mesolithic but was early even for Mesolithic Scotland and unknown that far north. I can remember thinking that if I didn’t know there was no Mesolithic in the area I’d think the site was Mesolithic. Of course, the penny eventually dropped, we continued to excavate a fabulous Mesolithic site (you can read the report for free here), and many other people have since examined other Mesolithic sites around there and even to the north.

But I had another surprise. I thought that my archaeological colleagues would be interested (even pleased) in my findings and so I phoned and wrote to people to tell them about the site and the dates. Curiously, people were divided: some wrote back with hearty congratulations; others were more cautious, telling me that I must have contaminated the samples, that it was just not possible for Mesolithic hunters to have penetrated that for north by that date, and so on. It was, I discovered, hard for some people to give up their fixed ideas of the order of things.

Change is always difficult, but I think we need to be cautious of thinking that we have worked out everything there is to know about the past. We need to be open to new interpretations and ready to rewrite our narratives. We need to remember that it is rarely the type fossils that are wrong, but rather our understanding.

So, the moral of my story is – to keep an open mind. To be open to the unexpected. And to be careful not to set the past into stone. The frameworks that we build to understand the people of the past need constant tweaking. That is what keeps us on our toes. That is what keeps it interesting.

And this is my last post on dating for a while – I promise!

Dating Problems

it can be easier to envisage time in a spiral
Time can be difficult for archaeologists to get a handle on. This depiction was drawn by Rhona Jenkins for my book: Fear of Farming.

I have a confession to make.

I hate dating…

To start with there are so many different types of date.

Take radiocarbon for example. I’m assuming you know how radiocarbon works (if you don’t there is a good explanation here). A raw radiocarbon date is a measurement of how much radiocarbon is left in an organic object, from which we calculate how long ago the object died. When you cite radiocarbon dates you can cite them as radiocarbon years: uncalibrated BP. But most people prefer to adjust them so that they equate to calendar years: calibrated BP; this is done by calibrating the radiocarbon date (which you will see referred to as a date, a determination, an assessment, an age or an assay), against organic material of known age (read about it here). A calibrated date is rarely a single year; it is more usually expressed as a range of years together with a probability in order to indicate the mathematical likelihood that the date falls within that range.

You have to know here that BP stands for Before Present. Except that it doesn’t – for uncalibrated dates it really stands for Before 1950, because that year has been accepted as a norm against which the decay of radioactive isotopes is measured. At the moment the little matter of 66 years seems but a tiny blip in the assessment of time, though we will have to do something about it as time progresses and measurements become more accurate.

Then of course you can adjust dates to the commonly accepted BC (Before Christ), except that in many countries it is now BCE (Before the Common Era). BC dates can be uncalibrated, or calibrated (usually: bc or cal BC). Uncalibrated bc dates are rarely used nowadays and usually achieved by subtracting 1950 from the uncalibrated BP date.

There are also calendar years (sometimes known as human years). And quite a lot of publications generalise as ‘years ago’. You will sometimes see dates quoted as 8 ka BP, which in general means 8000 BP and is usually calibrated, though the publication may specify that it is not.

Do you begin to see my problem? And I haven’t even begun to talk about the Bayesian analysis of dates (see here).

Add to the murk the fact that while archaeologists prefer quoting dates as cal BC, geoscientists will usually use cal BP. So, an interdisciplinary project can run into problems.

And not just interdisciplinary, multi-authored, projects; I’ve been reading a book that cites date ranges as uncal BP, cal BP, unspecified BP, years ago, 14C years ago, and cal BC; all in few pages of text. There are also a few uncalibrated original dates to play with.

It is a nightmare to make sense of it all. I’m not sure what the answer is, but I’m sure there must be a better way. For now, all we can do is plead for people to use one standard throughout their work, explain clearly what it is at the start, and stick to it!

I’m adding to this a link to the Archaeology Blogging Carnival that runs this month, perhaps there are other potential bloggers out there who like to think about archaeology!