Earliest Koran dating and the 'trouble' with carbon dating

[Imperial Dave]

A new entry on a long running 'argument' in archaeological circles....how accurate is carbon dating

http://www.bbc.co.uk/news/uk-england-35495035

[Patrick Waterson]

"Carbon dating is backed by scientific rigour, repeatable and verifiable." - Professor Cook.

All that is missing is the word 'accurate' ...

Seriously, a key potential problem lies in the assumption that "This stored C14 then starts to break down at a regular rate".  Other radioactive elements do not necessarily do so: U235, for example, has a half-life of 703.8 million years, but in critical mass it can break down a whole lot faster.  The half-life of C14 was measured or calculated in 1962 as 5,730±40 years, and while the variation is trivial, it indicates that the decay rate may not be wholly stable and more significantly may be accelerated or retarded by outside influences.

There are also other influences on the ratio; one of these is noted in the article.

This was thrown into sharp relief when Prof Cook's team worked on the headline-grabbing dating of Richard III's skeleton.

The first dates came back as AD1430-1460, when Richard was known to have died in 1485.

"But then stable isotope analysis of the bones showed the individual had eaten a lot of seafood," he said. "This gives a reservoir effect that makes the age too old.

The 'reservoir effect' means that when carbonates dissolve into water, they donate carbon which is free of C14 (or the C14 is a lot older and hence more scarce) and thus makes things look older because a subject consuming them ends up with less C14 to C12.

And then there is Thera ...

But, as Dr Michael Dee, Leverhulme Fellow from the University of Oxford's School of Archaeology, explains, its own date is a matter of fierce debate.

"Historians have traditionally put the eruption at BC1500 but carbon dating, using organic remains from the island, has placed it between BC1627 and 1600.

"That is six generations earlier and it overturns the idea of crippled societies being overrun. It gets quite heated."

Now here nobody has yet thought to invoke a 'reservoir effect' despite Thera being adjacent to the sea.  Presumably the southern Aegean is not rich in dissolved carbonates.

So is radiocarbon dating worth bothering with?  It would seem so, but with rather more caveats and less dogmatism than when first envisaged, and it would seem particularly risky to base dating on radiocarbon analysis alone.

[Erpingham]

The 'reservoir effect' means that when carbonates dissolve into water, they donate carbon which is free of C14 (or the C14 is a lot older and hence more scarce) and thus makes things look older because a subject consuming them ends up with less C14 to C12.

This is scarecely a new thing.  I remember having to measure C13/14 ratios when I (briefly) worked in radio carbon dating in 1978.  Ratio correction has probably improved since then.

So is radiocarbon dating worth bothering with?  It would seem so, but with rather more caveats and less dogmatism than when first envisaged, and it would seem particularly risky to base dating on radiocarbon analysis alone.

Of course C14 dating is worth bothering with.  It has its problems but by comparison with having to date everything by overlapping sequence, its still considerably more accurate.  Just perhaps not as precise as some of its more enthusiastic advocates have claimed.

[Imperial Dave]

in my day (a long time ago), we were taught to use at least 2 points of data if not more where possible. Therefore the 1980's view of radio-carbon dating was that it was a useful pointer but not uber accurate and really needed something else to help hone in on the date. Recent history (ie the past few thousand years) allows us to use things like coinage (althou8gh that can be fraught with dangers too). The best evidence to support radio-carbon dating (assuming it wasnt too far in the past) was always cited to be tree ring analysis if you were lucky enough to have some handy wood lying next to the thing you were trying to radio-carbon date! Having said that, there are many who reckon tree ring data is the most accurate way of dating stuff

[Jim Webster]

I would suspect that tree ring data, if you're lucky enough to have a really long local sequence, is probably as accurate as anything, and of course it can allow you to calibrate carbon 14 as well

[Erpingham]

I would suspect that tree ring data, if you're lucky enough to have a really long local sequence, is probably as accurate as anything, and of course it can allow you to calibrate carbon 14 as well

Long local sequence is the key - geographical variations are much more important than was appreciated when I was taught this all those years ago.  Sample size is also quite large compared with modern C14 and, of course, wood only survives in certain contexts - less commonly than bone or burnt material, for example.

[Imperial Dave]

hence why its better for 'our' timeframe, ie generally recorded history and less so the further back into prehistory you go

[Patrick Waterson]

Dendrochronology, interestingly enough, backs up the 17th century BC radiocarbon dating for Thera: 1628 BC is the 'year of no growth' which puts 1629 as the year in which a lot more than just Santorini went ballistic.

This also ties in (exactly, which is something of an achievement) with Hebrew records in pointing to 1629 BC as the date of the Exodus amid a series of world-affecting calamities, but that is another story ...

The dating of the Koran fragments or at least their paper could perhaps utilise another technique which seems to have been rather shouldered out by radiocarbon dating: pollen analysis (nowadays subsumed under palynology).  One unfortunate feature is that radiocarbon dating tends to be used to 'calibrate' palynology, rather reducing its value as a means of determining date.  Palynology is best when used to determine location, e.g. the starting-point of a textile discovered somewhere in Europe and of claimed Middle Eastern origin.