Re: Decalogue, Sabbath.

In message <87645a6xmg.fsf@xxxxxxxxxxxxxxxxxx>
Gareth McCaughan <Gareth.McCaughan@xxxxxxxxx> wrote:

See, I'm no expert on dendrochronology, but I have a pretty good
idea of how the sort of matching-and-checking involved would work,
and how C-14 dating might be useful for disambiguating matches,
and it doesn't look to me at all as if that would make it uninteresting
that they agree well with one another. It might turn out to, but I'd
want to see some numbers: how many "patches" a tree-ring chronology
stretching over (say) 20k years typically involves, and how much
overlap they have, and how good the match is, and how common matches
of that quality are, and so on.

I don't have that sort of information, but you may be interested in this
article that I wrote as a review/comment on something by Peter James. Feel
free to skip if you find it boring.

==========
In the July/August edition of "Minerva", a magazine about things
archaeological published in Britain, Peter James deals with the question of
dendrochronolgy and the ancient world. Peter James is the principal author
of "Centuries of Darkness", a book which pointed out the fact that every
culture so far investigated in the Middle and Near East has a gap of several
centuries.

Archaeologists explain this gap by claiming that the population packed up
and moved elsewhere, driven, perhaps, by famine or invasion or some other
cause. This is plausible and so long as each archaeologist stuck to his own
speciality, it was quite possible to believe it. Peter James, however, took
a broad view of all the cultures, all the nations, and quite reasonably
pointed out that if all the people from area A disappeared, they ought to be
found in areas B, C or D which surround area A. If, however, the people from
area A are not to be found in B, C or D and, in addition, the people from B,
C and D are also thought to have disappeared - and none of them are to be
found in nearby E, F, G or H - then something is wrong.

His contention was that the "something" was the chronology. Egyptian
chronology, upon which all the other chronologies are based, is too long and
this has led to the gap in the chronologies of the surrounding nations. Let
me dream up an example to demonstrate the problem.

In 1770 the American states revolted from their proper loyalty to the
British crown and set up a series of presidents. Meanwhile, however, the
British monarchs continued to reign and were acknowledged in nearby Canada.
Let us suppose that in another thousand years some historian, completely
misunderstanding the problem, thinks that the monarchs whose names he finds
on Canadian coins also ruled in America and concludes that the American
presidents must have come after them. So you would have George I, II and
III, various Edwards and then Elizabeth II, followed by George Washington
and so on down to Bill Clinton.

Meanwhile another historian is working on the history of France and
discovers that Louis XV ruled in the reign of George I and Napoleon in the
time of George Washington. "Hah!" he says. "We have established that there
were two centuries between the time of George I and George Washington, so
there must be two centuries between Louis XV and Naoleon Bonaparte. There is
no evidence of any history - no new buildings, no coins, no literature -
between those two rulers, so clearly the people of France, feeling
threatened by American nuclear power, fled from France and remained away for
two hundred years until Napoleon led them back again into their ancestral
lands following the death of Elizabeth II."

You can see the fallacy. In fact the French haven't been anywhere - there is
no gap! It is the mistake in American chronology that has made the gap
appear. Correct the American chronology so that George I and George
Washington are contemporaries and the gap disappears.

This is what Peter James claimed in "Centuries of Darkness". The gap in the
chronologies of Greece, Italy, Syria, Israel and all the rest of them is
artificial. Correct the chronology of Egypt and the gap will disappear.

The archaeological establishment, by and large, although shaken by James'
book, has clung to the old way of looking at things. They have claimed
support for the conventional dating from radio-carbon dating and from
dendro-chronology - and it is these pillars of orthodoxy that Peter James
attacks in his article.

Before dealing with his article, however, I need to explain what
dendro-chronology is and some of the problems it faces.

Trees, as every schoolchild knows, have rings and, so far as we know, one
ring is added for every year of the tree's life. Thus if you come across a
tree that was cut down yesterday, you can count the rings and find out how
old the tree is and from that calculate when it began to grow.

Furthermore - although most schoolchildren will not know this - the rings in
a tree will differ in width according to the climate. A year of drought will
produce a thin ring, a year when the rain came in good measure and the sun
shone brightly will produce a thick ring. Let us suppose that you find a
newly felled tree with 100 rings and the four innermost rings are thick,
thick, thin, thick. Then you notice an old piece of timber whose four
outermost rings are thick, thick, thin, thick and you realise that its four
last years correspond to the first four years of the newly felled tree. You
are now able to produce a tree ring sequence that incorporates the 100 rings
of the newly felled tree plus the rings in the old piece of timber.

Repeat this process often enough, using increasingly older wood, and you can
produce a sequence of rings going all the way back to - well, to a very long
time ago.

It all sounds very easy, but like everything in life, things are never quite
that simple. For a start, trees grown in different locations can display
quite different patterns of rings. It is unlikely that the rings from a tree
grown in West Australia would match the rings of a tree from Queensland's
rain forest. However we need not be quite as extreme as that: trees growing
on different sides of the same mountain can show very different ring
patterns if one side is in rain shadow and the other is not! So the first
caution is that when assessing tree rings, you have to be sure that the
trees came from broadly similar locations.

Then there is the problem of matching the rings. Obviously the four rings of
our example - thick, thick, thin, thick - make too common a sequence to
allow us to reach any definite conclusion. You could probably find such a
sequence in half a dozen different places in a single tree. You have to use
twenty or thirty or even more rings to make a good match. The trouble is
that it is highly unlikely that any two trees will have exactly the same
pattern of ring growth over such a period of time. One tree will be fine and
healthy in year X, laying down a good thick ring, while some kids playing
with matches scorched the back off the tree next door to it, which in
consequence only managed a thin ring. So if you take thirty rings from both
trees and twenty-nine of them match, can you say that the ring squence is
correct? What if only twenty-eight match?

Dendro-chronologists use statistical methods to determine the best fit -
it's called "wiggle matching", if you really want to know - and when they
get their computers to search for the best place in which to put a new tree,
they may be given half a dozen different possibilities, each one with a
probability percentage. So what do you do if two sites have the same high
probability?

The second caution, therefore, is to make sure that your wiggles are
correctly matched up.

And then we come to the problem of actually using dendro-chronology. The
common picture is of an archaeologist discovering a piece of wooden
furniture. He looks at the pattern of rings in the arm of the chair or the
leg of the bed, matches it with some master sequence, and can immediately
say, "This chair is 1,679 years old."

Would that it were that easy! You see, although it might - just might - be
possible to match the pattern of rings in the bed leg to a pattern that
occurred 1,679 years ago, we simply do not know how many rings the carpenter
removed as he planed and polished the chair. The two-inch wide bed leg may
have come from a two foot wide tree - and there may have been a hundred or
more rings in the missing one foot ten inches of tree!

The third caution, therefore, never forget the missing wood.

This is one reason why archaeologists were so excited by the tomb of King
Midas at Gordion. The burial chamber beneath the huge mound was made up of
enormous juniper logs that had as many as 918 rings. When the ring sequences
of all the logs were matched up, it was found to span a period of 1026 years
- and best of all, the logs still had their bark on, to it was possible to
determine exactly when they were cut down!

The dendro-chronologists were happy too, for they dscovered that the
earliest rings formed a sequence that they could assign, with a fair degree
of probability, to the last rings of a sequence they had constructed from
logs found at Porsuk in south-central Turkey, a sequence that covered 400
odd years. This meant that in total they had a sequence covering 1503 years.

The only trouble was, when did Midas live - and, more importantly, when did
he die?

You see, there is no master sequence covering the history of central Turkey.
Starting with living trees and working back through timber found in Turkish
and Byzantine buildings, the dendro-chronologists have been able to
construct a sequence that goes from 2002 AD back to 362 AD - and no further!
The wonderful 1503 year sequence is left floating, unattached.

So the dendro-chronologists turn to radio-carbon dating for help. Now this
raises two problems, a philosophical one and a practical one. The
philosophical problem is that dendro-chronology is supposed to be an
independent method of dating and is claimed to confirm the radio-carbon
dates. It is hardly surprising that dendro-chronology "confirms" the
carbon-14 dates when it is based on those very dates!

The practical problem is getting hold of a firm radio-carbon date. Professor
Peter Kunihom of Cornell University has been working on the
dendro-chronology of the Aegean and Anatolia for many years. He has taken
many samples from the juniper logs from Gordon and the timber from Porsuk
and had it analysed at Heidelberg University. To everyone's surprise, he
found that Midas must have been buried in 547 BC.

This was considerably later than most people thought possible, so Professor
Kunihom sent off some more samples and this time the answer came back, 757
BC. Now that is just over two centuries difference - about the length of
time usually attributed to the gaps that Peter James called "the centuries
of darkness"! Even though the radio-carbon date has since been refined to
718 BC (this was done in 1996) this isn't much help, because not only has an
even more recent analysis put the date of the tomb at 740 BC, the same
analysis tested the earliest rings as well as the latest and found that
instead of there being 1053 years between first and last, there are 1033 -
so do we trust the radio-carbon date for the early rings or for the later
rings?

As Peter James says in his article, "The offset is explained by regional
variations in the amount of C14 - a possibility that has long been
suspected. To use C14 dates callibrated by one tree-ring sequence from
central Europe, in order to fix in time another from Anatolia and then to
use that sequence to modify our wider understanding of C14 behaviour, is
clearly a perilous exercise." (Minerva vol. 13 no. 4 p. 18)

Personally, I think that Peter James has made a most restrained comment.

Interestingly, in the same paragraph he comments: "The lower dates fall
largely in the 8th century, which has long been known as the beginning of
the 'radio-carbon disaster area' - a flat stretch of the calibration curve
stretching from about 800 to 400 BC." (ibid)

In other words, the amount of radioactive carbon present in a sample falls
off the older it is and you can plot this on a graph. The resulting line
should rise more or less evenly, with the oldest samples at the lowest point
and the youngest ones at the highest point. However between 800 and 400 BC
the line is just about flat, apparently indicating that radioactive decay
halted for four centuries! The theory behind radio-carbon dating is
scientific, but clearly there is something terribly unscientific about the
way it is being applied.

Similar problems bedevil attempts to link the Aegean and Anatolian tree-ring
sequences to ice layer sequences from Greenland. Several years of abnormal
growth were matched with ice layers that had a higher than average amount of
sulphuric acid, on the basis that the explosion of Thera was responsible for
both features. This gave a date of 1625 for Thera and therefore for the tree
rings. Unfortunately closer investigation of the ice layers revealed tiny
glass particles embedded in the compacted snow - and these were not from
Santorini! There is another period of high sulphuric acid twenty years
earlier and the dendro-chronologists are now pinning their hopes on it, but
further investigation is needed. (It must not be forgotten that the first
dates for the eruption of Thera placed that event around 1450 BC, based on
the styles of pottery discovered in the buried ruins on Santorini.)

In short, the chronology of the Aegean and of Anatolia is by no means fixed.
Other dendro-chronology sequences may be more secure - Peter James does not
comment on them - but considering that the eruption of Thera was a pivotal
event for many cultures, our inability to date it precisely and securely
must mean that any chronology which includes Thera must be uncertain.
Despite the best efforts of dedicated researchers such as Professor Kuniholm
and his team, there are still gaps to be filled and large areas of
uncertainty.
==========

In the absence of that sort of quantitative information -- indeed,
of *any* actual information at all -- all I have to go on is your
confident but unsupported claims that it's all circular, which I have
to weigh up against the improbability that all the scientists involved
are idiots. Now, sure, most people[1] are idiots much of the time, so
it's *possible* that there's some gigantic scientific blind spot here,
but on the face of it that doesn't look like the most likely scenario.

I have never claimed - and would never claim - that scientists are idiots.
To be mistaken is an entirely different matter and yes, there are times when
all or a large majority of scientists are mistaken. One thinks, for example,
of the widespread rejection of plate techtonics when the idea was first
mooted.

God bless,
Kendall K. Down

--
================ ARCHAEOLOGICAL DIGGINGS ===============
| Australia's premier archaeological magazine |
| http://www.diggingsonline.com |
========================================================
.

Relevant Pages

  • Re: Decalogue, Sabbath.
    ... one, as I have described, is when a "new" tree fits the ... sequence equal well in two or more places. ... nice set of rings in the leg of a chair and from that you cleverly build up ... Critics of dendrochronology point out that even ...
    (uk.religion.christian)
  • Arguement from Authority Re: Carbon-14 fallacy in dating
    ... trees with a number of rings larger than 5766, ... scientists assembled lots of the rings of trees together from different ... of the oldest tree from this information, sort of like how they do ... Tree rings are not discrete quantities like DNA, so this methodology ...
    (talk.origins)
  • Re: Carbon-14 fallacy in dating
    ... trees with a number of rings larger than 5766, ... of the oldest tree from this information, sort of like how they do ... Tree rings are not discrete quantities like DNA, so this methodology ... pretty much bet the farm that the overlap is real and not illusory. ...
    (talk.origins)
  • Re: Re: The savage athiest evil of uranium
    ... archaeology class where we discusses dendrochronology and carbon ... except for the difficulty of correlating tree rings ... And to claim that Western Civilization is based on Christianity is ...
    (talk.origins)
  • Re: Poor Noah and tree ring dating
    ... lamination in the sense of to make or build up layers. ... form a material with the obvious strength and flexibility to face the ... Nobody is proposing a tree on that model. ... Rings are not ...
    (talk.origins)
Loading