I have been receiving further information on the site and it is clear
that not only are there problems in the C14 dates for the Iron Age in
Britain, generally, but this site contains some additional anomalies in
that respect and there is a possibility of some contamination. The C14
dates on the grave were for the pig bones (no later than the 3rd Cent.
B.C.) and the human bones (no later than the early 2nd Cent. B.C.). Both
of these dates are far from the 5th to 4th century B.C. date that was
reported not long after the discovery. I must presume that it was the
anomalies in the data that led to the earlier estimation. This also
calls into question the validity of the exceptionally late dates for
some of the bones in the surrounding ditch in the link I gave and which
cites early 3rd Cent. A.D. from recent C14 analyses.
With regard to the involute brooch typologies, there are several which
must be considered. Most important is the general chronology of the
type, itself. Involute brooches are all of La Tène II in that the foot
returns to be attached to the bow. Although involute brooches are a
British phenomenon, La Tène II brooches are well established in their
relative chronology throughout the central and western Celtic world. It
would be difficult, indeed, to make a claim that British La Tène II
brooches precede some La Tène I brooches from the Swiss Plateau and even
precede some La Tène I brooches from England (such as the Wessex type).
There are two factors to be considered about the internal chronology of
the involute type. The first is whether the catchplate is open or solid.
Jope (2000, p. 49) uses the example of the Woodeaton involute which has
a solid catchplate and that this feature "is hardly to be seen" on the
earlier continental brooches. He places the Woodeaton brooch at the end
of the sequence "though not necessarily later than the first century
B.C." To counter this, there is the early La Tène III brooches of Hull's
Glastonbury type which derive from the La Tène II type, yet have an open
catchplate. All Yorkshire involutes have an open catchplate so there can
be no chronology offered on this feature specific to the region.
The second factor is the whether the curve of the involute is shallow
(early) or extreme (late). As there are three regions which yield
involutes: the southwest, middle Thames, and Yorkshire, the situation
regarding this feature is unclear and it might to be a general
observation over all the regions, although I have not looked into this
to verify that statement. Be that as it may, what is important is to
demonstrate an evolutionary process in this feature -- something to say
that the extreme curvature became necessary because of some functional
consideration. Without that, there can only be whim or fashion and the
expressions of these could thus be variable and arbitrary.
Because of the dificulties and anomalies in other dating methods for
this period, nothing can be used as a check to verify chronologies. Very
broad chronologies (which place involutes under the heading of La Tène
II), because of their numbers and broad range across Europe. This can be
easily demonstrated to appear on the time scale in different positions
depending on geographical factors and yet all in a logical order which
corroborates the overall chronology.
There can be no external scientific procedure which could yield a
mathematical plus or minor error range for a typology and any given
would be completely guess-work. The best that we can say about the
involute type is that it is 2nd century B.C. with the possibility of
some being either slightly earlier or slightly later.
For my study of Coriosolite die chronology I expressed fixed and
variable chronological points thusly:
1, 2, 3 (4, 5,) 6, (7, 8, 9, 10, 11) 12, 13, (14, 15)
The numbers within parentheses are variable and their internal order is
based only on my best educated guesses while the numbers outside the
parentheses and the parentheses themselves are absolute and are
demonstrated through several overlapping evolutionary sequences where
the chances of error are statistically minuscule. Like my example above,
the real data gave only a maximum of five variable points within a
I maintain that better evolutionary traits of design should be plotted
for Celtic art. It is extremely varied and thus good chronologies can be
built by using a proper art-historical analysis. In works on Celtic art,
this technique is used passim, but is thus very difficult to follow as
most of the objects are grouped by type, instead.
Now, if we apply all of this to the Ferrybridge burial (setting aside
the outer ditch data), it follows that the pig bones cannot possibly be
any earlier than the late 3rd century B.C. to allow for the "gray-area"
at this end of the La Tène II involute brooch range. If the relative
chronologies of the bone material are pig followed by human, it presents
quite a dilemma in the interpretation. Why would we imagine that a human
being was buried to commemorate an event that was previously marked only
by a feast of pork?
Moving on to the facsimile parts in the burial. I have been informed
that these were some of the terrets and that they were not filled with
ceramic as the first news accounts said, but instead of silt that
appears to have been formed wet, then dried and contracted. There was
evidence of charcoal and hammer scale upon close examination. This
suggests to me that the same techniques of repoussé were being used that
emerged in the 3rd century B.C. and continued into the 1st century A.D.
The earlier objects, by and large, were rather thick in section whereas
the skill developed to allow very thin copper alloy to be worked in this
way at the end of the sequence. Gold work is another matter, being far
more malleable than bronze. The initial reports say "foil", so if it
was, indeed, this thin, then a very late manufacture is possible and a
very early manufacture less likely.
The next part is very "unscientific", but I feel that it is a mistake to
rely completely on the science when the model thus revealed lacks a
certain practical coherence. I demonstrated this point with coins from
the Trébry hoard which had been analyzed through neutron activation and
had been shown, statistically, to establish that the source alloys were
being debased by about 2% for each class. First of all, those particular
classes were a modern construct and did not reflect any intentions of
the moneyers to arrange the issues thus. Therefore, the 2% was a curious
coincidence and nothing more. secondly, the hoard had been constructed
from coins that had already been in "circulation" and were thus subject
to Gresham's law that bad money drives out good: the more intrinsically
valuable coins had been culled to profit on the metal. Thirdly, was the
psychological discovery that a repetitious task builds up to a certain
efficiency rating and then drops off when boredom sets in. This was not
just noticed in human beings, but an experiment with planarian worms --
which of course do not even have brains, revealed the same mechanism. I
mentioned this to an English professor friend of mine as proof that even
very low forms of life have a rudimentary intelligence. He dismissed the
idea, saying that he was not convinced that even some of his own
students had a rudimentary intelligence.
The second and third factors, together, created the illusion.
We must ask ourselves why such an apparently important chariot burial
consisted of mismatched chariot parts with some facsimile terrets when
other chariot burials of the alleged period do not. Furthermore, what is
to be made of the varying dates of all of the finds from within and
without the burial? Could some of the C14 anomalies have come from the
situation of a reburial during the Roman period? Was the site itself
really just a burial that had been commemorated, or was it a religious
or council site that had a staged burial as part of its features?
One thing that is certain is that none of the dates of the objects can
be much closer than than about 175 years later than the 400 B.C. that
was stated in some of the reports, or 275 years later than the initial
"500 B.C." If the human skeleton was not a secondary burial, then there
must be something wrong with his C14 date. Is it possible for such a
date to be as much as 250 years off?
I have also been informed that the skeleton was slightly flexed and not
completely laid out flat.
Andy Holland wrote:
> Whilst John's comments are very interesting and he poses some great
> archaeological questions......
> ....can I just add a cautionary note about the dating of archaeology.
> Firstly the radiocarbon dates of the British Iron-Age..... they are
> terrible! It's unavoidable and is because the dendro based calibration
> curves for Northern Europe and particularly Northern England at that
> time are very flat - so you get a very wide error range. No matter how
> accurate your C14 date the date is never going to be very precise for a
> British Iron Age sample and it's a real pain in the rear. (There are
> currently some interesting PhDs and research projects looking at Baysian
> analysis of multiple C14 dates in stratigraphic sequences to try to help
> improve the dating resolution a little).
> Also whilst I'm no expert on the intricacies of Involute Brooch typology
> can I also add a cautionary note about the use of typologies as a form
> of dating - they're only supposed dates for these sequences and they are
> only ever as good as the other dating technique used to date the
> archaeological context from which original examples of that type of
> brooch come from.
> Many typologies have a long history of study - which is all well and
> good, however, the earlier studies of a particular type will attribute a
> particular type to a certain date based on the dating evidence available
> at the time the study was carried out..... those dating techniques may
> now be suspect, prone to errors unknown at the time or just plain wrong.
> Hence the typology dates can be out by a considerable degree - which is
> why whilst they can be useful, they should not be considered absolute
> dates or used as a single piece of evidence
> I would make a plea for typologies to adopt the same scientific rigour
> of other dating techniques by the mathematical calculation of error
> ranges and deviation (multiply the errors from each dating technique
> used together to form the error for the typology overall).
> This is not to suggest that John's hypothesis is wrong (in fact I'll be
> asking some of his questions to my friends who have worked on the
> Ferrybridge samples about the dating next time I see them) but just an
> observation from an archaeological science trained archaeologist to
> throw into the melting pot for consideration.
> Many Thanks
> The statements and opinions expressed in this message are those of the
> author and do not necessarily reflect those of the Council for British