Chestnut Article
This is a rough Copy of an article I wrote for the Quarterly Journal of Forestry. The article actually contains more Photos.
If the Romans Did Not
Bring Sweet Chestnut to
Britain, Who Did?
John Pitcairn looks into this intriguing question following a
throwaway remark at the 2023 Whole Society Meeting.
At the recent excellent Whole Society Meeting hosted
by the South Eastern Division, we saw exemplary
sweet chestnut coppice silviculture and were
introduced to its economic and cultural importance. At
one estate the host, in an almost throwaway remark, said
that a Dr Robin Jarman had made a convincing case for
the Romans not being responsible for the introduction of
sweet chestnut into Britain and that the first evidence for
it was from the 6th or 7th century. This, according to my
schoolboy history, was when the Jutes, Angles and Saxons
arrived from Denmark and northern continental Europe.
These are not areas I associate with sweet chestnut, and
a Roman introduction seemed more likely to me. I realised
I had no idea what evidence there might be and decided
to do some research online. It was an interesting idea to
explore and focussing on papers that were freely available
I quickly located and downloaded a thesis ‘Sweet chestnut
(Castanea sativa Mill.) in Britain: a multi-proxy approach
to determine its origins and cultural significance’ by Robin
Andrew Jarman (Jarman, 2019). I did read all 57 pages
of text but not necessarily the seven pages of references!
I thought readers might appreciate a summary of what I
discovered.
Fortunately, the introduction did explain the importance
of determining the date of sweet chestnut’s arrival, which
I thought interesting but unimportant. Whether sweet
chestnut is indigenous, an archaeophyte (an ancient
introduction) or a neophyte (a modern one) is a significant
question. The question relates to its naturalness, a key
function in the Nature Conservation Review 1977 which
determines which species should be protected or regarded
as invasive; this might become relevant should sweet
chestnut blight take hold in this country.
The aim of the thesis was to determine:
When is the earliest verified date for sweet chestnut
growing in Britain?
Whence do the longest-established British sweet
chestnut trees derive?
Does new evidence for antiquity and origins alter the
ecological and cultural significance of sweet chestnut in
Britain?
 |
King’s Wood near Worksop, Sherwood Forest, Sweet Chestnut originally planted by the Dukes of Portland. |
I shall just cover the parts relevant to sweet chestnut’s
arrival and any information as to where it may have come
from, considering what the evidence is and why Dr Jarman
cast doubt on it. The debate on the origin of British sweet
chestnut has a long history. John Evelyn considered it
non-native and in his 1706 4th edition of Silva suggested a
Roman introduction. This became
the consensus view of the Royal
Society in the 18th century but
lacked evidence.
Oliver Rackham believed
sweet chestnut to be a Roman
import, or so I thought. Dr Jarman
reveals he worked with Rackham on and off from 1971, and
that Rackham had reservations about the species being
a Roman introduction. The idea was being quoted and
requoted in an academic echo chamber without the primary
evidence being checked. It became a ‘factoid’ as Rackham
called it (something everyone believes but is not, in fact,
true). As I am taking what Dr Jarman has written at face
value, I may be contributing to a new ‘factoid’! Dr Jarman
has tried to re-examine original archaeological material
and interview the researchers. He also visited semi-natural
woodlands as well as ancient chestnut trees both alive and
dead, to establish its cultural and social significance.
A literature review was undertaken, and archaeological
archives were reviewed looking for pollen, charcoal, nuts,
and timber references prior to 1350 AD. An attempt was
made to locate all recorded archived specimens up to
650 AD and to re-evaluate their authenticity with modern
techniques. This was carried out in partnership with English
Heritage.
The thesis was essentially a summary of a number of
previously published papers that
Dr Jarman had been involved
with, usually as lead author.
The thesis is divided into four
sections: (1) archaeological
records, (2) dendrochronology,
(3) genetics, and (4) historical
ecology. What follows is my understanding of the key points
Dr Jarman makes grouped into ‘archaeological records’
and then a combination of the last three sections of his
thesis.
Archaeological records.
Dr Jarman located 35 archaeological records of archived
sweet chestnut material from before 650 AD, i.e. the
Roman and post-Roman period. He tried to find and
re-examine them physically. For me, the most striking
finding was that there were no confirmed records of sweet
chestnut pollen prior to 650 AD.
There were only two reports of archaeological records
of sweet chestnut nut fragments in the pre-medieval
period. One was from an excavation at Great Holts
Farm in Essex in the 1990s. The find consisted of five
fragments of chestnut pericarp found in a Roman well that
had been filled with rubble and rubbish. The waterlogged
bottom of the well contained organic material including “a
few fragments of sweet chestnut nut pericarps, walnuts,
hazelnuts, olive stones, grape pips, stone pine nuts,
cherry stones, sloe, bullace and apple pips.” Some of
these items must have been imported, for what looks like
feasting. This cannot be considered good evidence for
a local source of chestnuts, not that it had been claimed
to be. The other reported sample, from Castle Street in
Carlisle, was radiocarbon dated to a much later period.
Reports of the sweet chestnut charcoal remains
also proved problematic, in part because many of the
recorded archived pieces could not be located. There
was also a degree of misidentification. Samples recorded
as sweet chestnut turned out to be ash or alder, but
mainly oak. Small pieces, particularly small branch
wood, may be indistinguishable from oak. Oak can
confidently be distinguished from sweet chestnut by the presence of multiseriate medullary rays. The problem is
that the absence of rays particularly in a small fragment
is not proof it is chestnut. Creating fresh surfaces for
examination was not possible due to the importance of
the specimens. The dating of some samples could not
be verified or was confirmed to be inaccurate. Some key
samples were from 19th century excavations and their
accounts had been regularly requoted but had not been
questioned. Reported wood and wooden artefacts were
confirmed; however, a writing tablet
or a tool handle could easily
have been imported.
There are waterlogged
chestnut stakes and
piles from the Alverstone
Marshes on the Isle of
Wight, which have been
carbon dated to the 6th and
9th-10th century, but this study had not been published
when Dr Jarman put his thesis together in 2019.
There is a possible sweet chestnut pollen grain from
Uckington, Gloucestershire that is dated to the 7th
century. This with the Alvestone Marshes wood could be
the only physical evidence of sweet chestnut products
that could have been derived from plantations established
in the Roman period. In summary, no definitive archived
material could be found to confirm locally grown sweet
chestnut in England and Wales during the Roman and
immediate post-Roman period.
 |
Sweet chestnut regenerates freely. |
Historical ecology/genetics/
dendrochronology. The earliest written record for growing sweet chestnut
is in the 12th century (1113 AD), from legal ecclesiastic
documents near the current Forest of Dean. It must, therefore,
have been established in the 11th century at the latest.
There are a number of large ancient sweet chestnut
trees on estates and in former deer parks, some shown to
be more than 400 years old. Dendrochronology does not
take any sweet chestnut back more than a few hundred
years; DNA analysis has confirmed ancient clonal coppice
stools but provides no date. DNA might provide evidence
for where introduced sweet chestnut originated and so
provide clues to who introduced it. DNA analysis has
been carried out on British and Irish sweet chestnut for
comparison with European records.
Sweet chestnut along with other flora retreated to refugia
ahead of the ice sheets of the Ice Ages.
Three refugia
genepools have been identified. An eastern (E. Turkey
and the Caucasus), a central (W. Turkey, Greece and part
of the Balkans), and a western (Italy, Spain, Portugal and
southern parts of France and Switzerland). DNA samples
from England and Wales are from the western area but they
represent two separate genepools. These genepools are
geographically mixed, suggesting at least two substantial
introductions. The older British DNA samples showed some
commonality with samples from Portugal to Romania. It is
a diverse genepool but whether that is because of multiple
introductions or because of mixing on the continent prior
to introduction cannot currently be
determined. Ancient parkland trees
usually came from NW Iberia
while plantations <200 years old
were more likely to be Italian;
one can imagine members of
the gentry returning from their
grand tour with pockets full of
chestnuts!
Sweet chestnut does not inform British myth or legend
but is often planted in parks or gardens in significant
places. It is of considerable economic and practical
importance. One of the reasons for calling sweet chestnut
an ‘honorary native’ is that it forms ecological associations
within semi-natural woodlands, which have been stable for
hundreds of years. They are also similar to forest vegetation
associations on the continent.
My initial reaction to the lack of pollen evidence may
have been misplaced. British sweet chestnut pollen
samples are very rare from the whole medieval period,
Sweet chestnut produces viable seed and naturally regenerates.
“If the Romans made
a concerted effort to introduce
sweet chestnut, there is evidence
they failed!”