TRANSACTIONS

PERTHSHIRE
of NATURAL

SPECIAL

ISSUE

TRANSACTIONS OF THE

PERTHSHIRE SOCIETY
of NATURAL SCIENCE

SPECIAL ISSUE

Papers presented on the occasion of
the 25th Anniversary of the
ARCH/EOLOGICAL AND HISTORICAL SECTION
in the City of Perth Museum and Art Gallery
on 24th March, 1973

CONTENTS

INTRODUCTION

hy R. Butchart, President 3

THE PREHISTORY OF PERTHSHIRE OF THE
4th, 3rd AND 2nd MILLENIA, B.C.

hy Margaret E. C. Stewart, Ph.d., F.S.A.Scot. 5

ROMAN “SIGNAL STATIONS ’’ ON THE CASK RIDGE

hy Anne S. Robertson, Hunterian Museum 14

PERTH: THE FIRST CENTURY OF THE BURGH

hy A. A. M. Duncan, Professor of Scottish History, University 30

of Glasgow

Cover: From early lithograph of Perth by D. O. Hill.
By kind permission of the Curator.

The Society wishes to acknowledge the generosity
of the Council of British Archaeology, the
Forteviot Trust, and the University of Glasgow,
for grants towards publication of these papers.

3

Introduction

At a meeting of the Council of the Perthshire Society of
Natural Science (now in the 106th year of its existence), it was
decided to form an archseological section and a meeting was
convened by Dr. Margaret Stewart in the Art Gallery on the 24th
February, 1948, which was attended by fourteen people.

The section was formally inaugurated at the meeting, its
objectives were set out and a committee, with Dr. Stewart as
President, was elected. The section became affiliated to the Council
for British Archaeology and a member of the Scottish Regional
Group of that body.

During its existence the section has arranged lectures in
archaeological and historical subjects, been responsible for, and
assisted in, excavations, visited sites of interest, held joint meetings
and visits with kindred societies and attended meetings and schools
in various parts of the country. For a number of years the section
has assisted the Archaeological Division of the Ordnance Survey in
looking over the countryside for unrecorded archaeological sites, so
that these could be entered, after confirmation, on new map issues.

In 1961 the activities were extended to include those of
historical interest and the section title was amended accordingly.

On 24th March, 1973, an open meeting was held in the Art
Gallery to celebrate the 25th anniversary of the founding of the
section, which 120 people attended. These included representatives
from many kindred societies throughout Scotland.

The meeting was opened by Mr. R. Butchart, President, who,
on behalf of the section, extended a welcome to those attending,
and invited Mrs. M. Morrison, a past President, to introduce the
first speaker. Dr. M. E. C. Stewart, M.A., Ph.D., of Perth, who
gave an illustrated lecture on “ The Prehistory of Perthshire.”
After a break for coffee Dr. Anne Robertson, M.A., D.Litt., of the
Flunterian Museum, The University of Glasgow, whose subject was
” Roman Signal Stations or Watch Towers on the Gask Ridge,”
was introduced by Mrs. D. M. Lye, a past President.

The meeting then adjourned for a formal lunch at the Royal
George Hotel. On the resumption of the meeting Mrs. K. S.
Simpson, F.S.A.Scot., another past President, introduced the
third, and last, speaker. Professor A. A. M. Duncan, M.A., of the

4

Department of Scottish History, The University of Glasgow, who
spoke on “ Perth. The first century of the burgh.”

The proceedings closed with a social gathering in the Museum
Rotunda where the celebratory cake, topped by a most remarkable
model of a typical excavation made by Miss D. Alexander, a
member, was cut with due ceremony by Dr. Stewart, the founder
of the section. The toast (in champagne) to the section was
proposed by Mr. J. Aitken, President of the Perthshire Society of
Natural Science, and was responded to by Mr. Butchart, President
of the section. Professor Duncan spoke on behalf of the visitors
and extended their congratulations to the section on attaining its
25th anniversary and good wishes for its future.

5

The Prehistory of Perthshire in the
4th5 2nd Millennia, B.C.

Margaret E. C. Stewart

Very little Mesolithic material has come out of Perthshire in
spite of the fact that the Tay must surely have attracted early
explorers. The lack of evidence for riverine settlement is all the
more surprising in view of what has been found near the estuary.
Mesolithic material has come from shell middens at Stannergate
and probably also from near Droughty Ferryh But further up the
river, apart from the canoe at Friarton near Perth^ and the founda¬
tions of a hut and an associated heap of the shells of edible fresh
water molluscs found at a depth of 20' below the General
Accident Buildings at the east end of High Street in Perth-^ no
other evidence has ever come to light. This lack of Mesolithic
material is all the more remarkable in view of the evidence for
North European Forest Cultures in the Forth valley.

Surface collected flints from the farm of Gothens near Blair¬
gowrie^ and from cultivated land below Dunsinnane have a
Mesolithic Neolithic facies but their chronological context is un¬
certain.

By the 4th and 3rd millennia B.C. the picture becomes much
clearer. In the second volume of The Chambered Tombs of
Scotland Miss Henshall lists six certain examples in Perthshire.
These are the tombs of Clach na Tiompain in Wester Glenalmond,
Kindrochet, Rottenreoch and Cultoquhey in the valley of the Earn,
Derculich and Fortingall in Strathtay^. To that list should be added
the tomb at Clathick in Strathearn, and in spite of its disqualifica¬
tion by Miss Henshall, the ruined tomb at Dull in Strathtay.

Clathick is unexcavated and lies exposed on moorland on the
north side of Strathearn between Crieff and Comrie. Like
Cultoquhey the inner end stone of the chamber rises for over 2'
above the flat slabs -which roof the probably two compartmented
tomb.

The site at Dull though it ended up as a 14th century corn
roasting kiln started as a polygonal chambered passage grave.

6

Some of the upright slabs of the north wall of the original entrance
passage were still in situ and it was noticeable that they splayed
outwards as they approached what would have been the margin
of the covering cairn. The same feature can be seen in some
polygonal passage graves in the Hebrides notably Barpa Langass
(UST6) and Loch a’ Bharp (UST22) where the splay, now covered
by cairn material, may originally have been an open funnel entry.
At Dull the passage slabs could have had no functional significance
for the corn roasting activities all of which took place in the
remains of the chamber^. From its position on the north side of
the valley the site at Dull would be complementary to the polygonal
passage grave chamber further east at Derculich.

Of Miss HenshalFs list of six, Clach na Tiompain, Kindrochet
and CuJtoquhey have all been excavated.

From Mr. J. G. Scott’s recent work^ on the typology of the
segmented gallery grave type of chambered tomb it is now possible
to regard Clathick and Cultoquhey as early rather than degenerate
examples of their type. Fragments of pottery from the latter
belonged to a wide-mouthed carinated bowl of Piggott’s form G,
the diameters of the mouth and carination being equal. The pottery
is undecorated except for wide rippling on the neck and upper part
of the bowl. On the line of the carination there have been vertically
perforated lugs of which two survive. Miss Henshall points out^
the rarity of this feature which so far is known from only two
other sites in Scotland. It occurs in the Neolithic wares from the
tumulus of Pitnacree in Strathtay*^ and from the chambered cairn
of Achnacree on Loch Etive in Argyll. Chronologically it is dated
by the Carbon 14 assessment of 2860 B.C. ±90 (uncorrected for the
old land surface at Pitnacree. Such vertically perforated lugs
are unknown among contemporary Irish wares or from the
Neolithic pottery of northern England and only occur in the
causewayed camps of Windmill Hill and Whitehawk in the south
of England. It might be worth noting that the Cultoquhey bowl
doesn’t have the hooked rim so typical of carinated bowls with
ripple decoration from Scottish west coast chambered tombs.
Miss Henshall has suggested that Clach na Tiompain and Kin¬
drochet are multi-period tombs and that the former may have
started as a row of four single compartmented stone cist graves to
which various structural additions including the encapsulating cairn
were added. If so then the earliest period at Clach na Tiompain
could be regarded as part of the primary settlement of Central
Perthshire.

How these chambered tomb builders reached Central Perth¬
shire is a matter of controversy^®. The starting point is the Clyde
estuary and Dunbartonmuir at the south end of Loch Lomond
where there is a group of chambered cairns. A paper on Strathtay
in The Second Millennium B.C. suggested that the route into

7

Perthshire was by way of Loch Lomond, Glen Falloch and Glen
Dochart. Any alternative to this either by way of Loch Fyne or
the Trossachs would involve crossing hazardous country. Granted
there is an impressive number of segmented gallery graves along
the west shores of Loch Fyne, to come eastwards from there
involves crossing a 2000' watershed. The route by Aberfoyle and
the Trossachs would have meant traversing the dense forest growing
on the clay lands which bordered the upper reaches of the Forth.
To-day Flanders Moss is an eloquent if minimal reminder of this
forest in decay. The builders of segmented gallery graves were
accustomed to travelling by water, after all they had arrived in
western Scotland by sea, and Loch Lomond would have seemed
infinitely preferable to the hazards, known and unknown of the
Forth valley forest. This penetration into Perthshire from the west
is a reminder of how accessible the area is to migratory movements
using waterways as lines of communication. The Tay and the Earn
and the Tummel are parts of a loch and river system which opens
the way between east and west. There are in effect three routes
across Perthshire. The northern route makes use of the Tay and
the Tummel and then goes by Loch Rannoch and Loch Laidon
across to the head of Loch Etive. The middle way is based on
Loch Tay, Glen Dochart and Glen Falloch to Loch Lomond. The
Earn and Loch Earn are linked to the middle route by Glen Ogle.
These routes are basic to an understanding of Perthshire prehistoric
archaeology.

There are several large circular cairns in Perthshire whose
exact cultural context is uncertain. In Strathearn there is the
example near Almondbank Station*^ which forms the focal point
of the recently recognised ritual complex at Huntingtower. Also
near Almondbank is a great cairn of water worn boulders^- which
together with a recently discovered cairn 90' in diameter and about
15' in height on the farm of North Mains^-^ might well be
chambered. There is a somev/hat smaller example near Strowan
bringing the quota in Strathearn to four. In Strathtay there is a
very large stone and earth tumulus at Balnaguard^"^ and above Blair
Atholl two similar cairns (one with a well marked boulder
peristalith) in Glen Fender and Strathgroy’\ In the absence of
excavation it would be idle to speculate about the date of these
burial mounds. Each year field work adds more examples to the
list, the most recent recruit being a badly ruined site on the out¬
skirts of Perth. Excavation of this expendable example might
provide the necessary information which would place these
monuments in their proper chronological setting.

A number of circular earth tumuli ranging in visible diameter
from 40' - 90' are known from the river valleys of the Tay and the
Earn. In 1964 Dr. Coles excavated one of this group at Pitnacree
in Strathtay. The mound was found to cover an oblong structure

8

Burial Mound at Strathgroy, near Blair Atholl.

of water worn boulders with a lintelled passage entry but previous
to this two massive post holes had held timber uprights associated
with scattered cremation burials. These features have obvious
affinities with the mortuary house revealed at one end of the
Dalladies long barrow near EdzelT^. The pottery from Pitnacree
is echoed at Cultoquhey and not far away at New Fowlis^”^ is
another large earth tumulus externally of the same type as
Pitnacree. These affinities confuse the picture of Neolithic man in
Central Perthshire and another excavation of one of the numerous
earth tumuli in this part of Scotland should be given high priority.

Two ritual sites possibly of the late third or early second
millennium have been recognised in Perthshire. One has been
excavated^^. In 1965 Professor Piggott examined the concentric
settings of standing stones at Croftmoraig near Kenmore. Two
surprising facts emerged. The first was that the earliest period of
the monument had involved the use of wooden uprights which in
subsequent periods had been replaced by monoliths. This initial
wooden structure may have been partly roofed perhaps open at
the centre for here at ground level was a hearth and evidence of
fire. Croftmoraig in fact may have been a provincial version of the
great timber temples or meeting houses now being reconstructed
from evidence like that obtained at Durrington Walls in Wiltshire.
The second surprising fact was that in its third and latest phase
the axis of Croftmoraig had been shifted through 90° from east/
west to north south.

The great ritual complex located by air photography in the
neighbourhood of Huntingtower just west of Perth was described

9

at the official enquiry held to discuss its partial destruction by
industrial development as representing a religious centre for a very
great part if not indeed the whole of Scotland around 2000 B.C.

The second millennium archaeology of Perthshire is rich in
material but surprisingly difficult to put into perspective. There
appear to be several groups of people with distinctive traditions.

The Scottish scene at this period is dominated by the advent
of the Beaker people on the east coast. Coming directly from the
coasts of Holland and North Germany they arrived in considerable
numbers. They were farmers and in north-east Scotland they
occupied the fertile lands bordering the coast and penetrated the
river valleys inland in search of good grazing and cultivable
territory. In their continental homes they had been accustomed to
use the earliest types of bronze tools and weapons. Flat axes and
short knife daggers had been part of their equipment and once
established in Scotland they created a market for these and for the
raw materials and craftsmen who could make them. Their arrival
must have upset the equilibrium of the existing population and
possibly it was the resulting disruption which forced small
indigenous groups to move elsewhere. This may explain some
types of monument and pottery recently identified in the Perthshire
valleys.

Beaker pottery is infrequent in the Earn and Tay valleys but
the example from Bailielands, Auchterarder^^ and the broken pot
from the centre of the “ four poster ” setting at Lundin Farm-® in
Strathtay were both decorated with twisted cord impressions and so
belong to an early group of Beaker migrants to the east coast. On
the other hand the Tippermallo beaker-^ from near Methven is late.
The panelled decoration on its long neck is reminiscent of a similar
design on the fragment from the cist grave at the centre of The
White Cairn in Glen Cochil above Aberfeldy--. These four beakers
together with the degenerate beaker found with cremation at Balna-
guard leave the impression of a sporadic and unplanned penetration
of central Perthshire. The inference is that the upper reaches of
the Tay and the Earn were already occupied and that in spite of
the attractive farming terrain bordering the Earn east of Crieff the
Beaker people had no incentive, perhaps even a positive deterrent,
to the colonisation of this part of the country.

An Appendix to the Lundin Farm excavation report listed
over twenty pairs of standing stones in Central Perthshire. Paired
stones can no longer be regarded as the last vestiges of stone circles
but are monuments in their own right. The Perthshire survey
brought out two characteristics. One was a tendency to be paired
broad and narrow. The other was that pairs thought to be typo-
logically late because of size or siting, instead of being aligned
with their broad faces in the same plane, were turned to face each
other. Recently as a result of excavation to restore a fallen pair of

10

Paired Stones at Dalchirla, near Muthill.

stones at Orwell in Kinross-shire-^ Dr. Graham Ritchie of The
Royal Commission on Ancient Monuments has suggested that
paired stones might be a final memory of the flankers which stand
on either side of the recumbent stone in the circles of north-east
Scotland. The suggestion is important because it can be linked to
three other hitherto puzzling phenomena in Perthshire. The first
is a tendency in certain Perthshire stone circles to emphasise the
south-west quadrant by placing the tallest stones on that part of
the circumference. This is a feature of the Clava type of chambered
tomb as well as of certain ring cairns and recumbent stone circles
of the north-east. In Perthshire it was first recognised at Scone
Wood-"^ where one of two tangential circles was excavated in 1962.
Since then it has been noted in the Moncriefife and Monzie stone
circles and in a recently recognised circle near the Loch of the
Lowes near Dunkeld. The second puzzling phenomenon is that
so-called “ flat rimmed ” pottery which is at home in the north-east
has turned up at Scone Wood. The third phenomenon is the “ four
poster.” This very appropriate name was coined by Dr. Coles to
describe settings of four upright stones which couldn't strictly be
classed as stone circles. In a recent paper to the Society of
Antiquaries of Newcastle-upon-Tyne-'^ Mr. Aubrey Burl has argued
for their origin in the north-east of Scotland. There are several in
Central Perthshire; at Auchnafree in Wester Glenalmond, at Lundin
Farm and below Dull in Strathtay, on the moor above the north
side of Loch Tummel with an eastern outlier at Glenballoch near

11

Craighall Rattray. In some distribution maps prepared by Mr.
Herbert Coutts formerly of The City Museum, Dundee^^ paired
stones and four poster settings occur in the valleys of the Isla and
the South Esk as though the southward movement suggested by
these Perthshire phenomena had followed the Vale of Strathmore.
It all looks like an influx of new people into the valleys of the Tay
and the Earn by the second half of the second millennium B.C.
The Scone Wood stone circle has a Carbon 14 date of 1200 B.C.

Another influence on Central Perthshire in the second
millennium B.C. is associated with the presence of bell barrows.
These occur on the Haugh below Pitlochry^"^, at FortingalP^ and
on flat moorland east of the village of Kinloch Rannoch^^. The
bell barrow is a circular burial mound surrounded by a ditch with
an outer bank, the ditch separated from the mound by a flat berm.
They must have been built by people who were either in touch
with or were influenced by the Wessex culture of Southern England.
Further indications of these people are the collared urns from
Lundin and Dull, the former with its plaited cord ornament having
affinities in south-west England. Because of their widespread
trading connections the power and prestige of the Wessex hegemony
must have been felt far beyond its physical boundaries. Trade in
Irish gold may have encouraged Irish prospectors to visit Western
Scotland in search of metal ores and it would have been easy to
follow the routes eastward which had earlier been pioneered by the
builders of chambered tombs. Gold and copper occur in Strath
Earn, Strath Tay and Glen Ouaich^®.

In the recent Beaker Pottery Corpus, David Clarke has drawn
attention to what has for many years been known as the Balmuick
beaker^k Balmuick is a farm north of Comrie. Clarke has dis¬
missed this curious handled urn with its ribbed decoration from
his beaker lists and claims it is in fact a clay replica of the famous
gold cup from Rillaton in Cornwall^-. Quoting a second parallel
this time between handled clay urns from Monquhitter in Aberdeen¬
shire and Gulval in Cornwall Clarke suggests they reflect a trade
in Cornish tin. It is perhaps apposite to point out that on the
moorland above Balmuick cup marks have been made on a large
number of boulders and rock outcrops. It lends point to Stuart
Piggott’s suggestion that these markings are a magic formula for
locating copper and tin and gold^k

Evidence for settlement towards the close of the second
millennium has come to light since this paper was read and should
be included here to bring the record up-to-date. A cemetery of ten
short cists at Almondbank produced two food vessels from differing
traditions and two simple jet necklaces. The site should be com¬
pared with what remained of a similar group of burials at Tulliemet

12

Yorkshire type food vessel from Tulliemet, Ballinluig.

near Ballinluig. These yielded an Irish bowl food vessel and one,
possibly two of Yorkshire type. The fact that cremation and
inhumation were found at both sites underlines the extent of the
mixed traditions already shown by the pottery.

The early prehistory of Perthshire is emerging. What is needed
is selective excavation to prove or disprove some of the hypotheses
now being put forward.

* * *

References

1. “ Tayside before History.” Herbert Coutts. A Dundee Museum and
Art Gallery Publication. 1971. pp. 9 and 10.

2. “Prehistoric Europe.” Geikie. London. 1881. p. 392.

3. The General’s Review. Spring, 1956. pp. 16 and 17.

4. Discovery and Excavation Scotland. 1964. p. 43.

5. “The Chambered Tombs of Scotland.” Vol. IT. Audrey Shore Henshall.
Edinburgh University Press. 1972. pp. 428-479.

6. Discovery and Excavation Scotland. 1968. p. 28.

7. “ Megalithic Enquiries in the West of Britain.” Liverpool Symposium.
Liverpool University Press. 1969. pp. 180-182.

8. See Ref. 5. p. 172.

9. Proc. Preh. Soc. Vol. XXXI. 1965. pp. 34-48.

10. Proc. Preh. Soc. Vol. XVII. 1951. p. 35.

11. Nat. Grid Ref. NO 069249.

12. Nat. Grid Ref. NO 072272.

13. Nat. Grid Ref. NN 927164.

14. Nat. Grid Ref. NO 945518.

15. Nat. Grid Ref. NN 903681 and NN 898652.

16. Discovery and Excavation Scotland. 1972. p. 55.

13

17. Nat. Grid Ref. NN 928233.

18. Proc. Preh. Soc. Vol. XXXVII. Part 1. July, 1971. p. 1.

19. “ Beaker Pottery of Great Britain and Ireland.” D. V. Clarke. Cam¬
bridge University Press. 1970. Vol. II. No. 17. p. 282.

20. Proc. Soc. Ant. Scot. Vol. XCVIII. 1964-65 and 1965-66. p. 126.

21. See Ref. 19. No. 609. p. 354.

22. Proc. Soc. Ant. Scot. Vol. XCII. 1958-59. p. 81.

23. Discovery and Excavation Scotland. 1972. p. 55.

24. Proc. Perthsh. Soc. Nat. Sc. Vol. XI. 1963-65. pp. 7-23.

25. Archasologia Aeliana. XLIX. 1971. pp. 37-51.

26. See Ref. 1.

27. Nat. Grid Ref. NN 948573.

28. Nat. Grid Ref. NN 734467.

29. Nat. Grid Ref. NN 668585.

30. Proc. Preh. Soc. 1951. Part 1. pp. 69-71.

31. See Ref. 19. Vol. II. No. 1081. p. 417.

32. See Ref. 19. Vol. II. Frontispiece.

33. “ Scotland before History.” Stuart Piggott. Published by Thomas
Nelson. 1958. pp. 60-61.

/

14

Roman “ Signal Stations ” on the Cask Ridge

Anne S. Robertson, Hunterian Museum

Introduction

The Cask ridge is the northern ridge of the valley of Strath-
earn, Perthshire. It rises from the Creel, or Innerpeffray ford of
the River Earn, at about 100' (30 m.) above sea level, to a maxi¬
mum height of c. 470' (143 m.).

Along this ridge a Roman road ran slightly north of east for
a distance of 8 miles. Earlier observers traced several stretches
of this road, noted widths varying from 10' (3 m.) to 24' (7.3 m.),
and remarked that at some points it was raised on a causeway k

More recently, O. G. S. Crawford has followed and recorded
the line of the Roman road eastwards from the fort of Strageath
on the west bank of the Eam^, and the Ordnance Survey have
marked it on their maps (e.g., 1" No. 55). Within the last half
dozen years or more, members of the Archaeological Section of
the Perthshire Society of Natural Science have carried out frequent
surveys of the line of the road, and have generously placed their
knowledge at the disposal of visiting archaeologists. No one who
has had the privilege of walking along the Roman road on the
Cask ridge under the guidance of members of the Archaeological
Section will forget the enlightening, thrilling experience, and the
magnificence of the prospect to the north and south. The view
northward in Roman times would have encompassed the Flavian
fort of Fendoch with its small watch-post at the mouth of the
Sma’ Glen; the view southward would have included the fort of
Ardoch and the “ crag and tail ” on which Stirling Castle now
stands.

The Archaeological Section have also cut a section across the
Cask road in Parkneuk Wood (NN 915185) under the direction
of Mr. James Thomson, Curator of the Smith Art Gallery and
Museum, Stirling. The road had a maximum width of 19'
(5.25 m.), and was surfaced with a cambered layer of river gravel.
A central rib 12' (3.65 m.) wide, had been formed of boulders
and large, flattish stones, over a layer of turves. On either side

15

of the central rib, stones had been embedded in a cambered layer
of clayey-loam^.

Earlier observers had also recorded several small circular
earthworks on the line of the Gask ridge road'^. Dr. David
Christison did in fact list seven of these (now identified as at
Parkneuk, Raith, Kirkhill, Muir o’ Fauld, Gask House, Witch
Knowe, and Moss-side or Findo Gask. “ They are all,” remarked
Christison, “ much alike in size and structure, the diameter over
aU averaging about 100', and the plan showing an inner flat area,
generally without any sign of a rampart or palisade at the edge,
and a rather shallow trench, the spoil from which has been thrown
outwards, forming a low mound.”

Excavation was carried out in 1900 on five of the circular
posts — Kirkhill (NN 967188), Muir o’ Fauld (NN 981189), Gask
House (NN 990191), Witch Knowe (NN 997195), and Moss-side
(NN 007199). The plans of two only were published — Witch
Knowe and Moss-side. At Witch Knowe four round post-holes
were found nearly in the centre of the internal platform, which was
44' (13.4 m.) across. Each post-hole was about 1.5' (.46 m.) in
diam., and 2' (.61 m.) deep. The post-holes, measured from centre
to centre, delimited a rectangular area c. 11' (3.35 m.) from east
to west by 9' (2.74 m.) from north to south. Scraps of charred
wood came from them. The post-hole area was encircled by a
ditch about 14' (4.27 m.) wide and 6' (1.82 m.) deep, with an
opening or causeway 6' (1.82 m.) wide in the south side, facing
the Roman road about 300' (91 m.) away. Upcast from the
ditch appears to have been thrown out on the outer side to form a
low mound only 2' (.61 m.) high.

At Moss-side, there was a level expanse in the centre about
22' (6.15 m.) in diam. “In it, but not in the centre, were four
‘post-holes’ in a square formation, with a base of 11' (3.35 m.)
measuring from the centres of the holes, but unlike the other
works in which ‘ post-holes ’ were found three of these holes were
connected by flat cuts, probably to hold beams.” The post-holes
at Moss-side were over 2 — 2.5' (.61 — .76 m.) in diam., and over
3' (.91 m.) deep. The circular interior, 22' (6.15 m.) wide, was
enclosed by a mound which in its surviving form was 14' (4.27 m.)
wide and 3' (.91 m.) high, made up of layers, some black (which
were probably turf). The outer encircling ditch was ploughed
out, but there was an interruption in the rampart for a causeway,
facing south towards the Roman road, c. 210' (64 m.) away.

The other three circular posts which were excavated, but
without published plans — at Gask House, Muir o’ Fauld and
Kirkhill — all lay to the south of the Roman road, at about 30',
45' and 150' (9.1, 13.7 and 45.7 m.) distance south of it respect¬
ively. In each of the three, four post-holes were located, in each

16

case defining an area averaging 9' (2.13 m.) square. At Kirkhill,
the encircling ditch was 14' (4.27 m.) wide, and 3' (.91 m.) deep;
at Cask it was again 3' (.91 m.) deep.

A sixth circular post, at Raith (NN 932185), lying 300'
(91.4 m.) above sea level, and 600' (183 m.) south of the Roman
road, was discovered in 1901, during the construction of a water-
tank. Four post-holes, each about V (.31 m.) in diam., were
found, going down about 1' (.31 m.) into rock. They defined an
area about 9' (2.13 m.) square. A seventh circular post, at
Parkneuk, also lay south of the Roman road, but was not investi¬
gated in 1900-1901.

Raith was the only one of the seven posts which produced
pottery. “ Some broken pieces of red pottery were also found,
but not of a well defined character.” These have not survived.
From all six sites (i.e., all but Parkneuk, NN 916184) on which
trenching took place fragments of charred wood were recovered,
of oak, willow and hazel.

The conclusions reached by Dr. David Christison were as
follows:

“ That the posts are intimately related to the road is proved
by their being aligned along and close to it, by their entrances being
towards it whether they lie to the north or south of it, and by their
identity of plan and structure. Their defensive strength as earth¬
works is but feeble, the trench being in all but one not more than
Sy deep, and the mound outside only a foot or two above the
exterior. Neither did we find any evidence of a rampart or other
defence round the inner area, except in No. 6 (i.e.. Moss-side)
which had a rampart 2' or 3' high, composed of layers of sod and
clay. The key to their purpose appears to be the four post-holes
found in the centre of the area of the five that we excavated, and
which in all probability held the supports of wooden towers about
9' to 10' square. The thick layer of black mould over the whole
inner area may have proceeded from the decay of other wooden
structures as well as of the fallen towers, but no other holes or signs
of foundations were found, although narrowly searched for. For
defensive purposes the small supposed towers could have held
very few men. It seems probable, therefore, that they were mainly
watch-towers, and this purpose seems the more likely, as the post
at Raith appears to have been intentionally placed further from
the road than the others, in order to have the advantage of an
extensive view''.” Dr. Christison also discussed a large rectangular
enclosure just south of the circular post at Cask which he identified
as “ no doubt Roman^.”

Almost half a century later, O. G. S. Crawford described the
course of the Gask ridge road, and its accompanying structures, in
his Topography of Roman Scotland North of the Antonine Wall,

17

1949, 51 If. By then, the number of “signal stations” identified
had been increased to nine, through the discovery by Crawford
himself of Ardunie (NN 946187) in 1937, and of Thornyhill,
Midgate (NO 021204), in 1940. Ardunie is on the south side of
the Roman road, Thornyhill on the north side.

Subsequently a tenth “ signal station ” on the Cask ridge was
discovered from the air, by Dr. St. Joseph, to the north of the
Roman road, on Roundlaw farm, between Ardunie and Kirkhill.
Air photographs revealed a circular enclosing ditch, with the marks
of four post-holes inside. A similar setting of post-holes within an
encircling ditch was also observed on aerial photographs of
Moss-side'7.

Dr. St. Joseph also recorded the re-discovery of the Roman
camp at Cask, south of the “ signal station.” It measured c. 500'
(152 m.) from east to west, and c. 425' (130 m.) from north to
south. To the east of the most easterly “ signal station ” hitherto
known (Thornyhill, Midgate), Dr. St. Joseph saw traces of the
Roman road running eastwards, with, at almost 3000' (915 m.) east
of Thornyhill, a circular crop-mark similar in size to those at
Roundlaw and Moss-side. This site (Westmuir) would have
Thornyhill (Midgate) in view, although without outlook to the
north. It may have been an eleventh “ signal post.” (See map.
Fig. 1.)

A useful list of ten posts from Parkneuk to Thornyhill (Mid¬
gate), with their distances apart was published by A. L. F. Rivet

Fig. 1.

18

in Arch. J. 1964, 196 ff. His distances vary slightly from those
given by the Archaeological Section of the Perthshire Society of
Natural Science, and by some other field investigators, from Dr.
David Christison (c. 1900) to Charles Daniels (c. 1970). In general,
however, it may be said with confidence that (if our series is com¬
plete) no two such posts were less than half a Roman mile apart
or much more than a Roman mile apart. The average distance
was in fact about four-fifths of a Roman mile. They were all (or
almost all) on high ground, and mutually intervisible.

1. Gask House

In 1965, the Archaeological Section of the Perthshire Society
of Natural Science initiated a project to excavate or re-excavate
at least one of the Gask ridge “ signal stations ” in the hope of
finding dateable evidence. Gask House was the site chosen, since
although it had been investigated briefly by Dr. Christison, further
trenching near the ditch-end might, it was thought, bring to light
some pottery or other dating material. Accordingly, in 1965, a
trial trench was opened up, and further excavated in 1966, by kind
permission of Major R. N. Jardine Paterson and Mrs. Jardine
Paterson, of Gask, and with the sanction of the then Ministry of
Public Building and Works. The work was supervised by Mr. James
Thomson, Curator of the Smith Art Gallery and Museum, Stirling,
and myself.

Since the ditch of the Gask “ signal station ” and the bank
inside it were visible on the ground, a surface survey was first made,
and a trench (Tl), 4' (1.22 m.) wide, was then laid out from the
ditch near its end on the east side of the north causeway, south¬
wards across the site. The trench^, 80' (24.4 m.) long, was aligned
slightly askew to the presumed axis of the signal post in the hope
of locating at least one of the internal post-holes. A post-hole was
in fact found in Trench 1, and two other smaller trenches (T3 and
T4) revealed two other post-holes. All three were over 2' (.61 m.)
in diam., and went down 2' (.61 m.) into subsoil. Charred wood
and some packing stones were present in each, and in two, in
Trenches 1 and 4, there was a flat stone on the bottom. The pre¬
sumed position of the fourth post-hole has been dotted in on the
plan (Fig. 2).

These four holes must have held the four great corner posts of a
wooden tower, whose dimensions may be given as about 10'
(3.05 m.), from east to west, and 8' (2.44 m.) from north to south.
It depends of course on whether measurements are taken from
centre to centre of the post-holes, or from their inner or outer edges.

The post-hole first found in the 80' (24.4 m.) Trench 1 was
partly under the east side of the trench, so that a small trench (Tie)

19

was cut eastwards to reveal the entire post-hole. The eastern tip
of Trench le located the inner edge of the encircling bank, and
established a point on its circuit.

Trench 1 also of course revealed the remains of the bank. It
was 9' (2.74 m.) wide, and had a layer of turf at the bottom, and
clay upcast from the ditch above. The turf had been laid and
presumably represented stripping of the surface before the ditch
was dug. The bank, with its turf underlay, survived to a height of
only T (.3 m.).

20

At the north end of Trench 1, the bank was evidently coming
to an end within the width of the trench, and had run out altogether
on the south-west side of the trench. The points established on the
inner and outer edges of the bank, and the position of the four post-
holes indicate that the area enclosed by the bank was not circular,
but rectangular with rounded corners. It measured 22' (6.7 m.)
from east to west, and 18' (5.49 m.) from north to south. Within
this sub-rectangular area the wooden tower framed by the post-
holes was set back from the centre, leaving a space of about 2'
(.61 m.) between the two southerly post-holes and the bank, and
about 5' (1.52 m.) between the two northerly post-holes and the
bank. Such a position would provide more space for men entering
and leaving the tower.

The encircling ditch on the other hand was shown by its visible
Trench 1 to have been more nearly circular, with east and west
sides somewhat flattened. It was roughly V-shaped. Its depth
averaged about 3' (.9 m.) from the top of the clay subsoil. Its
greatest width was on the south side — 11' (3.35 m.) — while its width
on the east and west sides was about 9' (2.7 m.), measured on the
surface. It narrowed to 6' (1.8 m.) in the section at the north end
of Trench 1. The ditch was clearly beginning to run out at the
north end of Trench 1, and its precise ending was located in an
east-west trench (T2) cut across the visible causeway between the
ditch-ends. The causeway (or more accurately the strip of ground
left undisturbed when the ditch was dug) was less than 9 (2.7 m.)
wide. There were some stones trampled into the surface of the
causeway and on the ditch-edges. There was natural silting in the
ditch.

The axis of the “ signal station ” is shown by the position of
the post-holes, and by the gaps in mound and ditch, to have been
slightly west of north. It was, in other words, tilted north-west, as
if to face roughly in the direction of the fort and watch-post at the
mouth of the Sma’ Glen.

The finds were scanty in the extreme — only a few flecks of
charred wood, three iron nails (much corroded), and a mortarium
rim. (Fig. 3, drawn by Mrs. Margaret Scott.) This was found
unstratified in dark soil about 2' (.61 m.) behind the bank in Trench
1 (at a point marked X). Fortunately the mortarium was of a
distinctive type. Mrs. Katharine Hartley most kindly supplied the
following note on it:

“ The Gask mortarium is particularly useful as both date and
origin are beyond doubt. It is in every way a typical Flavian
product of the extensive potteries near Watling Street, in the area
south of Verulamium, and including Brockley Hill and Radlett.
The rim-form can be most clearly matched in the work of Albinus
and it could well be one of his products, especially as he made more

21

mortaria than any contemporary potter. His activity can be dated
c. A.D. 70-95, and there can be no doubt that the Gask mortarium
was made within this period.”

\

\

Fig. 3: Mortarium rim. Drawn by Mrs. Margaret Scott. Scale 1/1.

2. Parkneuk

Although the mortarium rim from Gask House provided a
Flavian date for that site, it still seemed advisable to secure more
dating evidence from the Gask ridge. Accordingly, in 1968, a
small-scale excavation was carried out at Parkneuk, the most
westerly “ signal station ” known.

At Parkneuk, trenching was once again under the direction of
Mr. James Thomson and myself, with help from Dr. John
Mackenzie, now at the University of Lancaster, and with a digging
force of four Sixth-Formers from Daniel Stewart’s College, Edin-
urgh, and of several indomitable members of the Archaeological
Section of the Perthshire Society of Natural Science. Parkneuk
lies in woodland owned by the Forestry Commission, which readily
gave permission for excavation, and much assistance, greatly
appreciated. Sanction was again given by the then Ministry of
Public Building and Works.

The encircling ditch, and a bank inside it, were (and still are)
visible on the ground, although both have been much distorted by
tree-roots, and the ditch choked with reedy growth and part-filled
with water. A surface survey was therefore made, and then a
trench (Tl), 80' (24.4 m.) long and 4' (1.22 m.) wide was laid out
north-south at the one point where it could be directed right across
the site from ditch to ditch without striking a tree. In the end,
one tree-stump did cause an interruption 6.5' (1.98 m.) long, at
12.5' (3.81 m.) froih the north end of Trench 1 (Fig. 4).

Trench 1 showed the ditch to have been V-shaped, and 12'
(3.66 m.) wide, but its precise depth could not be determined, owing

22

PARKNEUK 1968

10

^ METRES

A SR
yllS

Fig. 4.

to its waterlogged state. It was probably 3' — 4' (.9 — 1.2 m.) deep.
There was a causeway, 8' (2.44 m.) wide, visible in the north side of
the ditch. The ditch enclosed an area nearly circular, but with east
and west sides slightly flattened.

Trench 1 also revealed the remains of a bank on the inner
side of the ditch, formed mainly of red clay upcast from the ditch.

23

It was 9' (2.74 m.) wide, and appeared to have been laid directly
on the subsoil or on Roman humus. In the southern sector of
Trench 1, the south bank was found to have its south edge straight,
not curved. The inner edge of the south bank curved round slightly,
and here there were several layers of turf, as if turf had been
stripped off the surface and laid at the rear line of the proposed
bank.

Since Trench 1 had had to be laid out askew to the north-south
axis of the “ signal station,” in order to avoid trees, it picked up
the south bank, part of the west bank, the curve at the north-west
corner, and the north bank. The material of the bank at all points
examined was of red clay upcast from the ditch. At the point
where the inner edge of the west bank was encountered, there had
apparently been a hollow in the subsoil (X), which had been filled
(or had filled naturally), with grey-white clay before the red clay
bank had been put down.

An eastward projection of Trench 1 disclosed one post-hole,
about 2' (.61 m.) in diam., and going down at least 1.25' (.38 m.)
into the subsoil. A continual inrush of water made it impossible
to reach the bottom. In Area 2, a second post-hole was found, and
a third in Trench 3. These were also 2' (.61 m.) in diam., and went
down at least 1.6' (.53 m.) into the subsoil. The presumed location
of a fourth post-hole has been dotted in.

Trench 3 also located the tip of the clay bank on the south-east
side of the entrance gap, and part of the entrance causeway. The
opening in the clay bank was 7' (2.13 m.).

To judge from the four post-holes which must have held the
corner posts of a wooden tow^er, the tower was about 10' (3.05 m.)
from east to west, and IT (3.35 m.) from north to south. These
measurements were taken from centre to centre of the post-holes.

The tower had stood within a level area measuring 22'
(6.71 m.) from east to west, and 18' (5.49 m.) from north to south.
This area was sub-rectangular, enclosed by a bank or rampart
which apparently had four short straight sides and rounded corners.
Unlike Cask House, the Parkneuk tower stood almost centrally
within the bank. There was only 2' (.61 m.) between the Parkneuk
tower and the north and south banks, but 5' (1.52 m.) between the
tower and the east and west banks. The plan gives the impression
that the Parkneuk “ signal station ” was tilted slightly west of north
like that at Cask Flouse, but in fact the axis was almost due north,
facing directly towards Fendoch.

There were no Roman finds (except minute scraps of charred
wood) from Parkneuk, but its plan is virtually a blueprint of the
Cask House plan.' The two must have been contemporary, that is,
of Flavian date.

A surface find of flint has been described as follows by Dr.
E. W. MacKie, Hunterian Museum:

24

“ The fragment is part of a large waste blade of white-patinated
flint which has been struck from a prepared core: the striking
platform and the bulb of percussion are visible as also are, on the
opposite surface, the scars of previously struck blades. There is
no sign of secondary working so the flint is not an implement. It
has been severely heated after being detached from the core:
thermal fractures are visible on all the faces.

“ It is not possible to give the flint an accurate cultural context
or date, since flint working of this type was practised in Mesolithic,
Neolithic, Bronze Age and Early Iron Age times in Scotland.”

3. Roundlaw

In 1972, opportunity offered itself of conducting a small-scale
excavation at Roundlaw, since the field in which it stood was in
that year under pasture. Permission was kindly given by the
owner. Sir James Denby Roberts, Strathallan Castle, and by the
tenant, Mr. J. Burt, Roundlaw Farm. The week’s work was
directed by Dr. John Mackenzie, University of Lancaster, and
myself, with help from Miss Kay Docherty (Hunterian Museum),
Miss Gillian Gibson (St. Andrews University), Graham Magauran,
M.A. (Glasgow University), Gordon Thomas (Edinburgh University),
and members of the Archaeological Section of the Perthshire Society
of Natural Science and of the Strathearn Archaeological Society.
Roundlaw was not scheduled at the time of the excavation — Sept¬
ember, 1972 — but has since been scheduled.

The aerial photograph supplied by Dr. St. Joseph showed a
circular ditch with an opening to the south, and apparently four
post-holes inside^. (Figs. 5, 6.) There are no surface indications
whatsoever.

The first trench laid out at Roundlaw was sited a little too
far west, since the “ signal station ” proved to have stood a little
way down the slight eastern slope of a small hump, instead of on
the highest part of it^®.

Trench 2, however, 84' (25.7 m.) long and 4' (1.22 m.) wide,
soon revealed the west ditch and, as it progressed, the east ditch of
the “ signal station.” The site was of solid rock, and since the
excavation followed several weeks of drought, the digging of this
“ signal station ” was the hardest physical work yet required on
the Gask ridge.

The reward came with the clearing out of the ditches in Trench
2. They were not the conventional V-shaped ditches, but were of
the “ Punic ” type, that is, there was an almost vertical drop on
the outer side, or counterscarp, while the inner side or scarp sloped
gently upwards from a narrow, flat bottom. The ditches sectioned
in Trench 2 were 1' (2.13 m.) wide, and had been hacked down to
a depth of over 4' (1.22 m.) into the solid rock.

25

The west and east ditches located in Trench 1 were 47.5'
(14.5 m.) apart, and in the space between them there was no sign of
a bank in situ. The rock surface was only about 8" (.20 m.) under
the surface, and the plough in its work over the centuries must have
skimmed oflP the upper part of any bank that had been present.

26

The nature of the bank was, however, made clear by material
in the east ditch. There was natural silting in the bottom to a
maximum depth of 2' (.61 m.), and above that fallen turf which
had evidently slipped into the ditch from a bank to the west of it.
The west ditch in Trench 1 contained no turf slip, only natural
silting. The reason for this was the slight slope in the ground
from west to east, which allowed material to slip eastwards, but
not westwards.

The use of turf for the bank of Roundlaw “ signal station ”
was made necessary by the unsuitable nature of the material dug
out of the solid rock in making the ditch. This material must have
been in the form of rubble and small stones, which would make
a very unsatisfactory, unstable bank. The bank must either have
had turf cheeks retaining a central core of rubble, or else have been
formed entirely of turf stripped from the surrounding area.

The perimeter of the encircling ditch was further determined
in Trench 3, to the north-east of Trench 2. A well-preserved
“ Punic ” ditch-section was again revealed, with the expected width.

Fig. 6.

27

7' (2.13 m.) and depth, T (.61 m.). Natural silting filled the lower
T (.61 m.) of the ditch, and above that there was fallen turf, and
earth with flecks of turf, the latter from wash-down into the ditch.

The interruption in the south side of the ditch for a causeway
was located in Trenches 4 and 6. The sharp “ Punic ” outline was
perfectly preserved quite close to the ditch-end on the east side of
the causeway, and there were turf flecks in the dark earth silting
of both ditch-ends. The causeway was only 6' (1.83 m.) wide.

No post-holes were present in the main long trench (T2), but
three were found in Trench 5, south-east of Trench 2, and two others
in two small squares (T7 and T8) cut on the north side of Trench 2.
Yet another trench (T9) just missed a fourth post-hole which must
have been located beyond the southern edge of Trench 9.

Since they had been chipped out of solid rock, the post-holes
were not of the conventional round shape, but had very jagged
edges. They were all about T (.61 m.) across, and probably T
(.61 m.) deep.

The wooden tower whose four corners were marked by the
post-holes seems to have been 10' (3.05 ni.) wide from east to west,
and as much as 14' (4.27 m.) long from north to south, measuring
from centre to centre of the post-holes. The tower in that case
stood much closer to the south ditch (i.e., on the causeway side),
the distance being about 14' (4.27 m.) than to the north, east and
west ditches, the distance in these cases being about 17' (5.18 m.).
No evidence was recovered for the width of the encircling bank,
but it may have been about 9' (2.74 m.) wide, as at Cask House
and Parkneuk.

Roundlaw “ signal station ” would have had Fendoch in view
had the outlook been to the north, but in fact it faced south over
Strathearn. It stood about 140' — 150' (42 — 46 m.) north of the
Roman road'‘. Presumably its southward link was with Ardoch,
or with other stations on the Roman road running south from
Strageath past Kaims Castle to Ardoch.

The only finds from Parkneuk were a few tiny flake and quartz
chips, wood scraps and iron nails. The two last came from the two
northern post-holes.

Summary

There have been identified on the Cask ridge, average height
200' — 300' (60 — 90 m.) above sea level, at least ten Roman “ signal
stations,” set on the line of a Roman road running eastwards along
the ridge from the -fort at Strageath in the direction of Perth, 14
to 15 miles away. These ten “ signal stations ” cover the eight
miles east from Strageath, leaving a gap over six miles long between
the most easterly of them (Thornyhill, Midgate) and Perth.

28

It is, however, possible that an eleventh “ signal station ” may
be identified at Westmuir, over half a mile east of Thornyhill,
Midgate, where a circular crop mark has been observed from the
air. This mark is similar to the circular ditch round the certain
“ signal stations,” for example, at Moss-side and Roundlaw.

The ten certain “ signal stations ” took the form of wooden
towers, averaging 10' (3.05 m.) square, probably with upper floors,
and a flat roof, or a balcony round the top storey from which signals
could be sent. The total height may have been not less than 20'
(6.1 m.).

Each tower was enclosed by a bank consisting mainly of
material dug out of the encircling ditch. Although the ditch is
circular on plan, or on an air photograph, the bank may have
flattened sides, so following the rectilinear outline of the tower
rather than the circle of the ditch.

Each “ signal station ” had one entrance-causeway, facing to¬
wards the Roman road. The distance from the road varied from
c. 30' (9 m.) at Cask House to c. 600' (183 m.) at Raith. In each
case the site had apparently been chosen to give an uninterrupted
view from one ” signal station ” to the two on either side.

Six “ signal stations ” stood on the south side of the road, and
four on the north side. Most of them had in view the fort and
“ signal station ” at Fendoch at the mouth of the Sma’ Glen, and
possibly the fort at Ardoch.

The only dateable evidence from the Cask ridge itself is a
mortarium rim of unequivocal Flavian date from Cask House. The
similarity in plan of the ten known ” signal stations ” places the
whole system in the period.

The fort at Strageath had several phases of occupation, one
of which at least is known to have been Flavian^^. The earliest fort
at Ardoch was of Flavian date. The temporary camp at Dornock^^,
over a mile up the River Earn from Strageath, and the camp south
of the “ signal station ” at Cask House are of a small size and
therefore perhaps more likely to be of Flavian than of later date.
On the other hand, the two camps west and south of Parkneuk^^,
and the camp at Broomhill-Forteviot, south-east of Thornyhill,
Midgate 16, are of the larger size, possibly assignable to Severus’
campaignsi"^.

In any case, the Gask ridge was in use at some time during the
Flavian period, but at what point within that period is not deter¬
mined. From the fort at Strageath a substantial road ran eastwards
towards Perth, and ultimately to the legionary fortress at Inchtuthil,
and the auxiliary forts at Cardean and Stracathro. The Gask ridge
sector of this road (and it may be other stretches of the road) was
studded with wooden watch-towers each set within a bank and

29

encircling ditch. They were intervisible, and also had a wide out¬
look to north and south as well as along the Roman road. Each
could have held a few men only, whose base must have been at a
permanent fort, perhaps Strageath, and whose function must have
been to maintain a patrolling watch or look-out. They could also
have received and sent on signals. Whether or not the Gask ridge
can be correctly designated a limes, as has been suggested in recent
times, the name “ signal station ” given to the small, circular posts
75 years ago may be allowed to stand.

* *

References

1. R. Sibbald, Historical Inquiries concerning Roman Antiquities in
Scotland, 1707, 37; T. Pennant, Tour in Scotland, Mil, Vol. II, 1776, 90;
D. Christison, “ Excavations Undertaken by the Society of Antiquaries
of Scotland of Earthworks Adjoining the ‘ Roman Road ’ between
Ardoch and Dupplin,” PSAS, 1901, 15 ff., hereafter referred to as
Christison.

2. Topography of Roman Scotland North of the Antonine Wall, 1949,
51 ff.

3. Discovery and Excavation, Scotland, 1967, 37; more detailed publication
forthcoming.

4. PSAS, 1898, 433 f.; Christison, 25 ff.

5. Christison, 39 f.

6. Christison, 41.

7. JRS, 1955, 87; 1965, 81 f.

8. The site had been enclosed at some time by a fence, and privately
planted with young trees. These have since grown so thickly that the
“ signal station ” is now almost inaccessible.

9. JRS, 1955, 87; 1965, 81 f.

10. JRS, 1955, 87.

1 1. Not 240' north of the road as stated in JRS, 1955, 87.

12. JRS, 1958, 90; Discovery and Excavation, Scotland, 1973, 42.

13. JRS, 1965, 81; 1973, 229.

14. JRS, 1965, 81 f.

15. JRS, 1958, 90.

16. JRS, 1965, 82.

17. JRS, 1969, 114 ff.; 1973, 230 ff.

30

PERTH

The first century of the burgh

A. A. M. Duncan

The period with which this paper deals is the middle century
or so of three hundred years of expansion in the economy of
western Europe. From about the year 1000 an increase in popula¬
tion was accompanied by a greater food production, a diversification
of manufactories and an increasing specialisation of function which
made available to both peasant and landlord a wider range of
goods more expertly produced. An economic and social
phenomenon of this kind can know no national boundaries and
those who would seek to write the history of Scotland without
taking account of such economic developments are likely to mis¬
represent the society as more unique than it really was and to give
it the curiously old-fashioned look of a set of institutions without
people. In the twelfth century the pace of expansion quickened and
took an even more dramatic upturn about the year 1180. One
marked feature of this period of change in both England and
France was the development of towns as increasingly busy centres
of trade and manufacture. In Scotland we may seem to be much
disadvantaged by the absence of good evidence for urban life in
the first half of the twelfth century; it is indeed the case that there
is no charter making a grant to the burgesses of a burgh earlier
than the 1160s in date, but the many surviving grants to religious
houses, monasteries or cathedrals, show that burghs were in
existence at Berwick and Roxburgh before 1124 and at various
other places, including Perth, not long after that datek The absence
of early twelfth-century charters to burghs is attributable not to
their being lost but to the fact that the town had no legal persona
to receive a charter. Within it there were burgesses paying their
rents to the king’s representative, the sheriff or to his agents, the
grieves or provosts. The king created a burgh when he recognised
that the inhabitants of the town were burgesses, enjoying the
personal liberties associated with that term, but most importantly
the right to devise their property, whether moveable or landed; but
there was no one to whom he might give a charter (which was the
record of the act of creation but not the act itself). I take my stand

31

then with those who regard David I as the creator of many burghs
in the sense that he gave the privileges necessary for the further
development of urban life on these particular spots though not by
charter. It is the purpose of this paper to see how far we can draw
a profile of Perth in the century after David I mentions its
existence, say from 1130 to 1230, but I shall not hesitate while
doing so to indicate how the missing lines might run on the analogy
of urban history elsewhere. The first and much the most important
of these uncertain lines is the earliest development of the town.

Throughout western Europe it is usual to find towns, and
especially those towns which flourish, at nodal points of com¬
munications. They may also be beside a fortified place or a large
religious establishment but it is usually an open question whether
the routes attracted the abbey or castle, or vice versa. In Scotland
early roads Jed to the castles built on such natural rocks as Edin¬
burgh and Stirling, at the latter of which there was an important
river crossing. At Inverkeithing the routes from East Fife and
Perth met before crossing the Forth by the Queen’s Ferry. At
Montrose there was a ferry to cross the mouth of the basin. At
all these places and at many others where burghs were granted, an
intersection of routes, and especially water borne and land routes,
was clearly of great importance in encouraging the growth of a
settlement. Now we may accept the silence of the sources about
settlements at these places before the time of David I and argue
that they were created by him de novo as towns as well as burghs.
On the other hand, we may notice that there are no sources before
the reign of David I, or at least none of Scottish origin, and that
therefore silence about these places argues nothing at all. When
we do hear of them it is because the religious are anxious to obtain
houses within their bounds and the patronage and profits of the
churches associated with them, a fact which is best explained if
they were already flourishing towns when they became burghs, with
a history of some decades, if not centuries, behind them. At least
when we are writing of urban origins we should distinguish between
the town which is a settlement and a way of life on the one hand,
and the burgh which is an agglomeration of privileges designed to
further that way of life, on the other. It must surely be plain that
David I did not choose the spots and create the communities ah
initio like so many twentieth-century new towns.

The picture of Scotland as a country held down by royal
castles garrisoned by sheriffs intent in protecting imported
merchants in fortified burghs is a curiously unreal way of explain¬
ing the coincidence of towns, castles and sheriffs. The fact is that
the sheriff was a demesne agent of the king, responsible for the
collection, storage,' consumption and sale of produce from the
king’s estates in a certain province. His castle was the security
point for housing produce-rents and was probably also a royal hall

32

and kitchen. And because the king was a sensible man who knew
that he would have surplus produce to sell and a need for skilled
manufacturers, he placed his hall, castle and sheriff at the place,
be it natural harbour, ford or bridge, where there was a community
specialising in production, selling and buying. In early urban
history town and castle were mutually attracted.

The twelfth-century kings resided frequently at Scone, an
ancient cult centre with a new abbey. At Perth they had a house
and a castle, and although there is no direct statement to this effect
it is highly probable that abbey and town acted complementarity
as a royal centre, a hall and palace to which the king’s rents in
produce were brought for the consumption of his court-. At Perth
the river Tay became shallow enough to ford, while, at the same
time, deep enough at high tide to bring in ships; it was the natural
junction of land and sea routes and it is therefore not surprising
that it appears by 1128 as a burgh whose church, already in
existence, is granted away to the grasping Benedictines of Dun¬
fermline abbey^ It may at first sight seem more surprising that
this burgh seems to have yielded more in revenue than any other
in Scotland. Yet even before David I became king, before, that is,
Perth was recognised as a burgh, Alexander I told the merchants
of England that the canons of Scone might have a custom-free
ship to reach them without impediment at Scone"^; it was a nice
idea though difficult to carry out without deepening the river. It
does at least show that by 1124 foreign traders were sailing up the
Tay at least as far as Perth. Then in order to sustain his
programme of monastic foundations David I had to give money
for the initial monastic colonies, money taken from town revenues,
but overwhelmingly from those of Perth, all the more surprising
since the monasteries did not lie around the Tay. A mark to St.
Andrews priory, then fifty shillings more, fifty shillings to Cambus-
kenneth abbey, sixty-six shillings to Dunfermline abbey, forty to
the abbey of Tiron in France, and a hundred shillings to Holyrood\
Much of this came not from the burgh rents but from the customs
charged on ships coming for business and breaking bulk — one ship
yielding a notional fifty shillings custom. If perchance the ships
did not come to Perth then the canons of Holyrood were to obtain
their hundred shillings from the rent of Edinburgh, Perth and
Stirling. Other burghs yielded rents, but Perth far outreached them
in ship-customs. Such evidence implies that the Watergate of Perth
(with the wharves along the river) was already in existence for
some at least of its present length between High Street and South
Street.

Behind it lay St. John’s kirk, attested as early as 1128 not long
before the chapel of the castle^. Each would lie in the place at
which it is later found, on either side of the High Street, of which
we have evidence when Baldwin the Fleming was given a small

33

piece of land, ten feet by twenty-four, next to the house of UilchiL
A few years later in the 1160s this passed to St. Andrews priory as
“ a booth in North Street at the castle,” and so we can place it
exactly at the intersection of High Street and Skinnergate^. The
castle site is delineated by the Castle Gavel on early maps and
stood roughly where the Art Gallery now is. Charters will take
us a little further than this even in the twelfth century; to be
precise a charter granted by King William to his new abbey of
Arbroath about 1180, of a full toft outside the burgh of Perth as
was perambulated (or, as we would say, demarcated) for them
between two ports (i.e. town gates). They were to hold this with
full rights to do business in Perth with the burgesses of Perth.
Where was this toft and where the two ports? Some fifty years
later it was evidently still outside the town between two ports but
otherwise its history is obscure until 1330 when it emerges that the
abbot’s house in Perth had gone, perhaps destroyed in the war, and
that he sought hostillage in the toft outside the burgh between
two ports in a street called Merkarraw; this must be the same
property as that described in 1414 in the Barkarraw outside the
Turret (i.e. High Street) port and in 1456 “ in the street outside the
port and bridge called the Turret brig on the south side of the
highway going from the port to St. Paul’s chapel.” In 1478 it lay
“ abune the Turretbrig betwix the land of Michael Lokart on the
west part and the commone stank of the burgh on the est part^.”
The common stank and the Merkarraw or Barkarraw now lie
beneath Methven Street, but this toft is as securely placed as any
can be on the site now occupied by and to the south of St. Paul’s
kirk at the top of the High Street. Hence the two ports of Perth,
the High Street port and South Street port, are firmly fixed as
already in being by 1180, and with them the two streets. By that
date the town not only had its two parallel main streets but also a
degree of repletion — built-up area — sufficient to drive the king out¬
side the walls for a holding large enough to meet the needs of his
splendidly endowed monastery.

In my time I have been seduced by that other charter of King
William, perhaps a little later in date, granting to Cambuskenneth
abbey a toft in “ my new burgh of Perth,” and have looked for
divisions within the town which may represent stages of its growth^.
One is obvious: the High Street first. South Street second, but on
the argument just presented this will hardly do for King William’s
time. Unfortunately, the toft of Cambuskenneth passes out of
record and cannot be placed upon the map, but since the Arbroath
toft was held in burgage even though outside the burgh, it seems
most likely that the “ new burgh ” in which the Cambuskenneth
toft lay was also this small suburb beyond the Turretbrig.

Charters, however, are not the only source which we can bring
to bear on the problems of the early development of Perth. The

34

town itself as recorded before the alterations of the fate nineteenth
century and twentieth century preserves not merely the street plan,
but also the divisions into holdings the boundaries of which would
pass unchanged, except for sub-division, through the centuries.
Rutherford’s map of Perth is as much a document of King David’s
time as it is of 1774 when it was made. Of course, we must subtract
new streets such as George Street, St. John’s Street and Princes
Street, but Skinnergate, Kirkgate and Watergate are all attested in
documents from the 1220s and along with High Street and South
Street they were there fifty or a hundred years earlier.

The striking thing about this Perth is surely its size. Small by
modern standards it may be, but it was large enough to contain the
population growth of the thirteenth and sixteenth centuries; indeed
its bounds were not really burst until the beginning of the nine¬
teenth century. No doubt there were many open spaces within the
precinct, but that the town existed on such a scale by 1180 is again
a comment upon its early history for it is difficult to cram the
earlier evolution of such an organism within a mere fifty years.
Recently the critical work of earlier scholars in a number of fields
has come under revision; nobody now believes that Scotland had
one eleventh century bishopric and eight created by David I.
Nobody believes that the stone castle was unknown till 1300 for
thirteenth and even twelfth century examples are identifiable. No¬
body believes that the resources and adminstrative structure of the
monarchy were a novelty in the twelfth century though they
certainly were being adapted. In their heart of hearts I doubt if
anybody believes that David I created some eighteen towns and his
successors many more, though the most that is admitted is
antecedent village settlements. On the evidence of Perth, and I
would add, Edinburgh and Dunfermline, the time has come to
consider the eleventh century origins of towns in Scotland.

For instance, the plan of Perth presents intriguing problems.
The communications of which Perth was a nodal point came from
the south and the west, entered the town, crossed the river and
turned northward. But whereas the old road from the west comes
straight down the High Street, that from the south comes downhill
from Station Square and approaches the old South Street port at an
angle. The twelfth-century traveller entering the town there had
to take a dog-leg course down South Street along the Watergate or
the Meal Vennel and then cross the river before he could journey
north. If, however, we take the line of the approach from the
south and as it enters the town extend it across the town, it meets
the width of the High Street somewhere about the Kirkgate at a
point at which the High Street follows the same line by taking a
slight curve to the north. The existence of a former road along the
line just indicated, joining the High Street in a fork slightly to the
west of Kirkgate is a matter of conjecture which could only be

35

PERTH. Based on Rutherford’s map of 1774. The broken line represents
the suggested pre-urban road-line, and the approximate line of the early
ford and bridge. Key to figures: 1, St. Paul’s Kirk. 2, Turret port and brig.
3, Meal Vennel. 4, Kjrkgate and (between High St. and Castle) Skinnergate.

5, Watergate.

36

put to the test of excavation, but, to press speculation a little
further, the diversion of such a road is likely to have taken place
as the Watergate and east end of High Street became places of
business and settlement, that is, as town life developed. The
Watergate route offered an alternative approach to the ford or
bridge for the road from the south but it would take royal approval
to shift the road to the line of South Street.

It is also noticeable that the burgages on South Street form
a continuous line between Watergate and Meal Vennel, so that
there was no approach from the south to St. John’s Kirk, or even
more important, to the market place. The rigs of the south side of
the High Street are longer between St. John’s Kirk and Meal
Vennel than they are to the west of Meal Vennel where the division
of rigs between South Street and High Street is more nearly equal.
Perhaps, therefore, the rigs to the east of Meal Vennel were laid
out before South Street was settled, whereas those to the west of
Meal Vennel were part of a careful allotment after South Street
had come into occupation. These speculations indicate that the
eleventh-century town was on the High Street between Meal Vennel
and the Watergate with the castle outside to the north, with St.
John’s Kirk just within the town to the south, each linked by a
spur-road to the High Street. Outside this town lay South Street,
perhaps gradually subjected to suburban sprawl and so by 1180
fully settled along with the area to the west of Meal Vennel. In
answering the question: when did this expansion take place? much
depends on our view of how active the king’s officers were in laying
out a complete burghal site around a pre-existing nucleus by
dividing it up into rigs for occupation. It could well be that a
burgh was laid out on a generous scale by David I in the area to
the west of Meal Vennel and to the south of South Street and that
open ground was steadily cut up into rigs and occupied over the
following fifty years. But whatever the correct interpretation of this
difficult evidence, there can be no doubt that at the beginning of
David Ts reign there already existed along High Street and Water¬
gate a flourishing urban community attracting merchants from
overseas and traders and craftsmen to settle within the town itself.

The craft element in the early medieval town is perhaps the
most elusive in the sources, but from incidental references in
monastic and other charters a few more lines may be drawn which
illustrate this part of the profile of Perth. Thus the abbey of Scone
was allowed about 1160 to have three craftsmen, a smith, a skinner,
and a tailor who while engaged on the abbey’s business were to
have the freedoms which the king’s burgesses of Perth have in or
out of Perth Notice the specialists whom the good abbot might
employ: a worker with metal, with leather and with cloth; if you
think about these, with pottery and wood they were the only
materials available to medieval society, and leather in particular
played a great part in manufactures and daily life. Leather kept

37

the rain off you, shod you, saddled your horse, made up your
armour, held your drink and even cooked your food, and leather
was produced in ample quantities by the cattle grazed in Scotland.
Clothing was equally indispensable but the canons had no weaver
or dyer, only a tailor to make up the finished product. Evidence for
Edinburgh supports this account of the town as a place where
specialists exercise their craft; division of labour is of the essence
of a town; it implies exchange and therefore marketing and money.
The market like the town was not a place of inter-regional trade
alone, but also a place where humbler wares were produced and
sold: it is possible to have trade without towns, but artisans and
crafts are of the essence of towns.

Among a number of Flemings whom David I attracted to
Scotland was one, Baldwin the lorimer who presumably came to
make the trappings for the king’s horses but who settled to urban
life in Perth with the king’s blessing. He bought a toft in the town
and the king freed it of services except for the burgesses’ duty of
keeping watch and of closing off the end of his toft with a palisade
so that the burgh as a whole was enclosed. He had to pay the king
one terret and two horse collars, but he was free to sell up and
leave when he liked ‘h A little later he was given the small piece
of land to which I have already referred next to the house of
Uilchil. It is tempting to people the burghs of Scotland with
Flemings brought in by David I, but a moment’s reflection will
raise some doubts. If the burgesses we know of are Flemish this
is because written titles and assurances were given to such settlers —
but not to others — to persuade them to contribute their skills to
Scotland’s towns. Baldwin was a craftsman but his skills had a
particular relevance to the king’s court and castle and it should
not surprise that his holding is described as “ a toft in Perth with a
booth in North Street at the castle,’' that is, at the intersection of
High Street and Skinnergate.

A town like Perth exists then to serve, and to profit by, the
needs of society as a whole, and while it lives by the rules necessary
for a manufacturing and trading community, it is in no sense
isolated from the countryside which supplies raw materials and
food nor from the peasants and landowners who must buy within it.
Many prelates had town houses in Perth from which to conduct
the temporal business of their church and magnates may have had
them also. King David I it has been pointed out had a castle and
house there as well as nearby Scone abbey; his grandson’s need
was even greater.

In 1174 King William gambled on making war on the English
king at a moment of difficulty for political concessions; the gamble
failed miserably when William himself was taken prisoner and to
buy his release had to acknowledge England as his overlord and to
surrender three castles to English garrisons: Berwick, Roxburgh

38

and Edinburgh. In effect these three key places ceased to be
available for the king, who however had many alternative castles
and manors. What he did not have was much by way of alternative
accommodation for those solemn or festive occasions of great
importance for one reason or another to which were summoned
the prelates and magnates of his realm. For prelates and magnates
too had to be housed, with their families, servants and animals,
and it was not always easy to do justice to the good opinion which
a man had of himself in the accommodation which was available.
Thus in April 1194 King William went to England to see Richard I.
“ And Hubert, archbishop of Canterbury, and Hugh, bishop of
Durham, went to Brackley, where the dwelling which he had had
for thirty years past by allotment of the marshals of king Henry
had been prepared for the bishop of Durham. And when his
provision had been prepared the servants of the king of Scotland
supervened, wishing to cast out the servants of the bishop; but they
could not. Yet they bought the king’s foods, and prepared them in
a certain house of the same parish. And when the bishop of
Durham had come thither, and it had been told him by his men
that it had so chanced, he refused to retire, but boldly entered his
dwelling, and ordered the tables to be served. And while he dined
came Hubert, archbishop of Canterbury, and offered him his
dwelling; and counselled him that he should leave that dwelling
to the king of Scotland, and depart. But when the king of Scotland
had returned from hunting, late, and it had been announced to
him that this had so befallen, he took it ill and refused to go
thither, but commanded his provision to be given to the poor. And
he himself went to the king at Selston, and complained to him of
the wrong which the bishop of Durham had done him. And hence
the king was much enraged, and reproved the bishop of Durham

Back in Scotland King William almost certainly had the same
kind of troubles; the fact that his constable had a toft in every
burgh shows where he was expected to find lodgings for most of
the peripatetic court. But with three towns gone, the pressure on
the rest increased. It cannot be said that the king’s charters show
him resident more frequently at Scone or Perth than at Stirling or
Forfar, but they do show some large gatherings there. Bishops
were chosen there, bishops who were not obvious curial promotions
and were therefore the result of wider consultation. On Tuesday,
30th October, 1184, the king made an assize, a new law, at Perth,
just as in 1177 he had made one at Aberdeen and in 1180 at
Stirling. For this and other occasions an enlarged royal court was
undoubtedly quartered upon the burgesses of Perth, and Perth
became, to borrow a modem phrase, an acknowledged conference
centre. It had no cathedral but in 1201 the papal legate held a
church council there, in 1206 the bishop of St. Andrews a diocesan
synod, and other church councils were held in later years; the

39

reason was the availability of Scone abbey to be squeezed for
hospitality by the king, the legate or the bishop, and of Perth town
houses to put up the rest. The burgesses of Perth must have made
connections and contracts with the royal household which were
much to their mutual benefit.

Long ago the great scholastic, John Major, in his history of
Great Britain said that in Scotland there was only one walled town:
Perth. In his own day that was untrue because Edinburgh at least
was twice walled, and on Major’s head has fallen the scorn of more
than one scholar since; the walled medieval town is something of
an article of faith with us though, as faith so often is, it seems to
be grounded upon the experience of others and not of ourselves.
For if we look at the footnotes to the article of faith the references
are almost without exception to the burgh records of the fifteenth
century, to the age, that is, of the cannon-ball and the English
invader. And the same footnotes will point to the evidence that
Inverness and Berwick had about 1180 and 1296 respectively, a
palisade and ditch around them; but that is evidence against, not
for, a stone wall. There are also references to the back-dykes of
burghs with which the inhabitants closed off their rigs and so
provided a continuous stockade around the town. In other words,
like so much else of the supposedly medieval town, like John
Knox’s house, Provand’s Lordship, the Fair Maid’s House, the
stone wall is a very late medieval phenomenon; Robert Bruce had
an ill time taking the castles of Scotland one by one from the
English but there is no suggestion that the towns were able to defy
him — except Perth, of which Barbour says the walls were of stone,
describing how Bruce himself carried a ladder across the ditch and
was first upon the wall-head. This was in 1313, at least the second
siege of the English in the town. John Major was right out of
time: in the thirteenth century Perth was the only stone-walled
town in Scotland (though it is possible that such walls existed at
Roxburgh) and the reason is probably to be connected with another
curious circumstance: according to the chronicler Bower, writing
much later but well informed about these few years, in 1210 — a
mistake for the year 1209 — King William had undertaken to pay a
large sum of money to the English king, to raise which he called
his magnates together to ask for taxation at Perth, at Michaelmas.
The weather broke, the rain fell in torrents, rivers burst their banks
and especially the river Tay with the Almond. Destroying a certain
“ mountain,” rnons, it carried away various houses, the bridge, and
an old chapel. King William and his brother sailed out of the
town in a very little boat seeking dry land, and with them a very
few of the magnates who were present. Other magnates who were
in the town saved, themselves as best they could, either in little
boats or in the solars, the upstairs rooms, of the burgesses. Then
the tide turning began to rise and the swollen river far from getting

40

away, surged through the streets of the town so that not only
cobbles and boats but big ships could sail freely through streets
and open spaces. The council was adjourned to Stirling where the
king got his money.

The light which this episode throws upon the town itself is
fascinating. It takes us into the very houses of the burgesses, and
documents also tell us of their solars and their cellars; one
document of 1219 confirms the seriousness of the flood: John
Ylbaren sold to Scone abbey the toft, twenty feet by seventy, on
the south side of the bridge next the Tay on the east, namely at
the very top of the street, a toft which the canons of Scone had
held before the great flood carried it away by violence and which
John had been unable to build up again to its previous state
because of the depth of the water. Another document of similar
date gives us our last mention of the castle: Henry Bald the gold¬
smith of Perth sold to Scone abbey, “ my two booths with the
solar above, on the land which King William gave me, the two
booths namely which are in front of the road leading from St.
John’s Kirk to the castle on the east side against (or perhaps
opposite) the house of Andrew son of Simon, namely those two
booths towards the north.” The site is not easy to place on the
map, but the description of a building with two vaults which served
as shops, with a living room above is unmistakeable. The rent is
a pound of pepper to the king’s household (which by implication
would be interested and available for such a payment) and a
half-stone of wax to Coupar abbey. The immediate interest of this
document however is its mention of the castle at the end of
Skinnergate; this is repeated from King William’s charter describing
the same burgage tenement granted before the great flood, and
therefore it does not amount to evidence for the continued existence
of the castle after the flood The castle we know was circular,
almost certainly an earthwork and in it there was a chapel as early
as the time of David I, and perhaps a good deal earlier; this surely
is the mans denuded by the Tay and the Almond in 1209 of its
palisade and buildings, including the chapel, its embankment eroded
and swept away in a maelstrom of mud, timber and stone carrying
the Tay bridge before it. It cannot have been a very high ring-
work — probably not more than ten feet, and the king would escape
readily enough to his small boat, an old man of sixty-five, with his
brother, David earl of Huntingdon, shouting at a few of his house¬
hold to pull harder on the oars and get out of here. Meanwhile
in the town the prelates, earls and barons of Scotland waded or
rowed up the High Street or South Street or if they were housed
in the Watergate scrambled upstairs to escape the rising torrent.
Here was a scene to inspire the muse of William McGonagal!

Some thirty years later one of the new evangelical orders of
the church, the Dominican or Black Friars had their church

41

consecrated on or near the castle and on the same day in 1244 the
king gave to the order his house in Edinburgh and his garden in
Perth^"^. From this date the area of the castle became the Black
Friary to which in 1266 for example the envoys of Norway came
to make peace with the King of Scots. Whether the king himself
resided at the Friary at this time, as so many of his successors
were to do, is not known. But Kinclaven was developed as a royal
castle to house the king’s stores, and act as a hunting lodge and
residence when he made his way along the Braes of Angus. Perth,
however, was left without a castle so that when Edward 1 took
over Scotland his instructions poured forth to the keepers of this
or that castle, but always to the warden of the town of Perth —
because there was no castle there. Unfortunately, there is
absolutely no documentation to reveal when the town wall was
built. It does not seem to have been very high and my own surmise
is that it was meant to do two things: to show the wealth and
importance of the town, to act as a status symbol, and secondly to
prevent a recurrence of the great flood of 1209, keeping out the
waters of the stank or mill-stream which flowed round three sides
of the town. If that second conjecture is correct then obviously the
waterfront must also have been secured by raising solid wharves;
in Rutherford’s map there is an interesting line of buildings running
parallel to and midway between the Watergate and the river and
at right angles to the burgages, a line which may represent a
strengthened waterfront for medieval Perth.

The evolution of the community of Perth in the fifty years
between Baldwin the lorimer and the flood of 1209 is not well
documented but some evidence of change can be gleaned from
surviving charters. Thus already when Malcolm IV granted the
booth in Perth to Baldwin he addressed his charter to “ the provosts
and all his other burgesses of Perth.” This phrase implies a great
many things. Within a burgh the king was entitled to rent from
each of those holding land of him, but from a very early date in
his reign David I refers to this rent as his firma “ farm,” that is to
say fixed amount, from his burgh. He had given to the responsible
officer a target figure to produce, but it was the officer’s responsi¬
bility to see to the collecting of the individual rents and if he made
a surplus or a loss then that was his good or bad fortune. With
some burghs the responsible officer was the sheriff and the grieves
or provosts remained his deputies or agents. In the case of Perth,
however, they had entered into a direct relationship with the king
who expected that they would be chosen from among his burgesses.
Moreover as King William’s charter of 1209^^ to the burgesses
shows, and as other burgh charters confirm, from about 1180 each
town was held responsible in common for payment of taxes or aids
to the king. These documents reluctantly reveal to us an active
burgh community accepting responsibility for the payment of its

42

own farm and its own taxes through the medium of provosts chosen
by it from among its own number. But only those of wealth and
resources could undertake the responsibility for such financial
burdens. So much was this the case that when in 1209 the king
granted his charter to the burgesses he explicitly laid it down that
those within the burgh who wished to deal with the burgesses in
the market should also have to be responsible with them for the
king’s aids. The implication is clear: within the burgh there were
burgesses and there were also a great many others who lived there
but who were not admitted to the privileges of burgess-ship and
who had impudently claimed that neither should they be admitted
to taxpaying. They would not be denied personal freedom nor the
right to deal with their property as they wished, but they were
excluded from the privileges within the burgh of determining the
regulations by which the inhabitants were to live and practise their
trades and of choosing the provosts who represented them to the
outside world. And so when we come to the charter granted by
King William to the burgesses of Perth in the year 1209, we
should understand that it was just that — a grant to some of the
inhabitants of the town.

With King William’s charter we seem at first to enter a world
of haphazard and quirky regulations thrown together with no very
obvious regard for logic or convenience, whether of the modem
scholar or the medieval business-man. But the charter as we have
it is a precious survival from the campaigns which flowed back
and forth over the town between 1296 and 1338 destroying not only
houses but also the documents of Perth’s early history. Behind it
probably lie several lost charters — at Inverness there were four
charters of King William — expressing individual grants or small
groups of grants which were brought together and strung into one
long document in 1209, so that the same subject recurs as it did in
those earlier documents rather than in one neatly rearranged set
of provisions. Thus a statement in the middle of the charter that
the burgesses are to have their gild merchant except for fullers and
weavers was not necessarily the first grant of this gild. But there
is a further caveat of which account must be taken: that the
evidence of town documents in England and the continent shows
that charters were often concerned with matters in controversy, but
passed in silence over privileges which were well established. They
express the ruler’s recognition of something which may have a
long previous history in urban society. Such was the gild beginning
as a convivial society for mutual assurance and developing into the
body among which the king can find responsible provosts, which
will negotiate with him over the farm, and which will press the
interests of the town and especially the town business men upon
him. The gild merchant of Perth may have its roots in the time
of David I, but its emergence into respectability probably came

43

in the reign of William, not in 1209 but years, even twenty years,
earlier, when the boom of the late twelfth century was well under
way. The name gild merchant means, be it noted, not gild of
merchants or overseas traders, but business gild, and the kind of
business which was particularly important to it is broadly hinted
at by the phrase “ except fullers and weavers.”

Medieval cloth manufacture involved a number of processes
of which the earliest, spinning, carding, washing, weaving were
common to any kind of cloth, though of course they could be
carried out with varying degrees of care. But no basic industry —
and cloth was an essential manufacture, with a rising demand for
the finished product — could function on haphazard methods of
production and marketing. And again English evidence makes it
abundantly clear that the twelfth century cloth industry was well
organised and controlled by a group of merchant capitalists within
each town, often those merchants who engaged in the importation
of dyes and mordants, in which, since they came from afar, con¬
siderable investment and risk was involved. These entrepreneurs
arranged for the spinning of wool domestically, put the yarn out
to weavers on piece rates, and had the cloth fulled, a laborious
process of pounding in the wet cloth in order to felt the fibres;
thereafter it might be teazed and shorn to give a fine finish. Much
cloth was used in its natural colours, but there was also a high
demand for cloth dyed either in the yarns or in the piece, and this
dyeing took place in the sheds of the employers of working dyers.
In brief there was both an employer class and a labour force, and
within each there were grades: wealthy capitalists who sub¬
contracted part of the process to employers of specialist workmen,
but who controlled the whole industry and sold the finished cloth;
tradesmen employed on the skilled work of teazing and shearing
the cloth, weavers throwing their shuttle throughout the daylight
hours to meet their quota, fullers sweating over the trough as their
feet pounded the broad cloth to felt and shrink it into a tougher
and denser fabric, resistant to weather and wear. So the gild of
Perth “ except weavers and fullers ” is by implication a gild of
employers from which those who labour with their hands and
feet to make profits for others are excluded — excluded because they
are not worthy to associate with men whose hands no longer labour
with the tools of a trade.

Cloth entrepreneurs were, however, more vulnerable to com¬
petition than they liked. When the Perth charter says that no
foreign merchant is to cut cloth for sale in Perth except from
Ascension day (usually in May) until 1st August when they may
sell in the market place doing business with the king’s burgesses,
it is warding off for a large part of the year the competition of
cloth from Flanders or England finer than Scotland could make,
and is taking up another point in the charter — no foreign merchant

44

is to buy or sell anywhere in Perthshire except in the burgh of
Perth — which is to say that the importer may not go direct to his
rural customer and either undercut or oversell his native urban
competitor. All must be sold openly in the town market place
where the native merchant can be sure of “ fair ” competition — a
proviso that applies to all merchandise and not just to cloth. And
what of the rural weaver whose coarse cloth could not be subjected
to quality control but for which nonetheless there was an insistent
peasant demand; was not his trade in rough fabric hurtful to the
interests and reputation of the town cloth-makers if dressed up in
the guise of finished cloth? Then let him stick to what he makes;
as the charter has it: let no one living in Perthshire outside Perth
make dyed or mixed cloth or cause it to be made except my
burgesses of Perth who are in the gild and are responsible with my
burgesses for my taxes; if contrary to this prohibition any dyed or
shorn cloth is found, I order my sheriff of Perth to take it and do
with it what was customary in David I’s time — a statement which
takes this protective measure, and the sophisticated business
structure which it implies, right back into the first half of the
twelfth century. Malcolm IV allowed Scone abbey to have a tailor
but made no mention of a dyer.

The final provision of the charter, that no-one from outside
Perth may buy or sell v/ool or hides except in the burgh of Perth,
is in part designed to protect the control of the cloth merchant over
his raw material. But its significance is wider, for Scottish wool
was already being exported in large quantities to feed the multi¬
farious looms of the Flemish cloth industry and the effect of this
provision was to exclude any foreign exporter from direct access
to the grower and to force him to deal with the Perth middleman.

The charter, of course, tells only one side of the story. Those
who read the burgh records of the later middle ages will be well
aware that on the one hand restrictive measures are continually
justified by pleas of quality control and that on the other hand all
consumer comment in literary and documentary sources is about
high prices and shoddy goods. Most business men believed not in
competition but in “ fair ” competition by which is meant “ com¬
petition for everyone else and protection for me.” The cloth
merchants of twelfth century Perth were clearly of this school of
thought so that for example they would brook no interference with
the one material for which the town depended upon the surrounding
country — fuel. I give my firm peace, says the king, to all those
who bring wood or timber to Perth; let no one disturb them coming
and selling after they come within one league of Perth. Now selling
wood involved high transport costs over even a short distance, but
quite substantial profits if that distance was kept down, and these
profits went to the owner of the woodland, usually, that is, to a
landowner. It is not clear who would interfere with the supplier

45

but there is a suspicion that what troubled the worthy town
business men was a combination among their suppliers to withhold
their cartloads until the price had risen satisfactorily. And finally
what of the antiquarian ring of the provision about taverns: let no
one have a tavern in any village in Perthshire except where a
knight is a lord of the village and lives there and then let there be
only one tavern? What has this to do with the burgesses? Again
if we turn to England the meaning becomes clear. A tavern was
not any alehouse but a place for the sale of wine, and wine had
its own profits and problems. Each year it was produced in France,
and especially in Gascony, in vast quantities and exported by sea
to Britain for sale. The merchant who handled it must be a man
of wealth and will usually become a man of greater wealth, for
the demand could not abate. This year’s wine turned into next year's
undrinkable vinegar and must be replaced by new supplies, for
there were no sterile bottles and corks to keep the vintage drinkable
year after year. It has been shown that the wine merchants of
London were at this time the largest single group in the patriciate
of that great city, and that one of the important factors in the
growth of their business was the connection they had with the royal
court through supplying wine — for, of course, it was consumed
only at the tables of the well-off. Now the wealthy London wine
importer could meet resistance when he reached port with his
tunnage from the merchants of that port: no foreign merchant may
have a tavern except on his ship, said Henry II to Bristol, Dublin
and other Irish towns, which is to say that the merchant may not
break bulk to sell his wine to the consumer, may not open a
tavern in the town. A monastery, a king, an earl, will buy wine
by the tun from him either on contract or at the quayside, but he
may not sell direct in lesser quantities. He must sell to a local
merchant, a middleman, and this restriction is, I think, implied
by the Perth charter which restricts taverns in the countryside to
those places where the social claims of a knight to such an amenity
cannot well be denied. In keeping the wine-seller out of the land¬
ward and confining the sale of wine to the town, the charter is
closing a loophole to those wine importers who seek to avoid the
town merchants’ monopoly of tavern-keeping by opening up in the
countryside — the end of another avenue of “ unfair ” competition.

So far as it goes the evidence of King William’s charter is
highly revealing but it tells us very little if anything about the
town government nor about the men who made up the vested
interests it reveals. We can complement it not only by stray
documents in the English records but also by the evidence of some
thirty charters of the early thirteenth century dealing with Perth
properties and scattered through the monastic collections of Scone,
Inchaffray, Lindores, Balmerino and Dunfermline abbeys^^. Some

46

of these documents have already been quoted, but their particular
fascination is that they reveal the names of about a hundred
inhabitants of Perth, some of their relationships, the tofts they held,
the rents they paid or charged and the names of their streets. There
are no surprises in the street-names, but the rents far exceed the
sixpence per toft given in the Burgh Laws, a tract of earlier date
and of particular relevance to Berwick. Higher rents reflect
increased demand for house and shop-room and ultimately the
increased prosperity of society and especially town society.

Among the hundred townsmen are some few with trade names
in the form “ James the Smith,” and many others with double
names such as John Cokyn, William of Lynn, or James fitz Uhtred.
There is only one case where the double name is followed by a
trade description, and that is Henry Bald the goldsmith whom we
met earlier in this paper and who occurs seven times in the thirty
charters, so was clearly a man of consequence. This indeed is to be
expected of a trade which handled such a valuable commodity;
the goldsmith is prominent in whatever town we turn to, because
of the capital he lays out on his raw material, and it may be that
he soiled his own hands by working with it but was still acceptable
to the gild. The other tradesmen are too humble to have surnames.
Four of them, the largest group, are dyers, which bears out King
William’s charter in making Perth a centre of cloth manufacture,
but these four are clearly not the capitalist entrepreneurs who ran
the industry but the managers or workers of dye sheds which took
in cloth to the order of the great merchants. The other trade names
include tenteman (a cloth worker), tanner, lorimer, helmet maker
and smith, workers that is with cloth, iron and leather but not the
two humblest cloth workers, weaver and fuller, webster and
waulker, nor any of the victual trades, flesher, baxter, brewster.
These apparently were too lowly to register even in the chance
sample of documents which survive.

Among the other names seventeen stand out by reason of the
frequency of their occurence, four times or oftener. It may be
a coincidence but it is I think a significant one that in 1296 when
the Scottish towns submitted to Edward I twelve burgesses spoke
for each town except Perth for which eighteen spoke, presumably
because Perth was more important than the others and its governing
group, the proto-council was larger — and the number, eighteen, is
strikingly close to the seventeen of half a century earlier^"^. Now the
business activities of these men are not described except in the
case of the goldsmith, and 1 attribute this fact to the activities
being variable and to their not involving anything which dirtied
the hands with labour; some of the seventeen must have run the
cloth industry but none would have thanked you for being
described as “ cloth-worker.” Others presumably bought imported

47

wine for retail, or bought wool for export either by themselves in
a company hiring or owning a ship, or by sale to visiting Scottish,
English or Flemish merchants. Most of the documents which
illustrate this trade relate to Berwick but their relevance to Perth
is indicated by one or two circumstances. Thus one of the Berwick
merchants, William de la Bataille who was very active exporting
wool and hides and bringing back in exchange com and wine, must
surely have been related to the Perth merchant John de la Bataille
for the name is not a common one. That is no guarantee that they
were engaged in the same (or any) trade but then we should in
any case probably eschew the notion of specialist traders who would
touch only this or that cargo. Among the Perth merchants was a
group with the surname Lenn or Lynn including a William and his
socius or business partner, Richard, who must either in their own
persons or in those of their ancestors have come from the Norfolk
port of Lynn which specialised in exporting the grain produced by
the midlands of England. King William’s charter makes no mention
of the grain trade but there is plentiful evidence that substantial
quantities were imported annually to Scotland throughout the
thirteenth century to meet the needs of a rising population and
restricted arable at home and the Lynns presumably represent
a family of grain factors settled far from home. In 1246 King
Alexander II promised that he would see that the merchants of
Perth paid all the debts they owed in the wine port of Bordeaux,
and the English king in return released the Perth ships he had
arrested as security at Lynn, a document which tells us that Perth
was as much involved as Berwick in the importation of wine and
grain paid for by exports of wool and hides^^. If Perth was like
Berwick, then the ships of London cornmongers and wine
merchants tied up at the Watergate of Perth in even larger numbers
than the ships of native merchants. King William’s charter does
indeed imply that the business men of Perth were stay-at-home
middlemen rather more than venturers in overseas trade, though
there certainly were overseas traders among their number.

Most prominent among the prominent seventeen is Henry son
of Geoffrey son of Martin who held the office of provost in a
document of the early 1220s, a document to which he is first witness
and to which the burgh seal is appended. In another document of
much the same time he also appears as first witness, as alderman
and as appending the burgh seaP^ Now the burgh seal betokens
the communal existence of the town, its capacity to authorise the
land transactions of individuals, or to undertake to pay the farm
to the king. The place in which such decisions were nominally
taken was the burgh court presided over by the provosts, apparently
three in number at Perth^^. But the alderman, by all comparative
evidence, was something quite different — the head of the gild, with

48

no inherent connection with the burgh seal. Apparently the gild
has effectively taken over decision-making, treating the court as a
rubber stamp and disposing of the burgh seal as though the gild
was the burgh — as it certainly was the burgesses. The provosts are
gildsmen, and one of them is alderman of the gild, whence, I take
it, he becomes the most prominent provost, the provost, while the
other provosts becomes just baillies, town officers. It is a neat
illustration of this point that in one place Henry fitz Geoffrey and
John de la Bataille appear leading the burgesses and although not
provosts at the time are prefixed with “ dominus ” — lord — as men
of wealth and distinction^^.

It is not known in what Henry fitz Geoffrey traded but
significantly he is found witnessing charters of two great land-
owners, the bishop of Dunblane and the earl of Athol with whom
he would not be on calling terms other than for business purposes.
But there is room for some further speculation. For much of his
reign King William had at the head of his household responsible
for the purchase of large stores and sale of surplus, a Richard of
the Provender who in 1203 was rewarded with the bishopric of
Dunkeld. After his death his successor who also had Perth
connections gave land there to the niece of bishop Richard and to
her husband Henry who maddeningly does not tell his sumame^^.
If it was Henry fitz Geoffrey fitz Martin then we should have much
light upon the growth of the family fortunes. Under the new king
Alexander II another office rose to prominence in the household
giving its holder very similar responsibilities — the clerk of the
liverance. In 1228 that office was given to a rising royal clerk,
Geoffrey fitz Martin of Perth who in 1236 became in turn bishop
of Dunkeld^^. He cannot have been Henry’s father but it would
be neat if he were related to the burgess, a connection that would
make him better at his job, useful alike to the king and the family.
Thus it is possible that in Scotland as in England, albeit on very
different scales, the fortunes of some urban capitalists were made
through provisioning the king and his household, and especially
through supplying the £400-worth (the equivalent of an earl’s
income) of wine which was consumed in a year by the court in
the 1260s.

History is all too often the history of the rich and successful:
the poor are with us as the rich see them and try to make them
go away with gifts of alms. In this paper there has been little
enough about the luckless weaver and fuller, and nothing at all
about the really poor and destitute, for these men never put their
names to parchment. The profile I have drawn has been incomplete
but not I hope without interest. Perth was the second town of
Scotland, second, that is, to Berwick, and far ahead of Edinburgh,
Aberdeen or Stirling in importance. It grew from a settlement of

49

the eleventh century or earlier where ships came up the Tay to
unload by the land routes to north and south. It flourished through
royal patronage, first by receiving the privileges of a burgh, and
then through business dealings with the king’s household. It was
the home of a small business community active in the manufacture
of cloth, import of wine and corn and export of wool and hides.
They were intent on making a sure profit and struggled to load the
conditions of business against both the less fortunate and the more
fortunate. But they were also an essential element in bringing
about the rising Jiving standards of their society.

But what I have given you is no more than a profile in two
dimensions, and I should like to call to your attention the number
of unsolved problems before you: the early street lines and evolution
of the town; the shape and scale of the town houses, the materials
of which they were built, the standard of living of those within
them, the places of trade and modes of manufacture of the crafts¬
men, these and a hundred other things have passed untreated in
this lecture because they are undocumented in the written sources.
But another dimension, a solution to some of these problems is
possible. It is emerging from the ground for two great twelfth
century towns elsewhere in these islands — Dublin and Winchester —
and for a seaport at Bergen in Norway. It is high time that we in
Scotland took our opportunities to record the urban past before it
is too late. In many towns it is too late because Victorian founda¬
tions went deep and swept away the archaeological record of the
past. But at St. Andrews it has proved possible to excavate a
town house and to set the archaeologists there upon the track of
other sites and even of medieval vaults and booths hidden behind
the eighteenth-century facades of town houses. The other town
which should offer the same opportunities is around us, a town with
a diminishing number of early buildings but still with medieval
street lines and perhaps even cellars, a town with aspirations to
attract a university but more successful in attracting an insurance
company. Don’t mistake me: I’m not a preservationist, nor hostile
to business, but I have a great affection for Perth for it preserves
the scale of a town in which one can live and move and do all
necessary business of man and still dawdle and gossip and sense
the existence of the real world of nature. It is perhaps the most
attractive town in Scotland, because its heart has the shape if not
the beat of King David’s time. Some things have changed but not
everything; it is still St. Johnstoun as well as General Accidentville.
What we have from the past we should study and value before
we decide to throw it away, but so far as archaeology is concerned
time for that study is running short. Money is available, and with
sufficient effort, manpower; is there any reason why they should
not come to Perth? Is there any better place to which they could
come?

50

References

1. Pryde, Burghs of Scotland, Nos. 1-5.

2. Regesta Region Scottorum (hereafter RRS), i, 27-28; Nos. 157-8; Lawrie,
Early Scottish Chapters (hereafter ESC), No. LXXVII.

3. Lawrie, ESC, No. LXXIV.

4. Lawrie, ESC, No. XLVIII.

5. Lawrie, ESC, Nos. CLXIII, CCIX, CLIII; RRS, i. Nos. 260, 223.

6. Lawrie, ESC, No. LXXIV, and the discussion in Proc. Soc. Antiq. Sc.
XC (1956-7), 57-58.

7. RRS, i. No. 171; St. Andrews Liber, 142.

8. RRS, ii, No. 233; Arbroath Liber, i. No. 107; ii. Nos. 7, 8, 52, 107,
202, 361.

9. RRS, ii, No. 278.

10. RRS, i. No. 246.

11. RRS, i. No. 121.

12. Anderson, Scottish Annals, 312-313.

13. Scone Liber, Nos. 82, 86, 45.

14. Anderson, Early Sources, ii, 520; Perth Blackfriars, No. 5; Edinburgh
St. Giles Registrum, No. 79.

15. RRS, ii. No. 467.

16. The following paragraphs are based upon: Scone Liber, Nos. 21, 45, 46,
79, 80, 82, 86, 88, 89, 90, 95-97, 169; Dunfermline Registrum, Nos. 144,
149, 166; St. Andrews Liber, 316; Arbroath Liber, i, No. 215; Inchaff ray
Charters, Nos. LXIX, LXX, LXXI, LXXIII; Lindores Chartularv, Nos.
LXVI, LXVII, LXVIII; Balmerino Liber, Nos. 21, 22, 25, 27, 28, 30;
Coupar Chrs., No. XXV.

17. Cal. Docs. Scot., ii, Nos. 813-815, 819-820.

14. Cal. Docs. Scot., i, No. 1694.

15. Scone Liber, Nos. 86, 97.

16. Scone Liber, No. 82.

17. Inchaff ray Charters, No. LXIX.

18. Dunfermline Registrum, Nos. 138, 140. Richard of Leicester, burgess
of Perth, was related to John, Bishop of Dunkeld, 1210-1214.

19. Scone Liber, Nos. 62, 93.

^ ^

Since this paper was written and set up in type I have come to
the conclusion that King William’s charter to Perth, which is dated
at Stirling on 10th October and which, as the witnesses show,
belongs to 1205-10, was granted on that day in 1209. The circum¬
stances are of great interest and are discussed in my forthcoming
volume of the Edinburgh History of Scotland, The Making of the
Kingdom.

PUBLISHED BY THE SOCIETY AT
CITY OF PERTH MUSEUM & ART GALLERY

1974

Printed by MUNRO & SCOTT, LTD., SKINNERGATE PRINTING WORKS, PERTEI