Gillian Hutchinson - Medieval Ships and Shipping-Fairleigh Dickinson University Press (1994)
Gillian Hutchinson - Medieval Ships and Shipping-Fairleigh Dickinson University Press (1994)
and Shipping
Gillian Hutchinson
BOSTON
PUBLIC
LIBRARY
Medieval Ships and Shipping
Frontispiece: A lead pilgrim badge, found at the Thames Exchange in London,
depicting St Thomas a Becket's return to England in a hulk (private collection -
photo: Museum of London)
MEDIEVAL SHIPS
AND SHIPPING
GILLIAN HUTCHINSON
A CIP catalog record for this title is available from The Library of Congress
ISBN 0-8386-3628-4
Acknowledgements viii
List of figures ix
5 Carrying trade 88
6 Ports 104
8 Fishing 129
Glossary 183
Catalogue 191
Bibliography 199
Index 211
Acknowledgements
When Helen Clarke invited me to write a book for Leicester University Press
neither of us imagined that it would take more than four years to write. I am
grateful to Helen and to the publishers for not despairing of ever receiving
the final manuscript while I enjoyed myself exploring a subject which extends
in so many different directions. The National Maritime Museum has
provided a great deal of support and allowed me three months study leave at
the end of 1993 to bring the book to completion.
I have benefited enormously from the knowledge, skill and enthusiasm of
List of figures
Bergen. 113
6.6 The twelfth-century wine vault on Castle Quay,
Southampton. 114
6.7 Southampton in the later Middle Ages. 116
7.1 Map of Britain by Matthew Paris, drawn about 1250. 118
7.2 St Guthlac travelling by boat in the Fens, twelfth century. 1 1 9
thought for the creaking, stinking wooden vessels which carried them across
heaving seas from one port to another.
The purpose of this book is to bring together information about material
remains of shipping activity and to explain their significance by setting them
in the context of medieval maritime Britain. Westerdahl (1992: 6) has coined
the phrase 'the maritime cultural landscape' to signify 'the whole network of
sailing routes, with ports and harbours along the coast, and its related
constructions and remains of human activity, underwater as well as
terrestrial'. Taken in its broadest sense, the maritime culture of medieval
Britain can be detected in areas remote from the immediate coastal zone.
Fresh sea-fish were distributed well inland, for example to Coventry, and
preserved fish and imported goods reached all but the poorest and most
inaccessible places. Shipping made it possible for wine to become a regular
item of consumption for more and more of the population of a country
which produced little wine itself. The import of wine increased from the time
of the Conquest and continued to grow even when, at the beginning of the
thirteenth century, the loss to the English crown of most of the territory of
northern France meant that ships had to make the longer voyage through the
Bay of Biscay to the English possessions in Gascony. Much of the
agricultural production of Britain was locked into the export trade, most
notably of wool, and monastic merchants' houses in rural settings were built
with the profits of seaborne commerce. Trade was the formula for turning
sheep into stone.
Material remains not only demonstrate the import of foreign items but
2 Medieval ships and shipping
they can also provide evidence for cultural diffusion by sea - the imitation of
foreign techniques, practices and style. In the Norman period, for example,
improvements in the shape and technique of wheel-thrown pots made at
ports and large towns near the coast, such as Southampton and Winchester,
were the result of trade contacts with Normandy. Meanwhile, further inland,
handmade traditions continued (Dunning, 1959: 34).
The pace of new discovery of material remains of medieval shipping has
been rapid. There is also a vast resource of historical and pictorial evidence
to draw on. Ethnography and continuing practice can also enhance our
understanding, particularly of subjects such as fishing, sailing and pilotage.
With all these information sources there are of course problems of
interpretation. What we have is not a perfect representation of the past. The
archaeological record is the result of accidents of preservation and the
opportunity to study complete ships, complete ports, must be
still less
subject, the degree of realism attempted and the limitations of the medium.
The format can impose distortions, as can be seen when images of ships were
made to fit onto circular coins and seals. There is also the problem of scale.
The medieval convention of exaggerating the size of human figures when
they were shown with ships can give the thoroughly misleading impression
that medieval ships were no bigger than modern yachts.
Friel (1983b) has provided a very useful survey of some of the types of
documentary evidence for medieval ships. Documentary sources, while
excellent for things which were subject to bureaucracy, such as the raising of
fleets and the taxing of trade, can be disappointing for physical details.
Medieval documents are very likely to tell you how much a ship's carpenter
was paid but not what he did; to tell you the price of fish but not how it was
caught.
Physical and documentary sources can provide some insights into both
what shipping meant and how it affected their
to people in medieval Britain
lives. When, from towns had seals made for
the end of the twelfth century,
applying the stamp of civic pride and identity to their official documents,
most inland towns chose to depict their walls while most ports had ships
(Brindley, 1938). Some, like Bristol and Scarborough, had both. Great
Yarmouth, meanwhile, chose the motif of three herrings for its coat of arms,
perhaps in imitation of Liibeck. After Edward Ill's victory in the sea battle at
show himself enthroned
Sluys in 1340 he had the gold coinage redesigned to
on a ship (Callender, 1912). The ports of medieval England had a high
proportion of their churches dedicated to St Nicholas, the patron saint of
seafarers, and pilgrim's badges remind us that a great number of people who
were not employed in shipping had first-hand experience of sea voyages.
Introduction 3
|
By comparing the open, single-masted ships in use at the beginning of the
medieval period with the multi-decked, fully rigged ships at the end, it is
,
plain to see that far-reaching changes had taken place. Tracing the progress
of those changes is not so easy, as it is confused by the diversity of regional
shipbuilding traditions and variations in vessel design to suit particular
purposes or conditions. Market forces and the imperatives of warfare meant
that ships for trade and fighting were under constant development. Ship-
building rivalled architecture as the most advanced technology of the Middle
/ Ages. Change and development were made possible by the interaction of
shipbuilding traditions from different parts of Europe and by the experiment
'
and innovation undertaken by shipbuilders. It can be argued that there was a
loss of diversity among the largest ships, so that by the end of the period the
products of shipyards throughout Europe were more similar than ever
before. They left behind them innumerable types of smaller vessels which
were well adapted to the work they had to do and which underwent little
change over centuries, some types surviving even into recent times.
The two opening chapters of this book introduce the technology of
medieval shipbuilding, the methods of construction and the structure of
hulls. Use of technical terms has been kept to a minimum and a glossary is
provided towards the end of the book. The fitting and rigging of ships is
treated separately in Chapter 3 and these three chapters together provide the
technical background for the regional examples of ships presented in
Chapter 4.
At the time of the Norman Conquest the boats and ships in use around
the coasts of Britain were clinker-built, which is to say that their hulls were
built up by lapping the bottom edge of each run of planking over the top of
the one below and nailing them together through the area of overlap. Their
hulls were similar in shape at each end, they were propelled by single square
sails and oars and steered by side rudders. The scant evidence for Anglo-
Saxon shipbuilding, primarily the Sutton Hoo ship (Bruce-Mitford, 1975:
345-435) and the Graveney boat (Fenwick, 1978), indicates vessel types
similar to those used in Scandinavia and developed by the Vikings and their
Norman descendants. Viking ships would have been familiar in those parts
of Anglo-Saxon England which came under Scandinavian settlement.
The Norman invasion fleet of 1066 was capable of transporting large
numbers of men and horses and its ships were the product of centuries of
refinement. The structure of ships of the eleventh century exhibits a high
degree of technical competence, as the wrecks excavated at Skuldelev in
Roskilde Fjord, Denmark, demonstrate (Olsen and Crumlin-Pedersen, 1968;
Olsen and Crumlin-Pedersen, 1978; Crumlin-Pedersen, 1991). These ships
were deliberately sunk to make a defensive blockage of the principal
Shipbuilding: traditions of the northern seas 5
navigable channel in the fjord. Of the five ships, two were cargo vessels, two
were warships and the other may have been a fishing boat (figure 1.1). The
warships were of lighter construction than the cargo ships and were long and
narrow. Their length to beam ratio was as much as 7:1 as compared to the
cargo vessels' 4:1, giving them a high speed capability under sail and oars.
Tree ring studies have shown that the warship Skuldelev 2 was built in
1060-70 in Dublin (Bonde and Crumlin-Pedersen, 1990). The cargo vessels
were quite different from each other. One was made of pine and may have
been built in Norway. The Skuldelev finds show variety and specialisation
within the clinker-building tradition. Shipbuilders had by the eleventh
century acquired a thorough understanding of the scope and limitations of
the materials, the properties of wood and the behaviour of fastenings, and of
the performance of hulls in the water.
The clinker technique was used for building boats and ships throughout
most of the Middle Ages in northern Europe. As long ago as 1914,
Hagedorn, as a result of a study of ships depicted on seals, identified three
different strands in medieval northern European shipbuilding and designated
them 'keel', 'cog' and 'hulk'. All three employ the clinker technique but they
represent different approaches to building the hull. They each had their
origins as regional variations before the beginning of the period under study.
They existed side by side throughout the period and came increasingly into
contact with each other and with the quite distinct southern European
shipbuilding tradition which used flush-laid, rather than clinker, planking.
Keel
The term 'keel' is used by ship archaeologists to denote ships, like those of
the Anglo-Saxon and Scandinavian traditions referred to so far, which were
built up from a keel, with planking running roughly horizontally between the
stem and stern posts. As this was the normal method of building in medieval
Britain the term may not have had much currency at the time. People would
be more likely to refer simply to boats and ships, or use the names of specific
types such as barges or farcosts which were variants within the keel
construction tradition. In Anglo-Saxon literature 'ceoP was used as a poetic
synonym for ship. Toll collectors in London in the eleventh century found it
useful to be able to distinguish between ordinary ships, 'ceols', and vessels of
another type referred to as 'hulcis'. In the later Middle Ages there were vessel
types called keels on the Tyne, Humber and Severn and in Lincolnshire and
London, and the application of the word to specific variants reinforces the
idea that the keel tradition is more easily recognised in retrospect than it was
at the time.
The sequence of construction of a keel is illustrated in figure 1.2. The first
Figure 1.2 The sequence of construction of a keel-type vessel (Viking Ship Museum,
Roskilde).
a potential area of weakness. The top of the keel was flat or dished from side
to side. In profile the keel was T-shaped (though in some vessels the T was
very squat) with the sides cut away to create a flange at the top. To the
underside of this flange the lowest run of planking, the garboard strake,
could be fastened. The keel could have parallel sides or it could be wider
<s Medieval ships and shipping
near the middle of the boat, or slightly forward of the middle, tapering
towards the ends.
To the ends of the keel were scarfed the stem and stern posts. Timbers
were chosen with grain following the long curve required for the finished
components in order to provide strength. These were often in two pieces
scarfed together: the stem post itself and an intermediate piece between the
stem post and the keel. The stem and stern posts of Skuldelev 3, shown in
figure 1.2, have false planking cut out of the solid wood to provide a secure
transition between the posts and the hull planking. This technique required
special skill and simpler posts were more common, as for example that of the
Poole Foundry boat (figure 8.7).
Once this backbone was in place a start was made on planking the hull.
Planks were made from radially split oak trunks or from tangentially split
softwood trees (for the selection and conversion of timber for boatbuilding
see McGrail, 1987: 23-43). A man can be seen preparing planking in the
scene from the Bayeux Tapestry shown in figure 1.3. The first strake was
fastened to the underside of the keel flange. A strake is a run of planking
from one end of the vessel to the other, which might be made up of several
planks scarfed together. The plank ends were tapered to a feather edge so
that when they overlapped at the scarfs there was no increase in the
thickness of the strake. Caulking (waterproofing material) was put between
the strake edges and into the scarfs as they were assembled. This was made
of animal hair or moss, generally mixed with tar. Samples of animal hair
which have been analysed, from Dublin and London, show that it was
frequently not sheep's wool but was often cow or even horse hair, probably
the by-products of the tanning industry. A sample which did prove to be
wool, from London, had been dyed and was probably textile waste (Musty,
1993: 33). At the Queen Street waterfront site at Newcastle, 73 small twisted
rolls and flattened pads of fibre have been identified as caulking material
from clinker-built vessels. Of the 25 samples examined for species analysis,
13 were of cattle hair and 10 of sheep's wool with two possible examples of
goat hair (Walton, 1988: 78).
The edges of the strakes and the scarfs were fastened with iron clench
nails. These nails had large round heads and were hammered from outboard
to inboard through partly pre-bored holes. On the inboard face of the
planking the shanks were hammered over and clenched against quadrilateral
roves. The extreme ends of the strakes were feathered for fastening to the
stem and stern posts with iron spike nails. The hull planking was built up as
a shell, probably largely by eye as in more recent Scandinavian practice,
without the use of formers. Levels and width gauges might be used to make
the shape of the two sides of the hull as similar as possible but exact
symmetry was rarely achieved.
When the bottom planks were in place the first transverse framing was
inserted into the hull at regular intervals. These frame timbers,which crossed
the bottom of the vessel including the keel, are called floors. They were cut
from grown crooks, following the run of the grain. Their undersides were cut
into steps or joggles so that they would be in contact with the planks as
much as possible. They were fastened only to the planking and not to the
keel by means of wooden pegs known as treenails. These were inserted from
10 Medieval ships and shipping
medieval period very large vessels, both warships and merchant ships, were
built in the keel technique. The English royal galleys of 1294 and the 'big
ship' from the Bryggen at Bergen in Norway, built in the first half of the
thirteenth century, were well over 30 metres in length.
Hulks
In the Anglo-Saxon period, the form of vessel discussed so far, with clinker
planking on a backbone of keel, stem and stern post, was referred to as a
'ceoP or 'keel'. The term 'ceol' seems to have evoked images of the shape of
the stem posts: 'high keel', 'steep keel'. Charters from London from the years
1000 and 1030 introduce the term 'hulc' to denote a type of ship which
differs from the keel (Ellmers, 1972: 59). 'Hulc' in Anglo-Saxon meant
'hollow' or 'cavity'.
Ship archaeologists have applied the term 'hulk' to ships which share
specific hull structural characteristics, rather than interpreting it as a general
name for ships of a certain size or shape. The word continued to be used
throughout the Middle Ages and it has been assumed that the ships termed
'hulk' in the fifteenth century had developed from those termed 'hulk' in the
eleventh century. The characteristics of the hulk have been deduced entirely
from iconographic evidence and the sole link between the verbal and the
iconographic data for the whole of the period is the town seal of New
Shoreham of 1295 (figure 1.4). The seal's inscription reads 'hoc hulci signo
vocor os sic nomine digno - By this sign of a hulk am called Mouth which
I
1972: 144; Brindley, 1938: fig 33) and that of the Provostry of Southampton
of about 1300 (Brindley, 1938: fig 15). The Hulksmouth ship has a thin
Shipbuilding: traditions of the northern seas 1 1
Figure 1.4 The late thirteenth-century town seal of 'Hulksmouth' or New Shoreham
(National Maritime Museum).
crescent-shaped hull with planking running parallel to the lower and upper
edge and finishing at the platforms for the castles, not at the stem and stern
posts as is the case with the keel type. The position of the nail heads in
relation to the edges of the strakes suggests that the hull was built in 'reverse
clinker' technique - that is, with each strake fastened to the inside of the
strake below, not to the outside as with normal clinker. This is shown more
clearly in the picture of Henry I's ship, drawn before c.1140, shown in figure
1.5. In both images, the ends of the strakes appear to be 'gathered' at
'collars', or bands across the ends of the strake runs at bow and stern. In
pictures pre-dating 1350 these collars are often shown surmounted by animal
head carvings but they were gradually replaced by castles, as on the New
Shoreham Seal, from the end of the thirteenth century.
Having established the basic characteristics of the hulk, earlier represen-
tations were sought out in back to the origins of the
order to follow the trail
L£U £
<o c. t vv
Scheldt in about 1180. On the Winchester font (figure 1.6) most of the
strakes end at a 'collar' at stem and stern and one of them ends at the sheer
line (upper edge of the planking). This Tournai connection served to direct
the search for the origins of the hulk to the Low Countries. Coins minted in
Dorestad in 814-840 show vessels propelled by both oar and sail with
'banana'-shaped hull profiles. A boat found in 1930 at Utrecht, not far from
Dorestad, has repeatedly been cited as the archaeological specimen of the
early form of the hulk. The hull has a log-boat base with two strakes on
each side joined to each other by a half-round wale. The small mast step
near the forward end of the boat was most probably for a towing-mast
rather than for sailing. A radiocarbon date of 790 ± 45 ad was obtained for
the boat in 1960 and it was claimed that the arm of the river in which it lay
was closed about 866. However, recent work has revised this dating and
shown the boat to be rather later, with a probable date of sinking in the
eleventh century (Vlek, 1987: 66). A very similar vessel was found in 1974 at
Waterstraat, another site in Utrecht, and has also been dated to the eleventh
century.
These boats from Utrecht are extended log-boats for use on rivers, not
designed to be while the much earlier vessels depicted on the
sailed,
Carolingian coins are seagoing sailing ships. Clearly the Utrecht boat can no
Shipbuilding: traditions of the northern seas 13
Figure 1.6 A hulk on the Winchester Cathedral font, made in Belgium in about 1180
(National Maritime Museum).
longer be cited as an example of the early form of the hulk. Indeed Vlek
rejects the idea that the Utrecht type has anything at all to do with the hulk.
He concludes that 'this kind of vessel cannot have been the prototype, either
in construction or in form, of the late medieval large seafaring hulk' (Vlek,
1987: 89).
The intense focus on the Low Countries may have diverted attention from
the hulk's very strong French connections. The Dorestad coins referred to
above are copies of those of Quentovic, a French seaport just south of
Boulogne. Although the reference collection of pictures of medieval ships
compiled at the National Maritime Museum contains only a very small and
probably unrepresentative sample of the images created, it does appear that
France has a considerably higher proportion of hulk pictures than other
countries, followed at some distance by Britain. Several of the British images
are of the return of Thomas a Becket from exile in France (e.g. frontispiece
and figure 3.2) and are conceivably intended to represent French ships.
What are the constructional implications of the features of the hulk, the
rounded hull, the run of the often reverse clinker planking to the sheerline?
The most basic problems of wooden ship construction are how to bend and
fashion straight planks to create a hull and how to fasten the ends of the
planks in a secure and watertight way. If a clinker vessel is dismantled and
its strakes are laid out flat it becomes clear that they are not straight strips.
14 Medieval ships and shipping
Figure 1.7 Conjectural diagram of the run of the lowest strakes in a hulk (National
Maritime Museum).
With keels, the strakes taper to the post and, because of the fullness of the
hull, tend to rise higher at the stems than in the mid-part of the hull. There is
that whereas the steps of the normal planking would resist the flow of the
water, slowing down the vessel and inviting leaks through the plank seams,
reversing the plank steps helps solve these problems. It is puzzling then that
after 1300 the instances of representations of reverse clinker are heavily
outnumbered by those of normal clinker-building. This may perhaps have
been related to an increase in size which made the friction of the plank laps
less significant.
To make a strong hull, the strake runs need to end on a rigid component.
The 'collars' depicted were probably breast-hooks of some sort but until a
hulk is discovered and excavated this aspect of the construction must remain
a mystery. It has been suggested that the ropes sometimes depicted coiled
round the 'necks' of hulks were for binding the ends together. This would
have been structurally weak and it is far more likely that they represent
mooring lines or anchor cables.
Until the middle of the fourteenth century, most pictures of hulks show
hulls which were similar in shape and in the run of the planking at both bow
and stern. Some arc shown with narrow stems and wider sterns and there are
(
Figure 1.8 The cog-like vessel on the town seal of Ipswich of AD 1200 (National
Maritime Museum).
1962, with the discovery of the Bremen cog in the River Weser in Germany,
that the constructional features of the cog were revealed (Fliedner, 1964;
Lahn, 1992). The ship appears to have been lost while it was still being built
and was perhaps swept away from the berth where it was being fitted out in
about 1380. Tree-ring studies have shown that the oak for the planking and
cross-beams was felled in the Weser Hills to the south, in 1378. The crooks
for the knees were from more local sources in the forests around Bremen
(Lahn, 1992: 19). The superstructure had not been fully installed, the ship
had not been ballasted and the only artefacts associated with the ship were a
selection of shipwrights' tools, a half-finished anchor stock, a barrel of tar
and a shoe. The excavated remains of the Bremen cog are remarkably
complete and show that the ship was 24 metres long, eight metres in the
beam and over four metres high. A long programme of conservation and
research has elucidated the building sequence and provided enough
information for the construction of an accurate replica (figure 3.7).
Shipbuilding: traditions of the northern seas 17
The keel was made up of three parts scarfed together. The central part was
8.3 metres long, with a maximum width of 0.47 metres. The two end pieces,
3.73 metres and 4.78 metres long, were each made out of a trunk of timber
with a branch coming out at an angle so that the change of angle between
the ends of the keel and the stem and stern posts was worked out of solid
timber. The stem and stern each consisted of an inner and outer post. The
inner posts were scarfed to the inner side of the keel knee but the outer posts
were not mounted until the planking was in place.
IS Medieval ships and shipping
The forward section of the keel was not completely level but was raised
The sides of the keel and the lower parts of the stems were rabbeted
slightly.
for the hull planking. The first four strakes to each side of the keel were
flush-laid. That is to say, they edge to edge and were not fastened
were laid
to each other in the midships area. At the ends of the ship the planks go
through a transition from flush-laid to clinker and this required very
specialised joinery including the stepping-out of the plank edges. The ship
had 12 strakes per side, each made up of three or four planks. The plank
seams were coated with caulking compound before assembly. When the third
strake was in place, further caulking was applied and inserted on the inside
and held in place by oak strips fastened on by iron clamps or staples. The
first four floors were inserted and held in place with small nails while the
treenail holes were bored and the treenails hammered in. The fourth strake
turned the bilge, then the remaining floor timbers, ending below the fifth
strake, were put in place.
The keelson, which had an integral mast-step, was treenailed to the floors
but not to the keel. Planking continued and, after strake eight, half of the
futtocks and five heavy cross-beams were inserted (figure 1.10). The cross-
beams were cut from oak trunks seven metres in length. They rested on
Shipbuilding: traditions of the northern seas 19
futtocks which had tenons projecting from their tops. Strake nine had cut-
outs to fit over the cross-beams. When
was in place further
strake ten
framing was inserted and inner planking was fitted, which gave longitudinal
strength. This was cut from less good and preused stock. The twelfth strake
ended on the framing rather than running right to the posts and the channel
wale for attachment of the lower end of the shrouds lay against it outboard.
The washboard planks above strake 12 were not clenched. A framework of
top timbers was left protruding in the area of the stern castle. The castles,
decking, windlass and capstan will be mentioned in Chapter 3.
The outer stem reinforcement was held on by iron bolts. It gave protection
to the strake endsand the fore-stay was fastened to it. The outer stern post
was much lighter than the outer stem post. It had gudgeons for the rudder
already mounted on it, showing that the cog was sufficiently complete to
have been launched. Further construction could be carried out afloat.
The most significant features of the hull of the Bremen cog are the use of
sawn, rather than radially split, planking and the construction of the
bottom with flush-laid planking and its transition to clinker. Also notable
are the upward-sloping keel and the elaborate jointing of the stem and
stern posts. The use of caulking battens held on by distinctive clamps or
staples seems to be diagnostic of the cog and the planking nails were not
clenched on roves but hammered over so that their tips entered the
planking again. Through-beams are known also from vessels in the keel
tradition, such as the 'big ship' from Bergen, but they seem to have been
essential to the cog. These structural features add detail to the general
visual impression of hull shape obtained by studying the pictorial evidence.
They also allow fragmentary finds which show evidence of similar features
to be classified as cogs; though it is not necessarily the case that all cogs
would have shared all these features. It is not known how far back in time
these features can be traced, nor over what geographical area. It is by no
means certain that cogs built in Britain or Ireland would have exhibited
precisely the same constructional traits as those built in the Hanse
homelands.
The lines of theBremen cog (figure 1.11) go some way to counter the
common misconception that cogs were unseaworthy boxes. The hull has a
fine entry, with a narrow bow to cut through the water and a deep forefoot
to maintain directional stability. The stern has a fine run, allowing the water
to flow past the rudder with the minimum of turbulence. The angular turn of
the bilge helps to resist heeling and leeway.
The earliest ship find yet identified as a cog is that which sank at Kollerup,
on the northern coast of Denmark, in the early thirteenth century, shown in
figure 1.12 (Jeppesen, 1979; Crumlin-Pedersen, 1979: 29-32). This did not
have sawn planking but was instead constructed from very wide oak planks
probably made by splitting a tree-trunk in half and using axes to cut a plank
from each half (Crumlin-Pedersen, 1989: 32). Six strakes either side of the
keel were flush-laid, with a transition to clinker planking at the ends of half
of them.
Numerous other boat and ship finds have been identified as cogs on the
grounds of having the lowest part of the hull constructed of flush-laid planks
and by the presence of re-entrant nails and caulking clamps. Among these is
20 Medieval ships and shipping
Figure 1.11 The lines of the Bremen cog (after Lahn, 1992).
Figure 1.12 The Kollerup cog under excavation (Viking Ship Museum, Roskilde).
Shipbuilding tools
parts of the ship other than the hull planking at least as early as the end of
the thirteenth century and sawyers were employed in the 1294 English royal
galley-building programme. Toolmarks on the probably fifteenth-century
assemblage of boatbuilding timbers found at the Poole Foundry site show
that stems were sawn out of tree-trunks. For planking, however, English
shipwrights seem to have used split, not sawn, oak boards throughout the
medieval period. The Grace Dieu, built in the early fifteenth century, has saw
marks on the sides of some of the frames but the grain of the planking shows
that it was still split radially.
Knives were used for cutting treenails and for fine-trimming of other
components. Augers were used to make treenail holes and a breast auger is
shown being used for this purpose on the Bayeux Tapestry (figure 1.3).
Clench nails also needed pre-bored holes, as they could split the planking if
they were driven in blind. Larger nails, especially when of extreme
dimensions, such as those for the fifteenth-century Grace Dieu which had
square-sectioned shanks 2cm across, certainly needed them. Treenails were
driven in with mallets and had wedges knocked into their inboard ends
before they were trimmed off flush with the inboard face of the hull. Clench
nails were hammered in and then, while they were held in position, pre-
punched roves were forced onto them. An axe hammer or a dolley, a tool
like a hammer with a hole in the middle, may have been used for this. Then
the end of the nail was hammered flat to clench against the rove and the end
was cut off.
Shipbuilding: traditions of the northern seas 23
Building sites
from merchants with connections with the Low Countries and the Baltic. The
mast and yard, oars, ship's boat, anchors and hawsers were also bought in.
The master and mates from the galley at Southampton went to Poole to buy
a sail and bought a mast and yard, expensive items, from the Isle of Wight.
In the final stages substances for coating the hull were bought, along with
pots in which they were prepared. At Southampton, rosin and grease were
introduced in the fifteenth week, followed by tallow, pitch and lard in the
sixteenth. Payments were made for painting the galleys and the Newcastle
account gives the most information about their decorative treatment. The
galley built there was whitewashed with lime inside the hold. The outside of
the hull was sized with egg white and oil and more than a dozen pigments
were used, as well as 18 pounds of varnish. Tinfoil was put over some of the
nailheads before painting (Whitwell and Johnson, 1926: 157). The
Southampton account records the purchase of clavis de tin in the tenth week,
by which time they would have been above the galley's waterline, and clavi
stannati (tinned nails) are also mentioned in the York account. Some of the
nailheads from the planking on the Dublin waterfront were covered in a blue
metal oxide and a yellowish-white substance coating the outboard surface of
the possible galley planking from the Dublin waterfront was analysed and
found to be composed of pine resin and calcium carbonate (McGrail, 1993a:
137). While the Southampton galley was being painted, 20 men were paid
wages for six days to make a ditch to haul the galley to the sea.
Henry V's great ship Grace Dieu was built, from 1416, in a specially made
'dok' surrounded by an enclosure of stakes. It is likely that the ship was built
on a low-lying strip of land with a temporary wall on the seaward side of
the dock. This would allow launching to take place at high tide after the wall
had been removed (Friel, 1993: 5). After the launching, the shores, piles and
stakes were sold off. William Soper, the Keeper of the King's Ships, who was
responsible for the construction of the Grace Dieu, built a storehouse at
Southampton in 1416. It was 126ft long, made of New Forest timber with a
tiled roof. It had an adjoining smithery, which may have been where the
nails for the great ship were manufactured. Carpenters and sawyers were
paid more than clenchers who in turn were paid more than holders and
labourers. West Country men had been forced to work on her and
absconded back home during the winter of 1416.
There is a potential medieval shipbuilding site, as yet unexplored, at
Smallhythe on the River Rother in Kent. Large vessels were built there in the
fifteenth century, when it was a thriving port, but by about 1600 silting had
reduced it to a hamlet. Clench nails have been retrieved from a rectangular
depressed feature, approximately 40 metres by 7.5 metres, which may be the
remains of a dock.
The only site associated with medieval boatbuilding which has so far been
examined in Britain is in Poole, in Dorset. Excavations on the Foundry site
on the Poole waterfront in 1987 revealed a store of boatbuilding timber
which had been neatly laid out on the medieval beach, not far from the
parish church, in around 1400. The excavation may not have uncovered the
total area of the boat yard site but the assemblage recovered consisted of 61
timbers. Of these, 11 had previously been used in boats. Most of the rest
were unworked timbers, with their bark removed, clearly selected on grounds
Shipbuilding: traditions of the northern seas 25
Figure 1.14 The medieval boatbuilder's timber store at Poole, Dorset (Poole
Museums).
«
Medieval ship pictures indicate that from about 1350 the clear divisions
between the keel, hulk and cog ship-types were breaking down. In docu-
ments a greater number of specific vessel-types are mentioned, all except
carracks presumably derived from one of the main clinker traditions, but
large numbers of vessels were simply referred to as ships (naves). Northern
European ships at this date were still single-masted and now had stern
rudders and more developed castles. While the length to beam ratio of
seagoing ships in each of the traditions was probably similar, there was
increasing differentiation in the shape of the bow and stern. There may
already have been some imitation of the hull shape, though not yet the
method of construction, of the ships from southern Europe which
northerners called carracks.
At this time, hulk hybrids, in which the strakes end on a post at one end
of the ship and at the sheerline at the other, become common. The 1418-26
seal of Thomas Beaufort (figure 2.1a) shows a ship with a hulk stem and a
keel stern. with hulk planking at both ends also continue to be
Ships
depicted. The seal of the Earl of Rutland, 1391-98, has hulk planking at
both ends, with the stern wider than the stem. The ships shown on the
fifteenth-century Admiralty seals of John Holand, Earl of Huntingdon, the
first of 1421 and the second (figure 2.1b) of 1435-42, also have hulk
planking at both ends.
Pictures are the prime source of information about the changes which were
taking place but they leave many questions unanswered. It is relatively easy
to identify ships built in the hulk tradition because the pictures often clearly
show However it is far less
the run of the planking to the ends of the ship.
easy to distinguish cogs from keels because the method of planking the
bottom is not depicted. The ships on the seals of Amsterdam of the early
fifteenth century might be presumed to be cogs but the hull shape is very
similar to that on the contemporary seal of Southampton, where the keel
building tradition was flourishing and the Grace Dieu was then being built.
The wreck of the Grace Dieu, which began building in 1416, lies in the
River Hamble in Hampshire, near Southampton (figure 2.2). It has been the
subject of several archaeological excavations, and the work so far has been
summarised by Friel (1993) and Clarke et al. (1993). This royal warship had
a keel at least 38 metres long, and was built as the English answer to the
carrack, such as those from Genoa used by the French. Grace Dieu was not
built according to Mediterranean skeleton-building practice, which will be
described below, but in an extension of the English keel clinker-building
technique, using planking three layers thick. This unique triple-thickness
Shipbuilding: interaction and innovation 29
30 Medieval ships and shipping
stage further and make strakes three planks thick, the problems would be
compounded (figure 2.3c). The nails and treenails would be excessively long
and the shearing pressures on the nails would be far worse. The planking
would not be uniformly strong as it would vary from six planks in thickness
at the strake overlaps to three planks thick in the mid-part of the strakes. So
this is not how Grace Dieu was built and it is not what is meant by 'triple-
thickness planking' in the context of this find. Grace Dieu was constructed in
a modified form of double-thickness planking (figure 2.3d). There were two
planks of full height and another plank, two-thirds the height of the other
two, on the inside of the strake. This provided extra thickness and reduced
the variation so that the hull was three planks thick in the mid-part of the
strakes and five planks thick at the overlaps. Most importantly, the inner
planks also served to relieve the nails by transmitting the downward pressure
onto the tops of the planks of the strake below. Watertightness would also
be improved since water would have further to penetrate and would not
simply have to run along a straight seam. This method of increasing the
thickness of the planking overcame many of the technical problems. Some
still remained, however. It was difficult to avoid boring the treenail holes too
close to the edge of the joggles. There was still a huge requirement for iron,
as the shanks of the clench nails were 2cm square and 15cm long, spaced
about 20cm apart along the strake overlaps, with an extra line of nails,
spaced about 60cm apart, along the midline of the strakes. The supply of 17
tons of 'clench and roof nails for the Grace Dieu and the smaller vessels
built at the same time is recorded in the building accounts (Carpenter
Turner, 1954).
It cannot have been easy to shape and assemble all the separate
components. The planking seems to have been made up of pieces not more
than six or seven feet (less than two metres) in length (Anderson, 1934: 165).
So far it has not been determined whether each layer of planks was scarfed
at the same place or whether the scarfs were staggered. Some clues as to how
the shipwrights might have set about fastening the layers of a strake together
and to the strake below can be seen in planks removed from the wreck
Shipbuilding: interaction and innovation 31
Figure 2.3 Clinker planking: (a) normal single thickness; (b) conjectural double
thickness; (c) conjectural triple thickness; (d) Grace Dieu (National Maritime
Museum).
(Clarke et al., 1993: 25-33). Small nails in these planks indicate that while
the composite strakes were being built up, the outer two layers of planking
were tacked from both inboard and outboard. The only small nail on an
inner plank occurs at a scarf. Tacking was necessary because the holes for
the 20cm square shanks of the clench nails had to be prepared. Holes were
quite frequently made in the wrong place or at the wrong angle and they had
to be plugged with wooden pegs.
A ship which was wrecked at Aber Wrac'h on the north coast of Brittany
in the first half of the fifteenth century (L'Hour and Veyrat, 1989) shares
some characteristics with Grace Dieu. Particularly striking are the deep rove
impressions on the inside of the planking, the concavities on the frames to
32 Medieval ships and shipping
accommodate the roves and the use of moss and sphagnum caulking. The
ship was large, with an original length estimated as 25 metres. The keel was
of beech, with the stem, framing and planking of oak. The building accounts
for the Grace Dieu record the purchase of more than a thousand beech trees,
though it is not known what they were used for (Friel, 1993: 4). The after
part of the Aber Wrac'h ship (figure 2.4) was not preserved. The hull
construction was of normal single-thickness clinker planking, with localised
strengthening at the places where the planking was pierced for the ends of
the through beams (figure 2.5). The strakes below and above the beams
(though not the ones mostly cut away by the beams) were reinforced by
small boards on the inside face, in the manner of the inner plank of the
Grace Dieu strakes. The weight of each beam was mainly supported by a
futtock and a stringer. The scarves between the keel and the stem and stern
posts were both missing from the wreck. The part of the stem post which
was present showed that the strake ends ran onto the stem post in the keel
rather than the hulk manner.
Pottery found on the wreck was of Breton manufacture; and six Castilian
and two Breton coins indicate that the ship probably sank in the first quarter
of the fifteenth century (L'Hour and Veyrat, 1989: 294). Tree-ring study,
although inconclusive for dating, indicated that the ship was perhaps built
near Bordeaux or on the north coast of Spain. There was a substantial
shipbuilding industry at Bayonne in the fourtenth and fifteenth centuries,
supplying English as well as local clients. While Grace Dieu was being built,
another great ship was built for Henry V at Bayonne. The accounts indicate
that it was of clinker construction.
Grace Dieu and the Aber Wrac'h wreck provide evidence for ships built in
the mature development of the keel tradition. The closest parallel to them is
the fifteenth-century wreck raised from the sea-bed off Gdansk in 1975-6
(figure 2.6). The Baltic is an area in which the cog was prevalent and the
Gdansk W5 wreck may provide an indication of the course ship construction
took when cogs were superseded. The town seals of Gdansk and Elblag for
this period show a change of ship type from the classic cog, with straight
stem and stern posts, to more rounded hull shapes.
Gdansk W5 has similarities with Grace Dieu and the Aber Wrac'h wreck
in the general appearance and dimensions of the components, including the
Figure 2.5 Aber Wrac'h wreck - diagram to show the strengthening or the hull
planking where it is pierced by the through-beams (M. L'Hour).
developed a large shipbuilding export industry and by the end of the century
was supplying English, Dutch and Italian buyers.
A model from the church of Ebersdorf near Dresden (Christensen,
votive
1987: 69-70) appears to represent a cog in a similar state of evolution to
that of Gdansk W5. The hull is round and beamy, clinker-built from the
keel, with a straight stern and curved stem post.
A ship found at Sandwich in Kent during pipe-laying operations in 1973 is
potentially important for the study of later medieval clinker ship construc-
tion. The shiptentatively dated to the fifteenth century on stratigraphic
is
evidence and with its keel five metres below the current ground surface in
lies
the silted-up medieval town ditch (Trussler, 1974). Although some frames
and planking, together with parts of the rudder, were removed at the time of
discovery, much of the clinker-built hull, which may be as much as 33
metres long, remains in the ground.
Shipbuilding: interaction and innovation 35
Figure 2.6 The intact part of the Gdansk W5 wreck after being raised from the
sea-bed (photo: L. Nowicz).
hull construction used in southern Europe was adopted in the north and
began to replace clinker construction for the largest ships from the mid-
fifteenth century.
The number of medieval ships so far excavated in southern
Europe it not
great, so regional and functional variations and chronological developments
are still being explored. Pictorial evidence is much more abundant and Italian
treatises on shipbuilding, largely concerned with galleys, survive from the
fifteenth century. Here it is appropriate only to make some generalisations in
order to show something of the technique which was to impact so very
significantly on northern Europe.
Ships of southern Europe had been following an entirely separate course of
development from those in the north. By the Middle Ages their planking was
flush-laid and the planks were not joined to each other but just to the
frames. This technique is known as 'carvel-building' and contrasts with
clinker-building in which the planks are fastened to each other, edge to edge.
The clinker method is termed 'shell-first' construction while its opposite is
'frame-first', where a complete skeleton of frames is erected, totally defining
the form of the hull, before the planks are applied to it. However, just as
northern clinker vessels only had to have some of their planking in place
before frames were inserted, southern carvel vessels only had to have some of
their framing planking began.
in place before
In antiquity, Mediterranean vessels had been shell-built, not with
overlapping strakes but with the planking of their hulls flush-laid and keyed
together with tenons locked into mortises in the edges of the planks. This
method persisted at least until the seventh century, as can be seen from the
Yassi Ada ship (Bass and Doorninck, 1982), by which time frame-first
construction had been introduced, as in the Saint Gervais II wreck (Jezegou,
1985). Frame-first construction in the eleventh century is demonstrated by
the Ser^e Liman vessel, which sank off Turkey in about 1025. Steffy (1991)
has determined that when planking began, only two full frames and eight
floor timbers were in place on the backbone of keel and stems, spanning only
the central 2.7 metres of the approximately 14.36 metre long hull. The
bottom strakes were nailed to these timbers, then framing and planking of
the sides progressed together. Planking the difficult area at the turn of the
bilge was left until some of the side planks were in place and was achieved
with mostly short planks of irregular shape.
A well-preserved vessel dating to about 1300 was discovered in 1898 at
Contarina in the Po delta, roughiy 50 kilometres south of Venice (figures 2.7
and was 16.5 metres long on the keel, flat-bottomed and very nearly
2.8). It
double-ended, tapering slightly more quickly aft than forward. The keel was
fairly slenderand shallow, with the floors and the keelson laid above it. To
below the turn of the bilge, at the area of overlap
either side of the keel, just
between the floors and the futtocks, were longitudinal members of similar
scantlings to that of the keel. Stringers were inserted at the corresponding
positions on the inside of the hull so that the ends of the floors and futtocks
were sandwiched between the bilge keels (or bilge wales) and the stringers.
Higher up, where the futtocks overlapped with the top timbers, there was
again a wale paired with a stringer. The keelson had mast-steps for lateen
main and fore sails (Bonino, 1978). This disposition of keel, bilge keels and
Shipbuilding: interaction and innovation 37
Shipbuilding: interaction and innovation 39
wales would give the hull longitudinal strength and reinforce potential areas
of weakness in the framing. The bilge keels and wales may also be vestiges of
up
the hull design process as the shape of a hull could be defined by setting
the keel and stems, establishing three key frames and running battens along
the sides of this framework from stem to stern.
Les Sorres X, a 10 metre long vessel found during the digging of the
Olympic Games rowing canal near Barcelona in 1990, is dated on pottery
evidence to the second half of the fourteenth century (Nieto, 1992). It shares
the constructional features of the Contarina 1 ship, except that the bilge keel
was reinforced by a double stringer on the inboard face of the hull. Similar
features can also be seen in the Lasize wreck, a Venetian galley which sank
in 1439 Lake Garda (Scandurra, 1972: 209-10). The bottom of the hull
in
has been excavated and indicates that the overall length of this galley was
about 39.5 metres. The cross-section of the remains shows a deep projecting
keel bolted through the floor to the keelson. At the overlap of the floors and
futtocks there is a thicker strake or wale but this time, instead of projecting
outwards, it is inset into a notch on the underside of the framing. A stringer
lies directly above, sandwiching the floors and futtocks.
Figure 2.9 The lower hull planking of the fifteenth-century Mataro model (Maritiem
Museum 'Prins Hendrik', Rotterdam).
process, was developed after the period under study, notably in England in
the late sixteenth century. It was not adopted in the Mediterranean until the
And from that time on the Genoese, Venetians and Catalans have made a practice
of navigating with cogs and have been leaving off use of the big ships to navigate
more safely and because they are less expensive. This has been a great
transformation in the shipping of our fleets. (Giovanni Villani, Chronica)
In fact the Genoese may have come to know this ship-type earlier because
of their dealings in low-value bulk cargoes likealum and grain. In 1286 a
navis sive cocha belonging to a Bayonne merchant in partnership with two
Genoese brought alum to Flanders. In 1292 another cocha loaded with 285
tons of alum at Focea to go to Bruges. The first mention of a cocha in the
Venetian documentation is from March 1312 (Balard, 1991: 119). Southern
shipwrights did not adopt northern clinker-building, however, but instead
created a flush-laid interpretation of the beamy and high-sided northern hull.
Because of this, coche could be built much larger than cogs. They also
adopted the square sail and the stern rudder, although coche de duobus
timonibus - that is, with paired side rudders - also feature in the Genoese
documentary sources until about 1375. The crucial characteristic identifying
a cocha in the earlier part of the fourteenth century appears to have been the
use of a square sail, with or without a supplementary lateen sail on a second
mast (van der Merwe, 1983). From 1310 lateen ships were rapidly
abandoned in favour of coche (Balard, 1991: 120) but the term cocha was
itself being replaced at the end of the century by navis, as coche types came
to be considered the norm. One- and two-masted coche are shown on the
Pizzigani chart of 1367 (figure 2.10) (Guilleux la Roene, 1957: 179-80). A
particularly good late example is shown on a Hispano-Moresque bowl, made
near Valencia in about 1430 (figure 2.11). To emphasise that ship
42 Medieval ships and shipping
called carakes' (van der Merwe 1983: 126, citing Burnham). There is good
documentary evidence up to 1396 that carracks were, in English sources,
ships which the Genoese referred to as coche or sometimes naves (Burnham,
1974: 269-78). 'Carrack' seems to have been a northern vernacular term for
large coche and had very little Mediterranean currency in the late medieval
period. The fact that the word can be traced back to thirteenth-century Spain
as a term for large (lateen) ships and that many Spaniards were involved in
Genoese coche voyages to England may be significant here (van der Merwe,
1983: 125-6).
The rounded hull shape we now associate with carracks is shown in
single-masted form in a French illustration of the siege of Constantinople,
dating to about 1400 (figure 2.12). However, as Pizzigani's chart shows, the
two-masted rig of square mainsail and lateen mizzen was then already
current. It is from this that the new three-masted rig was developed, by the
addition of a square-rigged foremast. Although this rig was to spread
throughout Europe from the fifteenth century, its exact point and date of
origin remains elusive. The earliest known illustration of a three-masted rig
(figure 3.10) is in a Catalan document said to date to 1406 (Mott, 1991:
111) but in practice the development may have occurred almost
44 Medieval ships and shipping
simultaneously at various places along the route from Italy to the Channel at
the opening of the fifteenth century.
Throughout the medieval period it was well known that north and south
Europe had different approaches to ship construction. Crusaders from the
north came in contact with carvel-built (see page 36) ships in the
Mediterranean and northern ships were trading beyond the Straits of
Gibraltar at least from 1300. Merchant ships from the Mediterranean visited
northern Europe regularly from the late thirteenth century and southern
warships were hired to fight in the north. Clinker-building never was
adopted in the Mediterranean but carvel-building became the standard
method of constructing large ships in northern Europe by the end of the
Middle Ages.
Carvel galleys had been built in northern France at the end of the
thirteenth and in the fourteenth centuries. Philip the Fair's galley arsenal, le
clos des galees, was founded at Rouen in 1293 and employed Genoese
shipwrights to direct local craftsmen in building galleys of the Genoese type
(Rieth, 1989). Italian specialists were also brought in to caulk the carvel
planking and to make oars. But as Rieth has pointed out (1989: 74-5),
although carvel-building was introduced at Rouen the method was not
effectively transferred to French shipwrights since they did not apply it to
any other ship forms.
Eight Genoese carracks were captured by the English in 1416 and 1417
and soon afterwards permission was sought to hire 'carpenters and caulkers
of a foreign country' to repair the carracks, 'for in this country we shall find
only a few people who know how to repair and amend them'. Payments
were made to Venetians, Catalans and Portuguese for carrying out the work
(Friel, 1983a: 131).
The term 'carvel' was applied in the fifteenth century to skeleton-built
ships, specifically those from Portugal where in the early fifteenth century an
apparently new family of ships, termed 'carvels' was developed (Paviot,
1991: 55). From the 1430s, carvels were being built in northern Europe, at
first by southern master craftsmen and then by the local shipwrights who
had learnt their skills. Between 1463 and 1466 a three-masted carvel was
built for Sir John Howard at Dunwich in Suffolk (Friel, 1983a: 134) and it is
quite possible that this was not the earliest to be built in England.
The carvel method was finally adopted in northern Europe because of its
suitability for building very large ships. There are limitations to the size of
clinker hulls. The larger the ship, the thicker the planking has to be to
achieve rigidity and this puts a great deal of stress on the nails, potentially
leading to problems with watertightness. Clinker construction of large ships
was extravagant with iron, as a great number of long and thick clench nails
had to be used. Carvel-building used fewer, less-massive nails, relying more
on treenails. Clinker planking in the 'keel' tradition was made from radially
split oak, requiring large trees, with straight grain free from knots. Cog
builders had already economised by using sawn timber and carvel-building
also used sawn planks, allowing the use of poorer quality timber stock. A
Shipbuilding: interaction and innovation 45
Figure 2.13 The ship found at Roff's Wharf, Woolwich in 1912 (after the
contemporary drawing by W.E. Riley).
46 Medieval ships and shipping
In the early part of the medieval period, ship's holds were covered by loose
boards, termed 'hatches'. The date of introduction of fixed decking is
unclear. The hatch nails itemised in the accounts for the galleys of Lyme and
Newcastle, built in 1294-96, indicate that these vessels had fixed decking,
perhaps in the castles. The main deck of the Bremen cog of about 1380 had
been partly laid, made of oak boards more than 3cm thick. Although no
traces of nailing were found it is assumed that, except for one board which
has a finger hole, the deck boards were intended to be nailed down. The
stern-castle deck was nailed down and caulked. The fore castle had a raised
gangplank rather than a full fore-castle deck (Lahn, 1992: 117-26). On the
early fifteenth-century Catalan Mataro model the main deck is laid with
planks running fore and aft and has a pronounced camber. The single
hatchway which gives access to the hold has high coamings to prevent water
from running in, and is capped by a sloping hatch cover (Culver, 1929: 215).
The stern-castle deck also has considerable camber but is laid with planks
running diagonally from a central fore-and-aft beam. The decks, like the hull
planking, are caulked with hemp fibre and payed with pitch.
Superstructure seems first to have been built on ships for defensive
purposes and Chapter 9 will discuss the introduction and development of
fighting platforms and castles. By the fourteenth century, castles were integral
parts of the hull and in the Bremen cog the framing of the hull extended
above the sheer to support the castles (Lahn, 1992: 90). Castles also
provided accommodation for those on board. Provision of cabins is recorded
as early as 1228, when a ship sent to Gascony was fitted with a chamber for
the king's 'things' and in 1242 chambers were constructed with panelling in
a ship for the king and queen's voyage to Gascony (Salzman, 1931: 230).
From the fourteenth century the south-west ports set up a thriving trade
transporting pilgrims to the shrine of St James in Compostella. They could
carry as many as 200 passengers at a time (Oppenheim, 1968: 20). A poem
about a voyage on a pilgrim ship, preserved in a fifteenth-century manuscript
at Trinity College Cambridge, has the owner of the ship calling a carpenter
to make cabins and little compartments here and there (Anderson and
Anderson, 1926: 91-5).
Most ships carried either a windlass or a capstan and larger ships had both.
A capstan has a vertical axis and is driven by men walking round the central
drum. It is stronger and faster than a windlass, which has a horizontal axis
and is worked by on levers on the drum. The Bremen cog, of the
pulling late
fourteenth century, was equipped with both a windlass and a capstan in the
48 Medieval ships and shipping
Figure 3.1 The early fourteenth-century town seal of Winchelsea (National Maritime
Museum).
for the Grace Dieu lists three cork buoys and five wooden buoys called
'dobles' (Friel, 1993: 9).
therefore more efficient, causing less drag and needing less effort to keep it at
the required angle, than the stern post rudder which is unbalanced because it
swings from its leading edge. Rudders mounted on the stern post were also
less effective because of the turbulence caused by the ship's hull, whereas side
rudders operated in clearer water. The adoption of stern rudders was a
necessary adaptation to changes in ship design and increases in vessel size
which provided greater carrying capacity at the expense of sailing
performance. The side rudder works well on a ship which is itself in balance,
with the centre of effort of the sail in the correct position in relation to the
centre of lateral resistance of the hull and with trim achieved by the careful
positioning of cargo and ballast (Andersen, 1986). As ships increased in size
the fine tuning of balance became harder.
There is a limit to the height of ship on which a side rudder can be used.
The whole length of the rudder has to be cut out of a single tree-trunk. The
longest side rudders yet found are a thirteenth-century example from the
Bryggen site in Bergen, Norway, which measures 6.70 metres (Christensen,
pers. comm.) and another, coincidentally of the same length, from Rye Bay,
East Sussex (Marsden, 1992: 126-7). Radiocarbon dating of the Rye rudder
suggests a date of manufacture in the twelfth or thirteenth century. Size
Fitting, steering and rigging 51
Figure 3.2 A hulk from a Life of St Thomas of Canterbury, c.1230-40 (J. Paul
Getty).
Bergen were made in two parts, with a narrow extension attached to the
afterside of the blade (Christensen, 1985: 152-6, 229). The 4.1 metre long
Rebseek rudder, an undated find from Kolding Fjord in Denmark (Solver,
1946), has traces of a similar feature. They may have been made this way
because the size of the parent log was not sufficient to allow them to be
made in one piece, or perhaps rudders were fine tuned to match the
performance of the ship they had to steer. The side rudder trawled up from
the sea-bed off Rye, East Sussex, mentioned above, has no holes either for
the tiller or for attachment to the ship (Marsden, 1992: 126-7). It is
possible, therefore, that it was not completed. Its blade is rather narrow and
it may have been intended to have a supplementary piece joined to it.
halyard hole, to help the rope to run smoothly and to reduce wear on the
mast, which was made of softwood.
The mast of the Grace Dieu was said to be 200 feet (nearly 61 metres)
high and seven feet (2.13 metres) in diameter at the deck (Friel, 1993: 17).
The great size of the mast means that it must have been of composite
construction. The mast of the Woolwich ship, which may have been the
Sovereign of 1488, shows how composite masts could be made. It consisted
of a spindle of pine surrounded by baulks of oak bound together with iron
bands (figure 2.13). The diameter was 1.32 metres (Salisbury, 1961: 85).
'Wooldings' or rope bindings around masts begin to appear on pictures of
large ships in the fourteenth century.
In vesselswhich did not have fixed decking amidships the mast could be
lowered and supported on a mast crutch. Mast crutches, sometimes referred
to as 'mykes', are Y-shaped, cut from forks of timber. Two large mast
crutches have been found in contexts dating to around 1200 on the Dublin
waterfront (McGrail, 1993a: 104, 109). They could have held masts of
75cm and 85cm. Smaller mast crutches have been found at Newcastle
(O'Brien et al., 1988: 104-6) and at the Poole Foundry site (Hutchinson,
1994: 29, 33).
Yards, the wooden spars onto which the sails were laced, are not
surprisingly elusive in the archaeological record. Broken or redundant ones
could be put to other uses or burnt as firewood and those on shipwrecks
would most probably be washed away. The same applies to bowsprits,
which seem to have been introduced in the thirteenth century to provide an
attachment point for a line led forward from the windward edge of the sail.
They supplemented but did not replace 'luffs' which were poles used to
brace the 'tack', the windward bottom corner, of the sail. The yard was
held against the mast by a parrel, a U-shaped piece of wood with a hole
towards each end for the bindings to the yard. The dimensions of the inside
of the curve of the parrel can give some indication of the size of the mast.
Two parrels, predating 1200, excavated on the Dublin waterfront, were
from ships with masts less than 50cm in diameter (McGrail, 1993a: 77).
Rope parrels with wooden beads threaded on them to act as rollers to
reduce friction as the yard was raised were in use by the beginning of the
fourteenth century. One is depicted in a French manuscript of 1307
(Villain-Gandossi, 1979: 217). By the fifteenth century parrels with several
rows of beads, spaced by strips of wood termed 'parrel ribs', appear in
pictures of large ships. Parrel assemblies of this type have been recovered
from the stores of the Mary Rose, which sank in 1545 (Rule, 1982: 140-
41, 145).
The which
'standing' or fixed rigging of a sailing ship consisted of stays,
supported the mast longitudinally, and shrouds, which supported it laterally.
The forestay, running between the bow and the top of the mast, prevented it
from tipping back. One or more backstays from the masthead to the stern
counteracted much of the force of the wind in the sail. The halyard, used to
haul up the yard and sail, could also be led aft and made fast as an
additional backstay. Holes for the shrouds are sometimes found on the
excavated remains of the upper parts of the sides of boats and ships. The
Bremen cog had a wale treenailed to the outside of the planking with 14
Fitting, steering and rigging 57
holes drilled through it for rigging. Remains of rope were found in all of
them (Lahn, 1992: 109). From at least as early as the fifteenth century,
shrouds could be set up on chains, as accounts and inventories for English
royal ships show (Friel, 1993: 9). As a result, the parts of the side of a ship
where the shrouds were attached came to be termed the 'chains', 'chain wale'
and 'channelling'. Wooden blocks called 'deadeyes' were used for tensioning
the shrouds. The 'running' rigging is that used to hoist and control the sails.
The principal elements of the rigging are marked in figure 3.6 on a
photograph of a bench-end from the Chapel of St Nicholas, King's Lynn.
The carving was probably carried out during the rebuilding of the Chapel
which was completed in 1419.
Viking Age ships in Scandinavia used woollen sails and rope made from
lime bast and walrus hide. It is possible that hemp rope and canvas sailcloth
were in use in Britain from the beginning of the medieval period but there
were some reversions to earlier practices for specific purposes or for local
craft. For example, the galley La Phelipe built at Lynn in 1336 had walrus
hide cords for the parrel (Tinniswood, 1949: 308) and woollen sails were
used in Kent in the late fifteenth century (Fenwick, 1978: 251).
From the thirteenth century or earlier, shipbuilding timber and hemp and
flax for cordage and sails were imported to supplement indigenous supplies.
The rope industry at Bridport was active from before 1200 and in the
fourteenth and fifteenth centuries Bridport supplied the ships of Plymouth
and Southampton (Lane-Poole, 1956). Much of the flax brought from
Ireland and landed at Bristol may have been destined for the Bridport rope
and sail industry. Some cordage also came from the Baltic. In 1226 there was
a market at Lothingland near Great Yarmouth selling nets and rope
(Salzman, 1931: 136). Cordage has been recovered from thirteenth-century
contexts at Newcastle, including a thick three-ply cord of vegetable fibre and
a thick, eight-strand plait of hair (Walton, 1988: 80-81). The hair rope has
strands of different pigments and compares with braided ropes found at
Bergen, worked from black and white goat-hair yarn.
It is known that Brittany canvas was brought back with cargoes of salt
from the Bay of Bourgneuf (Bridbury, 1955: 82); but evidence for sails in
medieval Britain so far comes only from documentary and pictorial sources.
Shipbuilding accounts sometimes specify the quantities of canvas bought for
a sail but uncertainty about the size of the 'ell', used to measure the cloth,
and the possibility that the cloth was used in double thickness create
problems in interpretation (Anderson, 1976: 48-50). When sailing in strong
winds vessels have to reduce their sail area to maintain stability. In the
earlier part of the medieval period this was done by reefing: taking a tuck in
the sail using short cords or ties, called 'reef points', sewn into it. This
literally shortens the sail to reduce its area and also allows the yard to be
lowered down the mast, so that the wind exerts less leverage or heeling force
on the vessel. Reef points are shown on seals from the twelfth century
(Brindley, 1912: 130). In the mid-fourteenth century a new approach to
changing the sail area was adopted for large ships. Instead of reducing the
area of a large sail when the wind became too strong, ships carried extra
strips of sail, called 'bonnets' which could be laced to the foot of the sail in
light wind conditions.
58 Medieval ships and shipping
14
15
16
17
4
18
5
19
6
7
20
8
21
9
10
22
11
23
12 24
25
13 26
Figure 3.6 The rigging of the ship on the King's Lynn bench end, c. 1415-20
(National Maritime Museum).
Square sails are so-called not just on account of their shape but because they
are rigged so that, when the wind is not acting on the sails, the yard and sail
hang at right angles (or square on) to the long axis of the vessel. The sails
are symmetrical about the vertical axis and either side can become the
leading edge depending on the vessel's heading in relation to the wind. By
contrast, fore-and-aft sails - for example, modern yacht sails and the lateen
sails of medieval Mediterranean ships - are set along the length of the vessel.
They have one leading edge and a trailing edge and are usually asymmetrical.
In sailing, the plane of a square sail and its yard are turned to make best use
of the wind. It is a misconception that the sails of medieval ships acted
simply like windbags and were only useful for propelling ships downwind.
By leading one edge of the sail forward to the stem post and sheeting the
other aft, the sail could behave in the same way as a fore-and-aft sail. When
the wind meets a sail at an angle, it is forced to change its direction. A
deflected air flow forms around the curved surfaces of the sail, with a
different speed and pressure on its two sides. The pressure on the windward
side increases while the pressure on the leeward side decreases, so that the
sail is being both pushed and sucked along at the same time. The suction, or
forward, side of the sail is in fact much more effective than the pressure side
and it yields the greater part of the driving force. It is which makes it
this
possible for a sailing vessel to make progress to windward by 'tacking' or
zig-zagging towards the wind. Sea trials with replicas of Viking ships have
shown that they can sail a course at approximately 55 degrees from the wind
direction. Leeway, the effect of the wind and swell pushing the vessel
sideways, means that the course made good is about 60 degrees from the
wind. At this angle the vessel has to sail twice the actual distance as the crow
flies (Andersen, 1986). Oared warships and small merchant vessels could use
oars as auxiliary power to make progress against head winds. The use of
oars will be discussed further in Chapter 9.
the hull and so normally located about half way along the keel. The force of
the wind on the sails is transmitted to the hull primarily through the mast.
The effect of having the mast stepped forward of the centre of lateral
resistance is that wind pressure on the sail will make the ship tend to turn
away from the wind. The rudder has to be turned at a considerable angle to
correct this tendency, which is termed 'lee-helm'. It can also make it difficult
to tack; that is, to change course by turning the head of the vessel through
the eye of the wind. If the mast is stepped aft of the centre of lateral
60 Medieval ships and shipping
Figure 3.7 The Kiel replica of the Bremen cog undergoing sea trials (photo: Jochen
Sachse).
resistance, wind on the sail will make the ship tend to turn its bows towards
the wind unless it is constantly corrected by the rudder. This is known as
'weather-helm'. Excessive lee-helm or weather-helm can both lead to the
dangerous situation of being 'taken aback', with the wind blowing against
the wrong side of the sail and driving the ship stern-first downwind.
Large ships became equipped with two masts in the mid-fourteenth
century. Some had a mizzen mast stepped near the stern; others had a
Fitting, steering and rigging 61
foremast. The Pizzigani chart of 1367 shows a cocha or early carrack with a
square-rigged mainsail and a lateen-rigged mizzen (figure 2.10). The earliest
two-masted ship in the English royal fleet was a carrack, probably Genoese,
acquired in about 1410. By 1420 the Crown had 13 two-masters, many of
them foreign-built carracks but including at least five English-built ships
(Friel, 1983a: 132). The single-masted rig is the most efficient way of
producing propulsive power, as sails always interfere with the wind-flow on
each other, but there were important advantages in dividing the sail plan.
Increases in the size and weight of the single mast, yard and sail required
more and stronger rigging to hold up the mast, control the yard and adjust
the sail, also placing great strain on the central part of the keel. Far more
importantly, the addition of extra masts aided steering. When a ship turns, it
pivots about its centre of lateral resistance, a point roughly half-way along
its keel. A sail set on a mast stepped at one end or the other of the keel will
exert a strong turning force when the wind blows against it. Most two-
masted ships appear to have had a square mainsail and a fore-and-aft lateen
mizzen at the stern, the typical carrack rig at the beginning of the fifteenth
century. Others, including a high proportion of hulk types, had the square
mainsail with a little square foresail. Pictorial evidence suggests that the
foresail was used on hulks rather than cogs.
Figure 3.9 shows the effect of the addition of fore and mizzen sails during
62 Medieval ships and shipping
Figure 3.9 Diagram to show the effect of the addition of masts on tacking
performance: (a) single-masted vessel; (b) vessel with mainmast and mizzen; (c) vessel
with mainmast and foremast; (d) three-masted ship (National Maritime Museum).
put hard over to bring the head of the ship into the wind. The sail is then
backed, to bring the head round onto the new tack. This has the effect of
driving the ship astern until it has turned sufficiently for the sail to be swung
quickly around and set on the other side of the ship. This is the dangerous
stage of the operation. If the tack is executed smoothly the sail fills with
wind and the ship gathers speed on the other tack. If it is not, the ship can
be swept helplessly down wind, at best losing a lot of ground and at worst
foundering on a lee shore.
The addition of a foresail (figure 3.9b) does little to increase the speed of
the ship but does make tacking more reliable. By backing the foresail the
it
head of the ship can be turned quickly and the foresail can maintain the
ship's heading while the mainsail is swung around. Adding a lateen mizzen
Fitting, steering and rigging 63
64 Medieval ships and shipping
relatively undisturbed air flows high above the ship. It provided extra driving
power and helped to lift the bows, counteracting the depressing effect of the
spritsail (McGowan, 1981: 13-14). Inventories of Henry VII confirm that
these sails were in use in 1485 and that large ships already had a fore-topsail
and a fourth mast, behind the mizzen (Oppenheim, 1896).
By combining the carvel-built hull with the three- (or more) masted rig, a
ship-type had been created which was not, unlike all those before it, a
regional product. It represented a fusion of technology and, with local
variations of detail, was built and used all over Europe.
4 British horizons, foreign ships
Shipping is essentially international and throughout the Middle Ages ships of
many different regions and shipbuilding traditions rubbed up against each
other in harbours all over Europe. The interchange intensified as the volume
of trade and the length of trading voyages increased. The following survey
sketches in the context of maritime contact between Britain and the other
regions of Europe, the flow of trade and the impact of war. Against this
background the seagoing ships belonging to the various regions are
described. The map (figure 4.1) shows places mentioned in the text.
Ireland
There were large fleets of ships in Ireland before the Norman invasion in
1169-70, notably those of the Ostmen, the Hiberno-Norse of Dublin and
Wexford. In the eleventh century there was trade between Dublin and the
ports of Bristol and Chester. This included the export of English slaves to
Ireland. By the first half of the twelfth century there was substantial seaborne
trade between England and Ireland. Ireland also imported wine from Poitou
and Bordeaux in exchange for hides and pelts.
Henry II granted Dublin 'to his men of Bristol' in 1171-72, probably in
recognition that the economies of Bristol and Dublin were mutually
important already. In 1174 the King granted Dublin freedom from toll,
passage and custom throughout the King's realms and this was a consider-
able stimulus to trade. These privileges were later extended to other Irish
ports. Customs records for the period 1171 to 1250 show a predominance of
food exports, especially cereals. Wheat and oats, beans, cheese and bacon are
listed.
the keel tradition and their features are closely paralleled in contemporary
Scandinavian boats and ships. Three individual vessels can be recognised
from the groups of hull planking: TG10, a late eleventh-century boat; TG9
with TG6, a late twelfth-century large ship (figure 4.2); and TG3, a
thirteenth-century large ship (McGrail, 1993a: 93). Tree ring studies have
shown that the components which have been sampled match the Dublin
master chronology. McGrail (1993a: 87) takes the view that the majority of
the timbers were from boats and ships dismantled and reused close to the site
where they were built. Some were unfinished, lacking fastening holes, and
this supports the argument that there was a building yard close by. The size
of vessel represented in the archaeological record increases over time. This
may be partly accounted for by the fact that the vessels are from different
sites, the later ones from Wood Quay and the earlier ones mostly from
Figure 4.1 Map of Europe (National Maritime Museum).
British horizons, foreign ships 67
Pimirp
rigurc ^.i
il 1 K pv
ivcy
Norway
The between the North Sea ports and Norway was established before
traffic
the Conquest. According to tradition, Bergen was founded about 1070 and it
rapidly became one of the largest medieval ports of northern Europe.
Artefacts traded from England have been found in the earliest layers. In the
twelfth century Norway had active seaborne contact with Britain and
Germany. King Sverri made a speech in Bergen in 1186 comparing the merits
of German and English traders. The Germans brought wine in such
quantities 'that it was no dearer than ale' and took away butter and dried
fish. By contrast, traders from England, Orkney, Shetland, the Faroes and
Iceland brought useful products: wheat, honey, flour, cloth, linen, flax, wax
and cauldrons. After the fire which destroyed much of Bergen in 1198,
English pottery gains predominance in the archaeological record, taking over
from continental imports (Herteig, 1959: 182-4). In the thirteenth century,
England supplied grain to Norway in return for dried fish, whale oil and
falcons, hawks and furs. The Hanse was also supplying grain to Norway and
there was a triangular trade: grain was shipped from Liibeck to Bergen, cod
from Bergen to Boston in Lincolnshire, and cloth from Boston to Liibeck.
Nedkvitne (1977 and 1985) has written extensively about medieval trade
between Norway and England.
In the rebuilding and extension of the wharf which followed a fire at
Bergen in 1248, ships' timbers were reused as building materials. They
provide important evidence for the type of large trading vessel used in early
thirteenth-century Norway. Christensen (1985) has demonstrated that many
of the timbers came from one very large ship which was perhaps quite new
when it was broken up. The timbers show little wear or use and fire damage
to a deckbeam and other components suggests that the vessel may have been
a casualty of the 1248 blaze. The use of pine and some similarities with
other finds, such as the thirteenth-century Norwegian Sjvollen boat
(Christensen, 1968), suggest that it was built in Norway. The principal
component found was the keelson (figure 4.3). It is 12.5 metres long, made
of two pieces scarfed together, and it has notches on its underside for 27
floor timbers spaced about 50cm apart. Parts of 19 cross-beams from just
above the keelson were also recovered. The cross-beams sat closely above the
floor timbers with the keelson in between, to give a very strong bottom to
the hull. The keel would have given the bottom a sharp profile, unlike a cog.
One deck cross-beam with rabbets for deck-boards was recovered and also
the drum of the massive windlass, 5.40 metres long with a diameter of about
50 cms, for raising the mast, yard and anchor. Fragments of at least two
other ships of comparable size were found at the Bryggen, represented by
cross-beams with protruding heads.
Christensen estimates the minimum total length of the vessel to have been
around 30 metres, with a maximum beam of between nine and ten metres
70 Medieval ships and shipping
Figure 4.3 The keelson of the 'big ship' from Bryggen, Bergen (Bryggens Museum).
and depth in hold of about three to five metres. This would provide a
minimum cargo space of 165 cubic metres (Christensen, 1985: 182;
Christensen, 1989: 18-19). The larger trading ship from Skuldelev had
about 30 cubic metres of cargo space in the hold and this may serve to
illustrate the increase inbetween the eleventh and the thirteenth
trade
centuries. It also goes to show that cogs were not the only large cargo
carriers trading to the north. Ships built in the Scandinavian keel tradition
could be of equal or larger capacity.
The Isle of Man and the Western and Northern Isles of Scotland were part
of the kingdom of Norway. In 1266 Norway rented Man and the Western
Isles to the Scots but the Northern Isles continued to owe allegiance to
Norway until the fifteenth century (Mitchison, 1970: 33, 76). The
shipbuilding tradition of north Britain was Nordic and timber for building
ships was imported from Norway.
At the end of the thirteenth century Norway surrendered control over
shipping to the Hanse. The Bergen trade fell largely into the hands of Liibeck
grain merchants. Customs accounts, which use 'buss' as the usual term for a
Norwegian cargo vessel, show that trade with England continued.
A Norwegian wooden calendar staff for 1457 (figure 4.4) has pictures of
four different large ships incised on it. The clearest of them is very likely to
be the earliest precisely dated three-masted ship representation known from
northern Europe. Leibgott (1973) has postulated that the calendar may have
belonged to a seafaring merchant.
72 Medieval ships and shipping
From the early thirteenth century the trade of northern Europe was
dominated by the Hanse. Hamburg and Liibeck controlled the ends of the
overland transit route across the base of the Danish peninsula and in the
early thirteenth century these towns formed an alliance with other trading
towns of northern Germany and the North and Baltic Sea coasts. This
became known as the Hanse or Hanseatic League and its function was to co-
ordinate the commercial initiatives and protect the interests of its members.
By organising in this way, the Hanseatic towns were able to defeat outside
competition. For example, at the beginning of the thirteenth century
Denmark had looked set to predominate in Baltic trade but the rise of the
Hanse 1980: 162) and Hanseatic trade
effectively put a stop to that (Unger,
in the Baltic expanded strongly between 1250 and 1300.
The overland route from the North Sea to the Baltic, by way of Hamburg
and Liibeck, was of importance in the twelfth and thirteenth
crucial
centuries, as trading ships seldom risked the dangerous passage around the
Skaw at the north of Denmark. In the Viking period, ships had avoided the
Skaw by sailing through the sheltered waters of the Limfjord which cuts
through the Danish promontory near its tip. However, in the twelfth century,
while ships were increasing in size, the Limfjord was silting up and it
eventually ceased to function as a through channel (Crumlin-Pedersen, 1972:
190).
The Baltic area exported grain, iron, copper, silver and lead, timber, tar
and hemp, flax, wax and fish. In return it obtained wool and
pitch, furs,
cloth, manufactured goods, wine and salt. French salt was needed to
supplement local supplies for the extremely important Baltic herring trade.
By 1262 tolls were levied on salt from the Bay of Bourgneuf at the
transhipment point at Hamburg. The 1292 shipping law of Hamburg
concerns journeys to La Rochelle for wine and salt. By the end of the
thirteenth century, English, Flemish and Frisian ships were shut out of the
Baltic. From 1294 Norway, defeated by a Hanse blockade, conceded the
right to carry all trade from ports south of Bergen. By 1300 the Hanse had
reached a peak of prosperity and control over northern trade.
The cog was the principal type of ship in use in the region. As early as
1197 German cogs went to rescue Christian prisoners from Beirut (Scammell,
1981: 77). This, incidentally, puts Villani's statement that cogs made their
firstappearance in the Mediterranean in 1304 into perspective (see page 41).
Town seals dating to the first half of the thirteenth century from Elblag,
Staveren and Wismar and from the second half of the century those of
Harderwijk, Stralsund and Gdansk, show ships with high sides, straight stem
and and no castles (Ewe, 1972). Castles make their
stern posts, stern rudders
first appearance on the town seals from 1300 onwards.
Several cog-finds dating to the thirteenth century have been found around
Denmark and in the Baltic. The cog found at Kollerup on the northwest
coast of Jutland (page 59) may have been
attempting to make its way into
the Limfjord when it was wrecked (Crumlin-Pedersen, 1979: 27-31; Moller,
1980). The Kollerup cog is narrow (about six metres) in relation to its length
(about 22 metres) and its mast is stepped very far forward (figures 1.14 and
British horizons, foreign ships 73
3.8).The cargo space was about eight metres long with partitions fore and
aft.The finds include a tally-stick and pottery. A cog found on the east coast
of Sweden at Bossholmen near Oskarshamn has been dated by
dendrochronology to the late thirteenth century (Cederlund, 1990; Adams,
1990). Its find-spot lies on the sailing route between Denmark and Estonia
(Cederlund, 1989).
In the early fourteenth century there was vigorous trade between the
Hanse and England. Hanse ships imported mixed cargoes, including timber
from Gdansk and some Rhenish wine, and exported cloth from Boston and
other ports. English kings gave the Hanse privileges in exchange for loans
(Scammell, 1981: 67).
Early in the fourteenth century, ships began to use the sea route around
the Skaw into the Baltic (the 'Umlandfahrt') to avoid the delay and expense
of transhipment at Hamburg. This caused the decline of the overland route
and of Liibeck. From the mid-1300s the Germans were trading to Poitou and
Gascony. By the late fourteenth century German merchants handled Polish
surplus grain through Baltic ports, shipping wheat and - even more
important - rye, as well as timber and herrings to urban centres of western
Europe, especially in Flanders. To have a full load of back cargo they loaded
salt (high volume but low value) at the Bay of Bourgneuf and cloth (high
value but low volume) in Flanders. The Hanse Bay Fleet was inaugurated in
the 1370s to import large quantities of Bourgneuf salt into the Baltic.
The cog of Bremen, already described (figures 1.13 and 3.7), sank in about
1380. At about the same time another cog was wrecked at Vejby, Denmark
(figures 4.5 and 5.6). Coin evidence indicates that the Vejby cog was built in
around 1350 in the Elblag region and sank soon after 1377. Contents of the
wreck suggest that it was voyaging from western Europe (see Chapter 5).
This cog was originally between 16 and 18 metres long, with a beam of five
to six metres. The bottom was flush-laid, with clinker sides above. The
strakes were of sawn oak. Only the bottom of the ship, consisting of the
outer planking, the floor timbers, the ceiling planking and the keelson, had
survived (Crumlin-Pedersen, 1976; Crumlin-Pedersen, 1979: 24-9; Crumlin-
Pedersen, 1985: 71-3).
By 1388 there was an English factory in Gdansk, set up by English
merchants whose aim was to secure in the Baltic privileges similar to those
enjoyed by the Hanse in England. Meanwhile Hanse trade with western
Europe was growing. In 1405 three Hanse ships sailed to Bruges with nearly
500,000 furs. In the fifteenth century, cloth made up 90 per cent of all
German exports from England, 6,000 to 12,000 pieces a year (Scammell,
1981: 50-52). The salt trade with France was very important in the fifteenth
century and in 1449 a major international incident occured when English
privateers captured the Bay Fleet. This consisted of 60 Hanse ships and 50
from the Low Countries (Postan, 1933: 127-30). The fleet was intercepted
between Guernsey and Portland and taken to the Isle of Wight (Hattendorf
et al., 1993: 30-31). The Low Countries ships and their contents were
released but the privateers kept the Hanse property. The Hanse retaliated by
confiscating English merchandise from the factory in Gdansk. A truce was
declared but English privateers captured the Bay Fleet again in 1458. Even
this did not cause lasting disruption to trade, however, and in 1471 Edward
British horizons, foreign ships 75
IV was restored to the throne with Hanseatic support and in 1474 a treaty
signed with the Hanse led to confirmation of their rights in the steelyards at
London, Boston and King's Lynn (Scammell, 1981: 49-67).
The wreck of the fifteenth-century trading ship Gdansk W5, described in
Chapter 2 and Chapter 5, provides some evidence for the measures that
Hanse shipowners had to take to protect their vessels from piratical attack.
Stone shot was recovered as well as cross-bow bolts, probably from the
weapons of the crew, found between the frames (Smolarek, 1979: 62).
Low Countries
with the Baltic and reduced that with France. War with France at the
beginning of the century meant that Flanders came to depend on Baltic grain.
The fourteenth century saw a decline in the Flemish cloth industry and
F.nglish merchants moved into that market. At the same time Dutch towns
developed a textile industry producing heavier, cheaper cloth better suited to
the market and climate in the Baltic. Dutch merchants voyaged round the
Skaw to the Scanian market rather than dealing with Hanseatic middlemen.
In the later fourteenth century the Dutch were rivalling the Hanse carrying
British horizons, foreign ships 77
trade, with salt among their principal cargoes. In 1384 Dutch towns, which
did not belong to the Hanse, were excluded from the Scanian market which
caused them to deal directly with other Baltic ports.
The presence of English merchants in the Baltic in the fourteenth century
had diverted Hanse attention from the more serious encroachment by the
Dutch, who had effectively destroyed the Hanse monopoly of Europe's
commerce with the Baltic by 1500. By 1498 Dutch-owned ships carried 50
per cent of the value of Gdansk imports. In the 1440s the Dutch had started
trading French and Portuguese salt into the Baltic, not just importing it for
their own fisheries. As we have seen, nearly half of the Bay Fleet which was
captured in 1449 was made up of ships from the Low Countries.
France
France is the country with which Britain had the most maritime contact
during the Middle Ages. Even before the Conquest, tolls and imports show
that there was a high level of cross-Channel trade with France as well as
with the Rhineland. As a result of the Norman invasion there was constant
traffic between the Norman landholders who had estates both in England
and in France. Le Patourel (1976: 164) estimates that William the Conqueror
made the Channel crossing about 17 times in the 21 years between his
coronation and his death. At the beginning of the thirteenth century the
Angevin kings of England held more than half of the land of present-day
France. After the loss of Normandy in 1204 (and with it Anjou, Maine,
Touraine and Poitou) and Brittany two years later, Bordeaux and Gascony,
which were vitally important to England as a source of wine, remained
subject to the English crown until 1453.
Northern France had by the eleventh century adopted Norman Viking
shipbuilding techniques. It is possible that hulk-building was established
earlier and that hulks continued to be developed as seagoing ships as well as
inland craft. Thirteenth-century stained glass in Chartres and Notre Dame
cathedrals depicts fishing vessels with characteristic hulk planking and a
1317 illustration of wine arriving at the port of Paris (figure 5.3) shows the
barrels being transported in small hulks.There is very little evidence for the
shipbuilding practices western and southern French coasts. It is
of the
possible that they were unaffected by Viking influence and continued Gallo-
Roman traditions. Henry IPs acquisition of Aquitaine in 1152 meant that
there was a regular traffic of English-built ships to the south-west of France.
The Association Bayonne merchants controlled Biscay from the early
of
thirteenth century despite a private war with the seamen of the ports of
Normandy. It is unclear what their ships were like. However, for the last
few years a watch has been kept on a wreck of about 1300 in the harbour
of St Peter Port, Guernsey, which is suffering gradual erosion as a result of
large vessels passing above it. The wreck has yet to be fully investigated but
is about 30 metres long and has yielded large quantities of Saintonge
pottery from the Bordeaux area of France (see page 99). This site has
exciting potential for the study of medieval shipping and the Bordeaux wine
trade.
78 Medieval ships and shipping
Throughout the thirteenth and early fourteenth centuries the trade between
Bordeaux and England remained the virtual monopoly of the Bayonnese,
with Southampton and Bristol as the chief English centres of the Gascon
wine trade. At its height, England imported more than 20,000 tuns of wine
each year from Gascony, or about a quarter of the region's total export. The
Hundred Years War between England and France, which began in 1336-37,
had an immediate effect on the Gascon trade, reducing exports of wine to
less than a quarter of previous levels (James, 1971: 15). It appears that the
volume of trade picked up after the first few years of the war. In the 1350s
and 1360s Gascony was still a major market for the expanding English cloth
industry. Because Gascony produced wine to the neglect of other agricultural
products some foodstuffs had to be imported. In the 1370s English ships
took food to Bordeaux and Bayonne and in 1372 Froissart recorded the
arrival of 200 British and Irish ships. Ships sailed in convoy, that is to say,
they were sailed in organised groups under escort of armed ships (Waters,
1957: 1-3) and this lasted until 1500 in the case of the salt fleets from
Bourgneuf Bay.
Despite the Hundred Years War, trade between Britain and France
continued, often handled by neutral middlemen. For example, in 1367 stone
for Rochester Castle was brought from Caen and in 1414 Nottingham
alabaster was being exported for the Abbot of Fecamp (Salzman, 1923: 87,
97). The loss of Gascony to the French king in 1453 created problems with
safe conducts for English ships and much of the handling trade went to
foreign merchants.
After the loss of Normandy and Brittany, the ports of Rouen and La
Rochelle were developed as bases for the French royal fleet. The clos des
galees was established at Rouen in 1293 for the construction, repair and
storage of war galleys which were intended Channel and the
to control the
route to Gascony (Rieth, 1989). The were carvel-built of Genoese
galleys
design but clinker-built Norman barges and galleys up to 30 metres long
with oars and a square sail were also constructed there. It was of
considerable importance during the Hundred Years War, until destroyed by
fire in 1419 (Sumption, 1990: 156-7). The French also boosted their
seapower by hiring contract fleets of Italian galleys and carracks.
Medieval depictions of French ships show a considerable variety of hull
shapes (see, for example, Villain-Gandossi, 1978; 1979). A large number of
them have hulk features and there are very few cogs. The only medieval
clinker-built ship to have been excavated in France has been dated by coins
on board to the first half of the fifteenth century. The wreck was found in
1985 at the Aber Wrac'h river mouth on the north coast of Brittany (figure
4.7). The tree ring curves obtained from the ship's oak planking best match
those from south-west France or the north coast of Spain (L'Hour and
Veyrat, 1989). Ten metres of the beech keel remained, with a detached oak
stem preserved to a length of 2.5 metres. Parts of 24 oak clinker strakes,
23cm wide and 3cm thick, were preserved on the port side. The vessel had
heavy frames, 15cm to 25cm wide, closely spaced. The depth of the frames
at keel level is 25cm, decreasing to 15cm. There were at least 52 frames on
each side as well as the six cant timbers piled up in the lower part of the
wreck. About half of the second futtocks were fitted before the lower ones so
British horizons, foreign ships 79
Figure 4.7 The Aber Wrac'h wreck (M. L'Hour - photo: Frederic Osada).
that their upper ends were correctly positioned to support the cross-beams.
On the outside of the hull the cross-beam heads were protected by cone-
shaped pieces of wood, made of elm and alder, fitted in front of the beam
with their tips pointing towards the stem. Part of the mast-step support was
preserved. Ceiling planking was treenailed through the frames to the outer
planking, up to the turn of the bilge. The thicker ceiling planks were in effect
stringers, joggled over the frames, and thicker stringers were located higher
up in the hull.
The excavators estimate that the overall length of the ship was about 25
metres with a maximum beam of eight metres. They have suggested a link
with the documented wreck of an English-owned ship which sank when
entering the Aber Wrac'h river in 1435 (L'Hour and Veyrat, 1989: 298).
Atlantic Iberia
By the end of the eleventh century and the beginning of the twelfth century,
Normans were using Iberian north-western ports during their involvement in
the Crusades. Chronicle evidence suggests that at the same time Saracen raids
from Seville and other southern ports were ravaging the Galician shore. The
Archbishop of Compostela sent to Italian cities for shipwrights. In 1115
first
two galleys were built by Genoese craftsmen for the defence of Galicia and
another was built in 1120 under the direction of a Pisan (Filgueiras, 1991).
So by the mid-twelfth century, when Lisbon was captured from the Moors,
seagoing ships of northern Spain and Portugal belonged to two different
80 Medieval ships and shipping
visiting Southampton. It may have been the organisation of the Cinque Ports,
along with their corporate identity, that the merchants of the Castilian ports
were keen to emulate, since by the end of the thirteenth century they
themselves had formed an association.
During the Hundred Years War, Castilians were among the privateers
attacking English ships. In 1350 a Castilian trading fleet was attacked by an
English fleet as it sailed out from the harbour at Sluys in Flanders. The
Spaniards lost 14 ships. This battle, which became known as 'Les Espagnols
sur Mer' is described in a contemporary chronicle. Unfortunately this gives
us no information about the appearance of the Spanish ships other than that
they were larger than the English ones.
From the thirteenth century England also had close trading connections
with Portugal, which was a source of olive oil, kermes dye, wine, fruit and
salt. By the 1300s Portuguese ships were voyaging both to the Canaries and
Mediterranean Spain
Figure 4.9 The Mataro ship model (Maritime Museum 'Prins Hendrik').
The votive ship model from Mataro (figure 2.9 and 4.9) of a substantial
Catalan trading ship of the early fifteenth century has been described in
Chapter 2. The 1.3 metre long model appears to have been two-masted
originally, with mainmast and mizzen. A third mast was added at a later
date, inserted through a hole crudely made in the foredeck (van Nouhuys,
1931: 343). Although recognising that this is not an exact scale model, de
Groot (1984) has analysed the stability and carrying capacity of a ship 16,
18, 20 and 22 times its size. He calculated that a ship of this form would
have to carry ballast equivalent to 20 per cent of its total displacement. On
British horizons, foreign ships S3
ship, that on a bowl now in the Victoria and Albert Museum and dated on
stylistic grounds to about 1425 (figure 4.10), was probably manufactured in
Malaga and was produced by Muslims living in medieval Europe. The ship is
depicted with the arms of Portugal on the mainsail and can be assumed to
belong to the southern European carvel-building tradition. Since Muslim
84 Medieval ships and shipping
ships are not known to have sailed into British waters they will not be
discussed in this book but the interaction of shipbuilding traditions on all
Italy
In the thirteenth century the cloth industry of Italy relied on imported wool,
especially that from England and Scotland. The wool was shipped across the
Channel and then taken overland to Italy until the sea route via Biscay and
the Straits of Gibraltar became an established alternative towards the end of
the century. Disruptions to trans-continental trade routes caused by Franco-
Flemish wars probably created an incentive for Italian merchants to adopt
the sea route. The first voyage of Genoese ships to Bruges and Southampton
took place in 1277-78. The Venetians followed this lead in the early
fourteenth century. By 1300 Genoa and Venice were both at a peak of
commercial prosperity and were also at war with each other throughout the
fourteenth century (Scammell, 1981: 110-11, 165).
In exchange for the wool exported from England and cloth from northern
Europe, the Venetians brought luxury goods from the Mediterranean and
from sources in the east, traded through Constantinople and the Black Sea.
Because they dealt in high-value, low-volume cargoes much of their trade
was carried in galleys (figure 4.11), of which they developed a specifically
merchant type with a cargo capacity of about 150 tons (Lane, 1934: 1-34).
The merchant galleys were effective sailing craft and the use of their oars was
largely restricted to entering and leaving harbour. Sailing in a strict, state-
controlled and seasonal convoy system, heavily manned and defended by
paid mariners, their high operating costs were justified by the value of their
cargoes and their extremely predictable voyage times.
The Genoese controlled the alum mines at Focea in what is now Turkey.
Alum was used in the cloth industry for fixing dyes and was increasingly in
British horizons, foreign ships 85
It was of astonishing size, full of treasures, which might easily have supplied the
needs of all the country; but the London merchants, so it is said, having in their
houses much stale merchandise of fruit, spices of various kinds, and so forth,
oil
lest these be thrown away on the arrival of fresh goods, made an agreement with
the Genoeses [sic] to leave the port and sail to Flanders.
4,
the carriage of French wines as well. This trade was mostly handled by
Breton and Spanish shippers but in 1485 the king ordered that wines of
Gascony and Guienne were to imported only in ships of England, Ireland or
Wales (James, 1971: 49). In this he was unsuccessful and the coasts of
Britain continued to be frequented by large numbers of foreign ships.
5 Carrying trade
Most medieval were built to carry cargoes on trading voyages.
ships
Specialised warships were very rare until the end of the period and most
ships used in war, for fighting and as troop transports, were merchant ships
requisitioned or chartered from their owners. The cargoes that merchant
ships carried, the routes they sailed and the arrangements for loading and
unloading them all had consequences for ship design.
Much is known about medieval trade because it was subject to a high
degree of government intervention, for raising revenue and pursuing foreign
policy objectives. Extensive documentary records survive and this is just as
well, since archaeological evidence alone would give a very distorted view of
the nature and relative quantities and importance of commodities traded. The
majority of them - wine, salt, grain and other foodstuffs, wool and cloth -
were organic and these goods leave few traces on the land sites which were
the destination of their travels. However, those which failed to reach their
destinations are rather more likely to be preserved on shipwreck sites.
Durable commodities, such as pottery and items of metalwork, did not make
up whole cargoes and it is not always easy to determine whether they were
imported as objects of trade or were personal belongings or gifts. Building
stone is the only commodity where the source of imports and the volume of
the trade can be reliably established from the surviving material evidence.
The customs accounts of England, unparalleled elsewhere in Europe except
for Italy, are an invaluable source for the study of trade from the late
thirteenth century onwards (see Clarke, 1983). Customs were first introduced
on a regular basis in 1275 and applied to the export of wool, woolfells and
hides by both native and alien merchants. In 1303 additional taxes were
introduced on imports as well, to be paid only by alien merchants. The
customs system is explained by Gras (1918). There are long lists of dues on
merchandise - ranging from bears to peppercorns - in the records of many
seaports, for example Norwich, King's Lynn, Hull and Southampton. Other
documentary sources for trade (and for disruption to trade) include laws,
tolls, aulnage accounts for cloth, charter parties between merchants and
Commodities
Packaging
The way in which cargoes were packaged for transport by sea and their
stowage in ships has been largely overlooked. It was certainly a technically
developed area, whatever the jumble of casks, bales, sacks and loose
metalware or pottery shown in the fourteenth-century illustration, figure 5.1,
would suggest to the contrary.
Wine was transported in casks, most commonly in tuns, though smaller
capacity casks were also used especially for high-value wines and brandy.
The tun, or the space it occupied, was the standard unit for measuring the
carrying capacity of a ship. It is not a simple matter to pin down the exact
tun as units of measurement varied from place to place and showed
size of a
a marked tendency to grow larger throughout the Middle Ages. This was, to
a large measure, in response to taxation. Taxes were usually levied on the
gross unit (e.g. the tun of wine, roll of cloth, chalder of coal) so by
increasing the size of the gross unit, the tax per sub-unit (gallon of wine,
yard of cloth etc.) was proportionately less (Salzman, 1923: 18, 240).
In 1423 it was enacted by statute that the wine tun should contain 252
gallons. The gallon itself was not defined but can be expected to have been
no larger than the Tudor standard of about 3.78 litres. On this basis the
90 Medieval ships and shipping
Figure 5.1 A 'ship full of all kinds of merchandise', redrawn from BL Sloane Ms
3983, f.!3r, second quarter of the fourteenth century (National Maritime Museum).
capacity of a fifteenth-century wine tun would have been about 954 litres
(Zupko, 1977: 29-30). Gascon tuns seem to have contained from 750 to
900 litres (Renouard, 1953, cited in James, 1971: xvi). A late twelfth-century
cask (figure 5.2), reused to line a pit at Goldsmith Street in Exeter, has an
approximate capacity of 815 litres (Allan, 1984: 313). Its height is
reconstructed to 1.42 metres and its maximum diameter to about 0.94
metres. By comparison, a mid-eighteenth-century tun of 252 gallons (by then
equivalent to 1145 litres), preserved at the Coopers' Company in London, is
1.52 metres high, with a maximum diameter of 1.17 metres.
The Exeter cask was made of oak staves bound together by hoops made of
split roundwood rods of coppiced hazel. The ends of each hoop were seized
by fine strands of split hazel or elder. Similar casks are depicted in figure 5.3.
Ila/el hoops and oak staves and ends from tour different sizes of cask have
been recovered from the early sixteenth-century Studland Bay wreck (Ladle,
1993: 16).
Empty tuns from the wine sold by the Gascons at Boston fair were a
hereditary perquisite (James, 1971: 73) but the Customs of the Sea refer to
merchants hiring casks for their trading voyages, with empty casks needing
to be returned within an agreed period (Twiss, 1874: 293). The shipowner
had to calculate whether it would be more profitable to carry the casks
whole and not have room for other cargo or to break them down into
staves, making room for more cargo, but incurring the expense of having
them rebuilt (Twiss, 1874: 531-3). The fifteenth-century Gdansk W5 wreck
92 Medieval ships and shipping
Figure 5.3 Casks of wine in hulk-like lighters at the port of Paris (photo:
Bibliotheque Nationale, Paris. Ms Fr. 2091-2, f.125).
contained more than 200 small planks for making barrel staves, manu-
factured in the Baltic region for export (Litwin, 1985: 46).
The ancient right of prise entitled the king of England to one tun of wine
from ships laden with between ten and 19 tuns; and two tuns (one from
before and one from abaft the mast) from ships carrying 20 or more tuns.
From 1302 foreign merchants were required to pay customs instead and this
was set at a rate whereby importers of more than 60 tuns would have been
better off paying prisage (James, 1971: 4). This reflects the increasing size of
merchant ships. In the mid-thirteenth century ships of 100 tuns were rare in
the Anglo-Gascon wine trade but by the early fourteenth century although
81 per cent were less than 150 tuns, 16 per cent were between 150 and 200
tuns, 3 per cent were over 200 tuns and the largest reached 300 tuns.
Medieval tunnage measurements of ships cannot be used to give more than
a rough guide to their dimensions. Tunnage literally meant the number of
tuns that a ship could carry.It did not refer to the weight of the cargo or the
displacement of the ship (Lane, 1964). The tun may have been a little larger
at the end of the period than at the beginning and, more importantly, the
Carrying trade 93
internal layout of a ship might affect its carrying capacity. It was not until
the sixteenth century that empiric tunnage was converted into the formula,
known as 'Baker's old rule', of (keel x beam x depth) 4- 100 = tons burden
(Friel, 1984: 135).
Casks were principally used for carrying liquids or wet products, including
herrings, but their strength and rigidity also made them useful for other
commodities. The cargo of a ship pirated on a voyage from Southampton to
Seaton in 1446 included a pipe (a cask half the size of a tun) of white soap
(Piatt, 1973: 162). More than 100 casks have been excavated from the
Gdansk W5 shipwreck, containing pitch, birch tar, wood ash (used to
dissolve dyes), resin and wax (Smolarek, 1979: 62). Casks on board ships
were laid on their sides with their ends facing towards the ends of the ship.
They were secured with wooden supports to ensure that they did not break
loose. So far the only medieval archaeological site where these supports have
been identified is the Gdansk W5 wreck (see below).
weighed down by the loose grey salt piled high above their gunwales' can be
dismissed as fanciful. Good stowage and ship-handling practice required
nothing to be carried on deck except the ship's equipment and the victuals
(Twiss, 1874: 245-7). During the fifteenth century ships in the Hanse Bay
Fleet could be as large as 800 tons.
In 1351 a large quantity of grain was shipped from Yorkshire to London.
Some of it was ground into flour at Hull and packed into 30 casks which
were carried to Grimsby in two boats, then loaded into one of the four ships
hired for the voyage to London. Laths were bought for dunnage as well as
'mattes large and small', planking, and litter for stowage. A total of 950
quarters of grain were measured into sacks and carried to the ships
(Salzman, 1931: 216). Sacks were also used for other foodstuffs such as
onions and beans.
Wool was sold by the sack, each weighing 26 stones (164 kg). The weight
of the standard package was therefore less than for grain or salt but, because
of the low density of wool, it may have taken up more space. A 'poke' held
the equivalent of half a sack and a larger package called a sarplar held the
same as 2Vi sacks (Carus-Wilson and Coleman, 1963: 22). The 1447-48
customs accounts list both wool and cloth by the sack (Power, 1941: 37) but
the 1414 list of lighterage charges at Southampton (Piatt, 1973: 83) show
that rolls of cloth were transported by the bale. A bale was a package
wrapped in a covering, perhaps of canvas, and trussed with bands or cord
(figure 5.5). A bale was a half-load for a mule or pack-horse and in Italy, the
source of much of the Southampton imports, a bale weighed 80-85 kg
(Mallett, 1967: 178). The 1414 Southampton schedule of lighterage charges
show that alum (a mordant for fixing dye), woad and other dyestuffs were
shipped in bales.
Carrying trade 95
Figure 5.5 Cargo handling at the quays (French, early sixteenth century. Photo:
Bibliotheque Nationale, Paris. Ms Fr 2810, f. 86v).
Furs and hides were also transported in bales, as were spices, including
pepper, anise, ginger and cumin. In the late fifteenth century a London
merchant bought 12 bales of pepper for a prodigious total price of more
than £200. He was understandably unhappy when one of the bales became
24 pounds lighter as it dried out and he found that the pepper it contained
was musty (Salzman, 1931: 180).
Fruit, brought to England in ships from Italy and Iberia, would have
required careful packing. On the Venetian galleys currants and molasses
were not to be stored in the hold but should be between decks instead
(Salzman, 1931: 231). Oranges were even more susceptible to spoiling
because of the motion of the ship. Fig seeds have been excavated from the
wreck of the Iberian ship which sank at Studland Bay, Dorset, in the early
sixteenth century (Hutchinson, 1991: 174). They were associated with scraps
of loosely woven material which may have been fragments of frails, which
were rush baskets for holding figs, raisins and other dried fruit.
Coal mined in the vicinity of Newcastle was packed loose into barges
called keels. From early times each keel held 20 chalders, the standard
measure of coal weighing between 18 and 20 hundredweight or slightly less
than a ton. Because export duty on coal was set at 2d per keel-load, it
became the practice to build keels of 22- or 23-chalder burden. This was
forbidden in 1385 and by Act of Parliament of 1421 the actual capacity in
96 Medieval ships and shipping
chalders of each keel had to be marked upon it. After that the size of the
chalder increased rapidly (Salzman, 1923: 18). The keels took the coal to a
point below Newcastle bridge where it was re-loaded into seagoing vessels
(Pelham, 1948a: 258). The average capacity of the vessels which carried coal
from Newcastle to the Low Countries in 1377-78 was a little less than 50
chalder but the shipping activity must have been great because 7,338 chalder
of coal were exported in that year (Salzman, 1923: 19).
Metals were transported in blocks and bars, stowed low in the hold.
Comparison of the merchants' cargo list and the English customs accounts
for the Florentine galleys which left Southampton in 1444 show that 62
pieces of lead went unnoticed by the officials, probably because they were
among the ballast (Mallett, 1967: 139-40). The size of the pieces of lead and
tin is not recorded. The port books show that worked tin and lead was
carried in baskets, crates and barrels.
Stone was carried in blocks. The cost of stone lay mainly in transport,
with the carriage cost perhaps twice the price of the stone at the quarry, but
it was no more expensive to import stone from Caen to London than to
carry it coastwise from Beer in Devon; and the traffic was considerable. The
Tower of London alone used 75 shiploads of Caen stone in 1278 (Salzman,
1952: 135-6).
Pottery and glassware, such as the fine tablewares brought in Venetian
galleys,were at the other end of the scale of fragility. It is not clear how they
were packaged for transport by sea. It does seem that pottery was generally
only a small part of a cargo and was perhaps a commodity which could be
packed in among other goods. Large quantities of Saintonge pottery was
included with cargoes of wine from south-west France. The best evidence for
trade by sea of the products of a British medieval pottery centre is provided
by the distribution of finds of thirteenth- and fourteenth-century Scarborough
ware. Some of these high-quality glazed and decorated vessels are found
inland but many more were taken by sea to Newcastle, Perth and Aberdeen
as well as to Norway and the Northern Isles (McCarthy and Brooks, 1988:
94-6). Wadmal, a coarse cloth used as packing material around fragile
goods, was produced in both Norway and Iceland from double-coated sheep.
A piece excavated at the waterfront at King's Lynn has been
of cloth
identified as wadmal (Clarke, 1984: 136) and other examples have been
found in London and Newcastle (Walton, 1988: 81-2).
Single commodity cargoes were probably quite rare. The ships which
loaded with salt at the Bay of Bourgneuf might also have taken on wine,
canvas and linen, fruit, sugar and oil (Bridbury, 1955: 127). Alum and
dyestuffs were generally carried as part of mixed cargoes, which might
include wine and salt from France or, for example, green ginger and cotton
from the Mediterranean.
The Nicholas was a ship belonging to a London merchant which was
attacked by some Flemish pirates while anchored near Sandwich in 1318.
The pirates killed most of the crew, stole the ship's contents and sank the
ship. The three merchants whose goods were on board submitted the
following valuation of the ship and cargo (Cal. Close Rolls, 1313-1318:
its
When the goods of more than one merchant were carried same ship
in the
they could not be stowed separately as the heaviest goods of each of them
had to be stowed low in the hold. Merchant's marks had to be put on the
packages before loading so that they could be identified. Such marks have
been found on casks from the Gdansk W5 wreck (Litwin, 1985: 47). The
Customs of the Sea acknowledged the dangers posed by vermin, ruling that:
'If goods shall be damaged by rats on board a ship, and there be no cat in
the ship, the managing owner of the ship is bound to make compensation',
unless the cat died during the voyage (Twiss, 1874: 99). The master of a
cocha bringing a bale of wool from Southampton to Genoa in 1384 was
obliged to pay compensation for its being damaged by water and rats. He
denied the charge and said in his defence that 'in eius nave habeat musipulos
prout debebaf (in his ship he had mousetraps, just as he ought to have)
(Liagre de Sturler, 1969: 610).
Ballast, additional to that stowed permanently to provide stabilityand
trim, was taken onto ships to compensate for the offloading of heavy cargo.
Ballast mounds consisting of stone from all over Europe must have formed
on the foreshores of British ports. Especially during the fifteenth century
Kingston-upon-Hull imported large quantities of stockfish, a light
commodity, and Hull's streets are have been paved with the
said to
Icelandic stone brought in with it (Buckland and Sadler, 1990: 118).
Metamorphic and igneous rocks of Scandinavian origin have been recognised
in medieval structures in King's Lynn (Clarke and Carter, 1977: 440) and no
doubt reached the port as ships' ballast.
Voyages
Ships on trading voyages faced dangers from hostile vessels and from foul
weather. A convoy system was instituted as a result of the war between
England and France from 1324, in order to supply Bordeaux with victuals
and bring wine back to England (James, 1971: 126). The convoys
concentrated the trading activity to two sailings per year, one in late autumn
for the vintage wines and another in early spring for the more mature reek
98 Medieval ships and shipping
wines which were strained off in January and February. Protection at sea
involved a double or even triple complement of men, including archers,
aboard each ship (James, 1971: 71). The fleets under convoy were often
large even during periods of active warfare: between May 1339 and July
1340 some 235 English-owned ships brought cargoes back to England
(James, 1971: 129). Ships did still try to sail alone before the rest of the fleet,
to gain commercial advantage by being the first to bring home the vintage
wine, until Richard II enforced the convoy system in the 1370s (James, 1971:
75).
Commercial pressure meant that trading voyages were undertaken all year
round, not just in summer months. Winter sailing was of course
the
especially hazardous. Off the coasts of north-west Europe today gales of
force 7 and above are eight times more frequent in winter than in summer
and rough seas may be expected on average every fourth day in winter,
compared with every twelfth day in summer (McGrail, 1987: 260). As well
as the risks to the safety of a ship caught in storms, rough weather causes
increased wear and tear on a ship and its gear and the crew operate less
efficiently when numb with cold. There are many documented winter
voyages, some of which ended in shipwreck (Piatt, 1973: 126; Rose, 1982:
35, 48, 51). In the late fourteenth century it was important for English
merchants to import wine from Bordeaux as early in the season as possible.
In 1384 the wine fleet left Southampton in September and returned there in
the following January and February (Piatt, 1973: 70, 74, 129). Despite
frequent official prohibitions of sailing between November and April, as for
example by the Hanse governing council in 1403 (Scammell, 1981: 54), the
practice continued. In the fifteenth century, London received more than a
quarter of its annual supply of salt between November and the beginning of
March (Bridbury, 1955: 112). The increase in the size of ships does not seem
to have relieved the problem; instead it made each loss a greater disaster. As
late as 1532 legislation was enacted against winter sailing for Bordeaux wine
because of the mounting toll of losses (Waters, 1967: 11).
James (1971: 119-24) has studied the routes that the wine ships took
between England and Gascony. The voyages were made hazardous by the
rocky coasts and prevailing south-westerly onshore winds. Ships had to be
prepared to wait a long time for favourable winds, so voyage lengths
varied considerably. The Solent was often the first stage in voyages from
the east coast and from London. Ships would assemble there to make the
passage together. From the Isle of Wight they would either cross to
Brittany then round the French coast; or follow the English coast along to
Cornwall and then head to Brittany. Mathieu was frequently a port of
St
call for victualling. The Breton coast was an important half-way house in
the transit of the providing havens against storms but also
wine,
harbouring pirates who
could be equally dangerous. The purser's accounts
survive for the voyage of the Margret Cely to Bordeaux in 1486-88. In
1486 the ship embarked meat and bread at the start of the voyage at
London, took on more bread, salt, salted and fresh fish and beer at
Plymouth and further bread supplies at La Rochelle, at He de Rhe and at
Blaye. At Bordeaux they replenished supplies of mutton, beef, bread,
beverage and ship's stores.
Carrying trade 99
The Channel Islands remained part of the Angevin Empire after the loss of
Normandy to theFrench crown and so St Peter Port in particular became an
important asylum for English and Bayonnese vessels (Williams, 1948: 271).
The considerable quantity of Saintonge pottery revealed by erosion on the
large medieval wreck at St Peter Port has enabled the loss of the ship to be
dated to 1290-1310 (R. Burns, pers. comm.). From the thirteenth century,
Saintonge in south-west France was one of the primary production centres
for supplying luxury pottery to Britain and northern Europe. This was linked
with the English acquisition of Gascony and the development of the wine
trade (Barton, 1963; Hurst et ai, 1986: 76). The Guernsey ship was
presumably outward bound from Gascony. Future excavation of the site may
reveal more about how pottery was packed for transport by sea and about
the rest of the cargo, which can be expected to have consisted mainly of
casks of wine.
The Aber Wrac'h ship, which was wrecked off Brittany in the early
was carrying at least 25 to 30 tonnes of ballast. Analysis of
fifteenth century,
this has shown that it was of such a mixture that it cannot be said to have
been loaded in any one region; rather, demonstrates the diversity of the
it
Veyrat, 1989: 293). Among this, 1200 fragments of animal bones were
recovered, from sheep, goats, deer, cattle, fowl and a small quantity of pig. It
appears that some animals, particularly sheep, were carried whole and
perhaps alive. This meat may all have been intended to be eaten by the ship's
company but some may have been cargo. Gascony needed victuals because it
grew vines to the neglect of anything else and meat, fish and grain needed to
be imported. Analysis of the sediments overlying the ballast led to the
identification of human faeces including remains of walnuts, hazelnuts,
chestnuts, grape and apple pips, plum stones and rye seeds. This suggests an
exceptionally and nutritious diet, with ingredients from southern
varied
Europe. According to the Catalan Customs of the Sea, mariners were to have
meat on three days a week, porridge on the others and bread with cheese,
onions or fish every evening. With this they were to have either wine or
prunes or figs (Twiss, 1874: 211-13). Merchants, who voyaged with their
goods, took along their own victuals as well as a servant, bed and chest
(Twiss, 1874: 107).
Two coins from the Elblag region placed in the mast step of the cog
wrecked on the Danish coast at Vejby indicate where it was built. The
presence of a cooking pot of Baltic origin suggests that the ship was fitted
out there. The finds on the wreck site support the interpretation that the ship
was on its homeward voyage at the time of loss in about 1380 (Crumlin-
Pedersen, 1985: 72-3). It appears that it failed to make the entrance to the
narrow Oresund between Denmark and Sweden and foundered in shallow
water at the foot of high clay cliffs. A part of the anchor cable, housed in a
leather sleeve, lay beneath the stern, showing that the main anchor had been
dropped in an attempt to save the vessel. Eighteen tons of ballast stones,
originating from somewhere along the Atlantic seaboard, were excavated
from the site. A concentration of metal finds lay around the stern, including
coins, pewter plates and tripod pitchers (figure 5.6). One hundred and nine
English gold coins of Edward III were recovered, ranging in date between
100 Medieval ships and shipping
102 Medieval ships and shipping
Figure 5.8 Bundles of iron bars from the Gdansk W5 wreck (photo: L. Nowicz).
1351 and 1377, and minted in London and Calais. There was also a gold
coin from Lubeck and a small Flemish 'mite'. The first ever English gold
coins had been minted less than half a century before and England was at
pains to attract and retain bullion. The export of coin and gold and silver
was constantly forbidden and merchants were required to export
merchandise instead. It was not until 1400 that merchants importing
goods into England were allowed to take half of the proceeds in coin,
provided that they spent the other half on English goods and obtained a
licence from the king (Salzman, 1931: 19-20). The Vejby wreck is a witness
to the difficulty of enforcing such regulations.
The Gdansk W5 ship appears to have sunk soon after leaving port, loaded
with a cargo of forestry and mining products. The wreck lay on the sea-bed
tipped onto its starboard side (figure 5.7). After the upper layers of cargo
were removed, a well defined and deliberate stowage plan revealed itself
(Litwin, 1985: 46-7). Oak planks were laid on the bottom of the hold.
Casks of iron ore and bundles of iron bars rested on them, with all the free
spaces between the casks packed with wedges to prevent movement of the
cargo. The very heavy iron bars, fastened into bundles with cord and copper
alloy bands (figure 5.8) were stowed towards the stern of the ship. The 52
bundles recovered contained an average of 80 bars each and there were also
some others found loose. The bars are about 80cm long with a rectangular
cross-section, tapering towards one end. Copper, transported in the form of
large disc-shaped ingots (figure 5.9), was stowed in the midships area. There
were 213 ingots raised. Above this were laid casks containing pitch, birch
tar, wood ash, resin and wax. The tar was transported in 69 and 99 litre
Carrying trade 103
Figure 5.9 One of the copper ingots retrieved from the Gdansk W5 wreck (photo: L.
Nowicz).
casks and the smaller ones were cylindrical rather than the typical 'barrel'
shape.
The and the shore-based
activities facilities involved in loading and
unloading cargoes were substantial and to these let us now turn.
6 Ports
Virtually all the towns of medieval Britain were served by water transport
along rivers but the term 'port' is used here to distinguish those settlements
whose economy depended on the operation of ships. At the beginning of the
period few of the major ports were sited on the coast. They tended to be as
far up estuaries or rivers as they could be while
allowing sufficient deep
still
increasingly concentrated in London and the trade of the North Sea was
progressively dominated by Hanse merchants. Plague, particularly
devastating in towns, was a set-back to urbanism all over Europe from the
mid-fourteenth century. Yet ports seem to have shown great resilience in
their recovery from other disasters, natural and man-made, such as the gales
of 1377 which washed away the harbour at Lyme in Dorset (Williams, 1948:
279) and the French raids on south coast ports in the fourteenth and early
fifteenth centuries.
Waterfronts
(Miller, 1977; Hobley, 1981: 3-7). Documentary evidence suggests that New
Fresh Wharf was the busiest sector of the medieval waterfront, already well
developed by the twelfth century.
On the foreshore at the Seal House site, just above London Bridge, the
firstwaterfront revetment was built soon after 1140, followed by a second
after about 1170 and a third after 1210. A series of waterfront buildings was
associated with the third structure, including a house with two ground-floor
rooms and a large shed with a drain emptying over the quayside. A ragstone
wall was built on the riverward face of the third revetment (Hobley and
Schofield, 1977: 37-9). Excavations on several other London waterfront
sites, notably Trig Lane (Milne and Milne, 1982) and the Customs House
106 Medieval ships and shipping
(Tatton-Brown, 1974 and 1975), have confirmed the general sequence. In the
twelfth and early thirteenth centuries the structures which were being built
along much of the waterfront incorporated a large proportion of reused
timber, including ships' timbers, while in the later thirteenth century the
timber was specially cut for the purpose. A complex array of carpentry
techniques was employed (Milne, 1982), bearing very little resemblance to
those used for shipbuilding. In the fourteenth century, stone riverfront walls
became common and docks were constructed. From the eleventh to the
sixteenth century reclamation advanced the bank of the Thames by between
50 and a 100 metres by a series of rubbish-backed revetments, with deposits
in excess of ten metres deep (Hobley, 1981: 7).
Waterfront structures were being built not only in London but in ports
throughout northern Europe from the twelfth century. In Britain they have
been revealed in excavations in Newcastle (O'Brien et ai, 1988; O'Brien,
1991); Hartlepool (Young, 1987; Daniels, 1991); Hull (Ayres, 1981); York
(Hall, 1991); Lincoln (Jones and Jones, 1981); King's Lynn (Clarke and
Carter, 1977; Clarke, 1979; 1981); Kingston (Potter, 1991); Portsmouth
(Moorhouse, 1971: 204; Fox, 1981); Poole (Horsey, 1991; Watkins, 1994);
Plymouth (Barber, 1971; Barber and Gaskell-Brown, 1981); and Bristol
(Ponsford, 1981; 1985; Jones, 1991). The excavation at Canynges House at
Redcliffe, Bristol, shows the rapid and extensive reclamation which took
T
Wlh C f NTURY
1 !lh CENTURY
4 -1—
14(h CENTURY
P=3
? \
T —
rri — >
r
_1l 1r
15lh CENTURY
place at a waterside property from the twelfth to the fifteenth century (figure
6.2). There is and the complexity of the
a very large tidal range at Bristol
waterfront development was increased because quays were constructed which
could be used at different states of the tide.
Timbers from boats and ships have been found reused in waterfront
structures in several ports, although no assemblage from a single site in
Britain is from Dublin (McGrail, 1993a)
as large or as informative as those
and Bergen (Christensen, 1985) which have been described in Chapter 4. In
London, a late thirteenth- or early fourteenth-century revetment at the
Custom House site included slabs of still-fastened boat planking. The bottom
of the boat, which was relatively flat, was used in large sections while the
curving sides had to be broken into smaller pieces to serve their new purpose
(Marsden, 1979: 86-7). At Kingston the first waterfront development,
provisionally dated to 1300, included a 11.5 metre long revetment largely
made up of reused boat planking (figure 6.3) Two other large sections of
boat planking were found at the Kingston Horsefair site, one of them as
Ports 109
110 Medieval ships and shipping
vessels whose components are found reused in waterfronts. Even when the
remains are not broken into very small pieces, as they were at Bristol, the
preserved elements of the vessels are those which were deliberately selected
for their lack of features. The preferred material was flat planking of fairly
constant width. Framing is rarely found, though part of the keel was
included with the planking reused in the thirteenth-century revetment at
Fenning's Wharf, adjacent to London Bridge. Indications of the size of a
vessel provided by the thickness and width of the planks and the
are
dimensions of the fastenings. The spacing of the treenail holes along the
strakes may show how robustly the vessel was built. (For the analysis of
fragmentary planking see Hutchinson, 1984 and the attribute list for clinker
planking published by McGrail, 1993a: 169-71.)
Docks
The earliest dock yet located is that constructed in about 1213 in Southgate
Area A in Hartlepool. It was more than 20 metres long and had stone side
walls. One of these was lined with timbering, including the boat planking
referred to above, as a protective fender between the docked vessels and the
stonework (Young, 1987; Daniels, 1991).
In London, excavation beside Baynard's Castle revealed a dock basin of
the fourteenth and fifteenth centuries (figure 6.4). The dock was a
rectangular basin surrounded by stone walls, with the castle on its east side
and a large gravelled area for unloading on the west side. Oak rubbing posts
lined the dock to protect the boats from being damaged by the rough
stonework. On the west side these posts had been renewed several times
before the dock was filled in at the end of the fifteenth century. An oak
mooring post was found just outside the dock entrance, while towards the
dock head were two posts which perhaps once supported an unloading
platform that had projected over the basin. Fronting the unloading area on
the inshore side was a stone building with an arcade along the ground floor
(Marsden, 1981: 14-16). Another dock, built at about the same time as that
at Baynard's Castle, has also been excavated in London at the Sun and
Ports 111
Topping's Wharf site on the south bank of the Thames. There is docu-
mentary evidence that this dock was in existence by 1323. It formed the
entry to a gated landing place known as the Watergate (Sheldon, 1974).
Fourteenth-century docks have also been located at Bristol Bridge and
Canynges House (Jones, 1991: 23).
The dock found in excavations at Oyster Street, Portsmouth, was a gully
or inlet with banks supported by timber and wattle fencing. A wooden quay
extended along the shingle shoreline from either side of the dock, which fell
out of use in the fifteenth century (Hurst, 1970: 176-7; Moorhouse, 1971:
204).
A financial account is preserved for the building of a dry dock for the
King's shops at Portsmouth in 1496 (Hattendorf et al., 1993: 116-18).
Timber, iron, stone and clay were used. The dock was lined with timber and
had inner and outer gates at its head.
The construction of quays allowed ships to moor for loading and unloading
cargo. Large ships anchored off and smaller vessels, called lighters, made the
112 Medieval ships and shipping
intermediate journeys to the quays. The royal butler's accounts of the reigns
of Edward I to Edward III and fourteenth centuries record
in the thirteenth
charges for unloading wine casks at quays and by means of boats from ships
anchored off, as at the Pool of London (James, 1971: 139). Lighters were
used for unloading carracks, moored to piles out in Southampton Water
(Cobb, 1961: xxxiv).
The Laws of Oleron, the earliest surviving maritime code, was probably
compiled in the twelfth century and gained currency throughout Atlantic and
northern Europe. Much of it is concerned with the responsibilities of
shipmasters and one section says that when a ship arrives in port, the master
must show the merchants the ropes that he will use to hoist the cargo. If he
does not, he has to pay full compensation to the merchants for any losses
caused by the ropes breaking (Twiss, 1874: 16-17).
Cranes may have been introduced to waterfronts from the second half of
the twelfth century. For the rebuilding of Canterbury Cathedral in 1175,
special cranes were invented for discharging the boatloads of Caen stone
(Salzman, 1923: 104-5). Ellmers (1989) has traced the origins of harbour
cranes to three different models, each of which was previously used for
other purposes. The first type, the hoisting spar, or 'wippe', was a mast
with a yard which could lift goods and swing them round (figure 6.5). The
Bergen town law of 1250 refers to cranes of this type (Ellmers, 1989: 48).
The second type, the windlass, had been used for raising goods to the
upper floors of merchants' houses. This type of windlass was operated not
by turning the drum by means of handspikes but remotely, by turning a
large wheel linked to the windlass drum by a chain or rope loop. The
windlass could be mounted on a support with the wheel below, with a
turning jib. The third type was the crane with a treadwheel. Cranes like
these had been used for building churches and were adopted in
Netherlands harbours in the first half of the thirteenth century or perhaps
before.
In London were cranes on public and private wharves in the
there
fourteenth and fifteenth centuries (Schofield, 1981: 31). At Southampton
Quay there was a crane from the early fifteenth century (Cobb, 1961: xxxvi).
Very occasionally at Southampton, tuns of wine were recorded as being
exempt from cranage charges because they were 'raised by hand'. Unlike
tuns of wine, bales and sacks were charged wharfage rather than cranage,
indicating that these were unloaded manually or with the ship's own lifting
gear. The crane garth at York, the place where all waterborne cargoes
owned by non-citizens had to be landed, weighed, checked and charged toll,
is first mentioned in a document of 1417.
The Southampton steward's accounts for the end of the fifteenth century
record that the town hired out equipment called 'styves' (from the Italian
stivare, to stow) for unloading the Italian galleys at a rate of £5 per galley
with extra charges for breakages. The accounts for the making of new styves
in 1493 show that elm, oak, ash and elder timber was used and that the
Figure 6.5 Reconstruction of a hoisting spar used for cargo handling in Bergen
(Universitetet Bergen Historisk Museum).
i
In the larger ports, separate areas of the waterfront were used for handling
different commodities. In London, for example, Queenhithe was the place
where grain was landed. In 1300 there were eight master measurers of grain
there, each with three assistants. They measured the grain into cylindrical
vessels called bushels, levelled at the brim. Eight of these were equivalent to
one quarter of grain (Salzman, 1931: 56). Individual ports had their own
measures and weighing equipment. In an attempt to impose uniformity in
weights and measures, weighing beams were distributed from the Exchequer.
In 1352, £40 was spent on making 12 new beams for ports (Salzman, 1931:
59). The name given to the Hanseatic headquarters, the 'Steelyard', refers to
the weighing equipment which was to ensure fair The Custom
dealing.
House quay at London was wool trade perhaps
the control point for the
from the time that the Great Custom was imposed on wool in 1275. In the
late thirteenth or early fourteenth century a large braced structure at least 2.2
114 Medieval ships and shipping
metres high was built on the foreshore, incorporating the hull planking
mentioned above. There was also a large timber jetty running north-south,
most probably the main jetty for the Wool Quay (Tatton-Brown, 1974).
Specially built granaries and warehouses would have been needed at the
accredited hythes for goods awaiting distribution or transhipment.
Warehouse ranges with open arcading have been located at Billingsgate
and the Custom House. Merchants also had private warehouses. In London
the wine trade was concentrated at the Vintry, where there were large houses
with vaults for wine storage. They had access to the wharf on one side and a
shop frontage on the other. There is documentary evidence for at least ten
vaults there in 1376 (Schofield, 1981: 25). Several medieval stores survive at
Southampton, including the late twelfth-century vaulted wine cellar which
still stands on the former Castle Quay (figure 6.6). The vault is 5.7 metres
wide, 17.05 metres long and 4.0 metres high and now lacks all of its ribs
and most of its corbels. At King's Lynn, in the thirteenth and fourteenth
centuries, documents and standing buildings suggest that merchants' houses
were often built as 'divided properties' with their living accommodation on
the landward side of the street while their warehouses and quays were
between the street and the river. The fifteenth-century warehouse at
Ports 115
Hampton Court in King's Lynn was built parallel to the river with an open
arcade of eight bays as its west wall, so opening directly onto the river
(Clarke, 1979: 162-3).
Foreign merchants often stayed several months in port and rented cellars
and houses. At Southampton the oarsmen of the galleys had their own
quarters and a memorial stone from their burial ground has been found at
North Stoneham (Williams, 1948: 277). Hanse merchants had establishments
in Britain at London, King's Lynn, Boston and Hull. In London their
Steelyard complex, on the site of the present Cannon Street station, included
a council chamber, a wine tavern, houses for merchants and rows of
warehouses running to a quay on which there was a crane. At the south-west
corner of the site stood the Hanse Master's house, which is known from
engravings to have resembled that which still stands at King's Lynn
(Schofield, 1981: 30). The King's Lynn Steelyard, a fifteenth-century brick
and timber structure, is the only surviving building of the Hanse in Britain
(Clarke, 1979: 163).
Functional zoning
t I I — — —1=1 —
<
<
f
0 M*tr«a lOO
Figure 6.7 Southampton in the later Middle Ages (from Piatt and Coleman-Smith,
1975).
Although journeys by boat could often be very slow, rivers were vital for
taking goods for export from their place of production to a seaport. Having
an adequate transport route to the coast must have been a crucial factor in
the viability of many production sites. Exports travelled downstream to the
ports so that the vessels were heavily laden when travelling with the flow of
the water and were much lighter on the return journey against the stream.
The distribution of imports from the ports where they were landed was not
reciprocal to the exports, since a variety of commodities was coming in, with
different and more widespread destinations. Their transport was not
necessarily so closely linked to waterways.
The royal butler's accounts for the first half of the fourteenth century
show that weather had an important influence on the route that consign-
ments of wine took from the quayside to their destination. Whenever the
state of the sea permitted, wine was shipped coastwise until it could be
transferred to river transport. In the first half of the fourteenth century it cost
as much to transport a tun of wine by road for 40 to 54 miles as it did to
bring it by sea from Bordeaux. A cart carrying a tun of wine might require
as many as six horses at a time and as a rule it was only the last stage of the
journey which was completed by road (James, 1971: 147, 149).
The balance between the use of road and water transport varied in
different parts of the country according to the nature of the terrain and the
waterways. In the Fenland of eastern England, transport was predominantly
118 Medieval ships and shipping
Figure 7.1 Map of Britain by Matthew Paris, drawn about 1250 (Cotton Claud.
D.V1, f. 12v. By permission of the British Library).
Figure 7.2 St Guthlac travelling by boat in the Fens, twelfth century (Harl Roll Y6
Roundel 4. By permission of the British Library).
were brought by packhorse, not down the Tamar. In the Severn area,
transport by water was much better than by road. The Severn was used for
exporting coal, in 'keels', and iron manufactures through Gloucester from
the Forest of Dean.
From the thirteenth century the Tyne valley shipped coal to London, to
other east coast ports and also abroad. Osier has detected a medieval
survival in the Tyne keel (figure 7.3), a square-rigged clinker-built vessel,
used until recent times for carrying coal along the river (Osier and Barrow,
1993: 24). He suggests that official regulation of the size of these vessels to
prevent the evasion of export duty (see Chapter 5) was one of the factors
which inhibited change.
Building stone was a commodity which relied heavily on water transport.
The great numbers and sizes of cathedrals, churches, religious houses and
castles demonstrate that the traffic in stone during the Middle Ages must
Inland water transport 121
Figure 7.4 Blocks of Barnacle stone lost in Wittlesey Mere (photo: N. Mitchell).
have been huge. Some of this was transported solely on inland waterways,
never having to reach the coast. Building accounts which refer to the carriage
of stone by water (Darby, 1940: 105) provide little information about the
boats which were used. The cost of stone lay mainly in transport. The York
Minster Fabric Rolls reveal that stone was brought by rivers from named
quarries, then carried by sleds into the mason's yard (Salzman, 1923: 85).
Barnack quarry, between the Welland and the Nene, was a most important
source of stone for the buildings in the Fenland and East Anglia. The monks
of Sawtry Abbey made a canal, or 'lode', for carrying Barnack stone to their
house by way of Wittlesey Mere but in 1192 the Abbot of Ramsey ordered
the blocking of Sawtrys lodes where they crossed Ramsey Abbey land. In
order to regain access, Sawtry had to promise to put up no buildings except
one rest house for the men on their stone barges. When Wittlesey Mere was
drained in the nineteenth century four large blocks of Barnack stone were
revealed (figure 7.4). They were presumably lost in a medieval barge
accident. Each weighs approximately one ton, with the largest measuring
94cm x 73.5cm x 73.5cm (Jenkins, 1993: 260).
In inland fisheries boats were used for operating nets, setting traps and
managing fish The Domesday survey records fishing activity in the
weirs.
Mere the abbot of Ramsey had one boat, the abbot of
Fens. In Whittlesey
Peterborough one boat and the abbot of Thorney two boats (Bond, 1988:
80). Boats were not only used on large expanses of water but were also
employed on fish ponds. In 1256-59 the king ordered the bailiff of
Woodstock to cause 'oak to be chopped down in the park of the king at
Woodstock and to cause a certain boat to be made to be placed in the
122 Medieval ships and shipping
fishponds of the king'. At the same time a boat was provided by the warden
of the houses of the king and of the fishpond at Havering, for the king's
fishermen when he sent them down to fish (Steane, 1988: 48).
By the early thirteenth century, fish weirs had in some places become an
obstruction to navigation and this problem was especially serious on larger
rivers like the Severn and Thames. An Act first passed in 1346-47 for
'remedying annoyances in the four great Rivers of England, Thames, Severn,
Ouse and Trent', and orders that 'all wears, mills, stanks, stakes, and kiddles
which disturbed the passage of ships and boats in great rivers should be
utterly pulled down', had little effect and needed to be repeated nine times
up to 1495 (Bond, 1988: 86-7).
A ferry across the River Tamar is recorded in a 1337 survey of the Duchy
of Cornwall. 'Three burgesses of Esshe [Saltash] hold. . . the passage of
Esshe with rent pertaining to the same passage with a boat and four oars for
the said passage'. It must have been a sizeable boat, for we learn that 'John
de Ferar shall cross there and his whole household with all carts' (Hull,
1971: 119, 122-3).
Boat finds
Numerous small boats have been found in inland contexts in Britain. Many
of them are logboats. There has long been a popular assumption that
because simple logboats are products of very basic technology they must be
of an early date. McGrail (1978a) has compiled a corpus of the logboats of
England and Wales, obtaining samples for radiocarbon dating where
possible. Results obtained have shown that many logboats previously
thought be be prehistoric are in fact medieval.
Nine logboats found within a 20 kilometre stretch of the River Mersey
have produced radiocarbon dates ranging from the ninth to the twelfth
centuries. Their period of use probably falls within the eleventh century
(McGrail and Switsur, 1979: 93-115). They were made from short oaks,
given rounded ends with protrusions resembling stem and stern posts.
Horizontal holes for mooring ropes were bored through their forward ends.
An early eleventh-century date seems likely for the large logboat, more than
five metres long, found at Stanley Ferry, Yorkshire, in 1838 (McGrail, 1981:
164).
An extended logboat was found in the drained bed of the former Kentmere
Lake, near Kendal inCumbria, in 1955. This logboat has been radiocarbon
dated to 650 ± 120 bp (D-71) or about 1300 ad (Wilson, 1966). It is a
hybrid between a logboat and a plank boat. The base is a shallow oak dug-
out hull, 4.25 metres long and 0.61 metres wide, which is flat bottomed with
a rounded stern and slightly pointed prow. To this are attached, on either
side, five wash strakes. The strakes were each made of two planks, joined by
a short (10cm) scarf in the centre. No logboat with so many side strakes is
known from a continental European context. Four ribs made from grown
oak crooks were treenailed to the hull and to the top strake. The strakes
were fastened together clinker-fashion with round-headed nails which did not
have roves but which had their ends hammered over. A rowlock was found
Inland water transport 123
attached to the top strake on the better preserved side of the boat, suggesting
that there was a central thwart for the oarsman, and an ash plank found
near the boat might have been a thwart. At the junction between the top of
the hollowed log, which formed the bottom of the boat, and the lower edge
of the planking there was a strip of wood treenailed onto the outside of each
side of the hull, running along about three-quarters of the boat's length.
These strips would have enhanced the boat's stability, making it less likely
that it would capsize. Like the River Mersey logboats mentioned above, the
prow of the Kentmere boat had a projecting cutwater through which there
was a hole, presumably for a mooring rope.
Only about 40 km from the find-spot of the Kentmere boat, the
Giggleswick Tarn logboat (figure 7.5) was similarly found in the bed of a
drained lake, in 1863. It has been radiocarbon dated to 615 bp ± 40 (Q-
1245) or about 1335 ad (McGrail, 1978b: 29). It was made of ash and its
overall length is 2.45 metres with a maximum beam of 0.63 metres. It is
interesting because of the strengthening timbers fastened across the ends and
because of external longitudinal timbers at the level of the sheer. These
would have provided strength and would have assisted stability when the
boat was heavily laden.
A logboat found on the shore of Oak Mere, Cheshire, (figure 7.6) has
been radiocarbon dated to 1395-1470 ad (McGrail, pers. comm.). It
demonstrates that logboats continued in use until a very late date and that
their builders had developed a thorough understanding of the raw material
and how to shape it to produce boats with good stability and performance
characteristics (McGrail, 1978a: 249). The logboat was 3.6 metres in length
with a maximum beam of 0.79 metres. It had a slight keel carved out of the
solid wood, with hollows on each side, which would have helped it to
maintain a straight course. Both ends were raised and pointed to provide a
cutwater and to prevent swamping. A maximum amount of wood had been
hollowed away, producing a light boat with thin sides. Extra thickness was
provided in two places and the excavator described these as 'ribs' cut out of
the solid (Newstead, 1935: 210); they would have provided extra strength at
the seat positions.
The number of logboats in the archaeological record is probably
disproportionately large because they are solid and readily identified when
they are found by chance. Fewer plank-built boats have been recorded,
probably because their remains can be slight and fragmentary. Several pieces
of oak boat planking were found during archaeological excavations in the
moat of Southchurch Hall at Southend in Essex. Although scattered, they
were in similar stratigraphical positions, and some of them were sealed under
bridge footings which were constructed in the early thirteenth century. The
excavated planking was very fragmentary and fragile. There are nine main
pieces, several consisting of more than one plank. All the finds share
sufficient common characteristics for there to be no doubt that they are all
from the same boat. By the direction of the scarfs and the presence of hood
ends- it can be inferred that the planking came from the starboard side of the
boat adjacent to the stern post. The edges of the planks exhibit strong
curvature for building a generous curve into the side of the boat, giving a
round stern.
124 Medieval ships and shipping
GiggleswickTarn - remains
E F
1 9ECTION OF C 'NOt AND Ot POE'T 2, 3. Pl<" AND ELE VATlON IllltL ItCTlONON OOTTE D LtNt)
4 Section amidships
Figure 7.6 The logboat from Oak Mere (from Newstead, 1935).
The Southchurch Hall boat was a small, lightly built, round-sterned vessel
which had had its useful life prolonged by major repairs. Although we know
it had a stern post, we do not know whether it had a keel or a flat-planked
bottom. The length and beam of the boat and the shape of the bow are also
unknown. It is possible that it had been used for pleasure, as shown in figure
7.7, beforeending its days in the moat.
Threave Castle, built on an island in the River Dee in Galloway, had a
shallow harbour cut out of the bedrock to allow boats to moor in a place
protected from the fast river current. Excavations within the harbour have
produced parts of an oar blade and a paddle blade, probably deposited in
the late fourteenth or the fifteenth century. The object illustrated as a boat's
rudder by Good and Tabraham (1981: 121) is more likely a window shutter
or a box lid.
Many castles, other than those where the elevation of the site was part of
the defensive design, were sited so that they could be reached by boat. The
stone to build them was
usually brought by water and provisions continued
to be supplied by
means. Familiar examples include those castles built
this
beside major tidal rivers, such as the Tower of London on the Thames and
Rochester on the Medway. Less obvious ones are Castle Acre in Norfolk
(Coad and Streeten, 1982: 138) on the River Nar, which runs through King's
126 Medieval ships and shipping
Oi fixnctcfKtafmccottmta
cotpm mam <mmw n cr
Figure 7.7 Pleasure boating in the fifteenth century (BL Add. Ms. 54782 f.54. By
permission of the British Library).
Figure 7.8 The boat found at Caldecotte, Milton Keynes (National Maritime
Museum).
were well thought out but the fastening was not very carefully done, perhaps
suggesting that boats of this type had been made for a long time before this
example.
The boat was lying towards the edge of an old watercourse, roughly
parallel to the bank. In the Middle Ages the surrounding area was
marshland. There was a mooring chain and spike fastened to the inside of
the stem and left dangling overboard and three pieces of hemp rope were
found under the boat.
The remains contained insufficient tree rings for dendrochronology. The
widest oak components, the planks, were through-sawn boards so that many
of the same tree rings occur twice in the same cross-section. A radiocarbon
date (HAR-5201) of 410 ± 60 years bp or about 1480-1600 ad was
obtained, indicating that the boat belongs to the medieval or early post-
medieval period.
The closest parallel for the Caldecotte boat, in shape and construction, is
the Somerset turf boat (McKee, 1983: 107-9), a type still being made in this
century. It has mooring chains similar to that on the Caldecotte boat at both
ends. The question of the survival of medieval boat types into recent times is
intriguing and the case of the Tyne keel has already been mentioned. The
appearance of some other vessels, for example the Fenland lighters (Jenkins,
1993), suggests an ancient origin but medieval examples have not yet been
excavated. Good pictorial evidence for vernacular craft can often be found
extending back to the seventeenth century but rarely much before. There is a
need for further research on pictorial sources to elucidate changes in local
128 Medieval ships and shipping
types in the post-medieval period. This would serve as a bridge between the
traditional craft which survived into the twentieth century, which have been
extensively documented, often just before they became extinct, and the small
craft in use in the Middle Ages.
8 Fishing
phases. Plaice also occurs in most samples, as does haddock and the conger
eel to a lesser extent (Wheeler and Jones, 1976: 212). Cod was the dominant
species recovered from the midden at Castle Acre Priory (Wilcox, 1980) and
also at Kirkstall Abbey (Ryder, 1969: 387). Excavations at the Austin Friars
in Leicester produced cod, ling, haddock and plaice or flounder bones but no
remains of fresh-water fish (Thawley, 1981: 174).
Fish can be caught with hooks, nets and traps and different gear is
appropriate for different species. Large demersal fish which are bottom
feeders, like cod and haddock, could be caught with hook and line. Smaller
pelagic fish which swim in shoals, herring especially, were more suitable for
catching in nets. Eels, flounders and small plaice could be caught in shallow
water with spears.
For fishing with hook and line at sea, hand lines and long lines were used.
No medieval fishing line has been found in Britain. It would need to be fine
and to have a considerable breaking strain: horse hair and bast would have
been suitable and also good quality hemp line. Medieval fish-hooks have
been found at numerous sites. They are mostly of iron though some are
bronze. They are not small; a hook must not be too small because the fish
may swallow it whole and bite through the line. In London the Trig Lane,
Billingsgate and Custom House sites have all produced medieval fish-hooks.
Their average length is about 50mm with some larger ones up to 75mm.
They are all barbed and, instead of having eyes like modern fish-hooks, their
ends are thickened or flattened so that the line bent round them would not
slip off. Forty-five fish-hooks from the period 1000-1200 were found on
excavations at Fuller's Hill, Great Yarmouth and three of these are shown in
figure 8.1. They are similar in shape to the London examples and vary in
length from 45mm to 75mm with one very large example 122mm long.
Fishing 131
Figure 8.1 Fish-hooks from Fuller's Hill, Great Yarmouth (after Rogerson, 1976).
Hooks of this size would have been effective for catching the largest fishes
represented by the remains from the site, such as spurdog, conger eel, ling,
cod, large haddock, turbot and halibut (Wheeler and Jones, 1976: 221).
Excavations at Hartlepool have produced two sea-fishing hooks in an early
thirteenth-century context, one 75mm long, the other, which has a barb,
50mm long. There are also three iron hooks from Pevensey (Dulley, 1967)
and one from Sewer Lane, Hull (Steane and Foreman, 1988: 146-8). A
pewter spoon from Beverley is engraved with an image of three fish about to
take hooks (figure 8.2).
Long lines with many hooks attached and weighted at the end, were towed
from fishing vessels, including the English doggers which worked Icelandic
waters from the fourteenth century. Hooks were fastened singly or in small
numbers on hand lines, between a weight and a float to keep the hook
submerged at the desired depth. The line might be wound on a hand frame,
such as those found in sailors' chests in the wreck of the Mary Rose. They
have wooden side pieces joined by dowels and have floats 20mm in
diameter, 60mm to 90mm long (Steane and Foreman, 1988: 149-53). On
the Thames foreshore large numbers of lead fishing-line weights, lost by
anglers from the waterfronts, have been recovered. They are of various
different shapes - pyramidal, conical or pear-shaped - and include a group
of more than 40 ascribed to the fifteenth century (Steane and Foreman,
1988: 155).
Evidence for nets and floats, netting tools and hemp twine have been
found in the Netherlands, Norway, Russia and Poland (Steane and Foreman,
132 Medieval ships and shipping
Fishing 133
1988: 162). The manufacture of nets with the weaver knot, which is still the
most widely used netmaking knot in Europe and America, has been carried
out in north-west Europe from very early times (von Brandt, 1984: 115-16).
There are scant remains of actual nets and contemporary illustrations do
little to throw light on the nature of the gear or its method of operation.
setting process then attaching the draglines to the ends of the mast laid along
the boat, so that the mouth of the net is kept as open as possible (figure 8.4).
A line of floats was attached to the head rope, along the top edge of the net,
and a line of sinkers to the foot rope along its bottom. The central part of
the net, for holding the fish while it was drawn in, might be in bag form.
The wings and the long draglines attached to them helped to drive the fish
into the centre.
A seine (or scoop) net like this could also be worked from a beach. This
was the method which was used in the Cornish pilchard fishery in the
sixteenth century. A boat took one end of the net out into the sea in a wide
arc then back to the beach. The nets worked most successfully when set a
long way out but with their feet weighted so they touched the bottom and
their heads buoyed at the surface. The prevalence of beach seine netting in
the Elizabethan period has led to the assumption that it was a new invention.
It seems to be the case that seine nets were in use long previously but that a
Sea traps
Nets were also used to make tidal traps. Carew, in the sixteenth century,
records that in Cornwall stakes were set in the mud at low water, across a
creek from shore to shore. A net was fastened to the foot of the stakes and
1
•
N03502OO 1
N0350IOO
N030000 f t
NO 349900
N034980ot
N0349TO0to to
(M a
O o
1
N0349600 ,uj
at high tide the upper part of the net was pulled up so that it formed a
curtain across the creek. When the tide ebbed the fish were left on the mud
(Halliday, 1953: 115).
There are numerous tidal fish-traps around the coast of north and west
Wales (Jones, 1983: 28-9). They are dry-stone barriers constructed in the
inter-tidal zone in areas of considerable tidal range, usually in excess of five
metres. At high tide they fill up with water and with fish, then as the tide
ebbs the water drains away and the fish are trapped. Most of the Welsh fish-
traps are semi-circular or semi-oval in plan with enclosures at the seaward
end to contain the catch. Jones (1983: 30) has identified four different types.
The fish-traps were self-baiting because marine worms colonised the silts
which accumulated in their seaward ends and these attracted small fish. The
small fish established habitats in the walls and were preyed upon by larger
fish.
The origins of the traps are lost in folklore but they were certainly still
being built in the medieval period and some were used by the Cistercians.
Aerial reconnaissance in 1989 revealed a semi-circular structure lying just off
the coast in Caernarfon Bay, North Wales. The anomaly has been
demonstrated to be a fish-trap, constructed at about the beginning of the
thirteenth century (figure 8.5). The enclosure wall has a central 'backbone'
between one and two metres wide with collapsed boulders spread on either
side. The boulders making up the wall were in the size range 0.30m to one
metre across. The wall forms an incomplete ring and the area within it
measures about 300 metres in diameter (Momber, 1991). A similar rubble
Fishing 137
stone sea-fish trap exists at Minehead. A net was placed over the entrance, to
catch fish as the sea ebbed (Aston and Dennison, 1988: 401).
Stone anchors were in use from prehistory to the post-medieval period and
are important but not fully explained marine artefacts. The very large ones
were probably for anchoring boats while the smaller ones may have been
used as additional weights on anchor cables or for anchoring fishing nets or
pots. It is worth distinguishing between stone anchors, where the stone is
used incombination with wooden stakes which enable it to grip the bottom
and so resist being dragged along, and stone weights which rely on gravity to
keep them in position.
Archimedes' principle states that any object immersed in liquid experiences
an upthrust equal to the weight of the liquid displaced. This means that a
large stone, which a man could hardly shift on dry land, becomes far easier
to lift on the sea-bed. The difference in weight of an object in water and in
air depends on its density, since to calculate the weight of a submerged
object, the weight of the volume of water displaced by the object must be
subtracted from its air weight. Lead is so dense that its weight is scarcely
affected by immersion but the weight of stone certainly is. The sinkers on the
foot-rope of a fishing net have to be sufficiently heavy to keep the slightly
buoyant hemp netting hanging nearly vertically, without submerging the
floats on the head-rope.
Stone anchors or weights have so far been found in the sea around Britain
only as stray finds, without any rope or other associated organic material
which can be used for dating. Recording the depth of water in which the
weights were found, and the nature of the bottom, may give some clue as to
the type of operation for which they were used. Markey has reported on two
stone anchors from Dorset and notes the existence of at least 16 others
(1991: 50). The considerable weight (37.5kg and 21kg) of the two which he
describes and the presence of rectangular holes which were probably for
wooden tines, suggest that these were for anchoring boats. They contrast
with the group of nine holed stones found by divers in about four metres of
water near the foot of a small cliff at the south-east tip of Guernsey. These
irregularly shaped stones, about 30cm long and 18cm thick, each have only
a single hole and weigh about 2kg (Hurst, 1969: 194-5).
Five undated stone weights have been examined at the National Maritime
Museum (figure 8.6). 'A' is disc-shaped and made of sandstone, about
48cm in diameter and 7mm thick with a central perforation. It was found
by a diver 400 metres off Seaford, Sussex. 'B', also of sandstone, was found
off Brown's Bay, Cullercoats, Tyne and Wear, in about 12 metres of water
(McCumiskey, 1980). 'C was found by a diver off Plymouth, Devon. It is
made of red sandstone with faceted sides and a hole at one end. This
anchor is different in shape from any recorded in the literature and has
been carefully designed to avoid chafing of the rope during use. 'D' is a
triangular unperforated stone weight found on the beach at Dunwich,
Suffolk.
138 Medieval ships and shipping
Nance (1913; 1921a; 1921b) has described the recent use of killicks,
composite anchors using stones and wooden components, in Cornwall. He
documented a great variety of forms. They were used for fishing boats
because the rough, rocky sea-bed claimed so heavy a toll of lost anchors that
more expensive ones could not be afforded (Nance, 1913: 298). He also
described how Marazion pilchard fishermen used a disc-shaped 'bully stone',
repeatedly dashing it into the water to frighten back the fish that attempted
to break out of the still unjoined ends of the seine. They chose stones of as
light a colour as possible to cause maximum alarm to the fish (Nance, 1913,
296-7). The pale disc-shaped stone from Seaford (figure 8.6a) would serve
very well as a 'bully stone'.
Fishing 139
Fishing boats
The design of fishing boats determined by whether they work off beaches
is
or out of harbours, the type of fishing gear operated, the nature of the
fishing grounds and the length of time they spend at sea. Local boatbuilding
traditions were strong and the same boat-types remained in use for centuries.
Excavation of fourteenth- and early fifteenth-century layers on part of the
medieval waterfront in Poole, Dorset, (Watkins, 1994) revealed a store of
timber for use in boatbuilding, as already described in Chapter 1. The
boatyard produced a number of boats of similar size and shape and there is
sufficient evidence to attempt a reconstruction of the type of boat represented
by the majority of the used timbers and rough-outs (Hutchinson, 1994: 35).
The reconstruction (figure 8.7) takes as its basis a keel 4.15m (13ft 6in) long,
scarfed to stem and stern posts each 1.83m (6ft) long, giving an overall
length of 7.81m (25ft 6in). The have been the same
stern post need not
shape as the stem post; a straighter post would have been more suited to a
stern rudder. The boats would have had ten or eleven strakes per side and
the beam is reconstructed as 2.45m (8ft). Boats of this size and shape would
have been suitable for fishing: their hull shape would provide stability and
their flat bottoms are suitable for beaching. A rope attachment hole low
down on the stem post is suitably placed for hauling a boat onto a beach.
Four Y-shaped mast crutches were found, intended to be used for supporting
masts when they were unstepped and laid along the boat. Depictions of
boats fishing, including those on a sixteenth-century map of Poole Harbour
(Hutchinson, 1994: 37), show boats working their nets with their masts
140 Medieval ships and shipping
that the net fell into the wreck from the quay, where it had been hung to
dry, but surely the more likely explanation is that was part of Blackfriars
it
Ill's equipment. This need not mean that Blackfriars III was exclusively a
fishing boat; there are plenty of documented examples of the use of boats for
various purposes at different times of year.
142 Medieval ships and shipping
To bring the catch of fish to shore in a fresh condition the Dutch adopted
The earliest known reference to these in
a type of boat called the waterschip.
the written sources occurs in 1339 and about ten examples have been
excavated in the IJsselmeerpolders. They are characteristically broad-
bottomed forward with a sharp stern. The stem is curved and the stern is
straight. A medieval example, probably of the fifteenth century, was
excavated at site MZ 22 in the polders in 1978. The midships part of the
boat was a 'bun', a fish-well compartment sealed from the rest of the vessel
by heavy bulkheads, with holes in the hull planking to allow sea-water to
flow through and thus keep the fish alive. This type of boat continued in use
into the nineteenth century but the post-medieval examples were built with
flush-laid planking rather than in the clinker technique (Reinders, 1979: 41-
2; Reinders, 1982: 21).
Documentary sources occasionally give information about fishing
practices. Heath (1969) has studied the accounts of the income of the
parish church of Scarborough for the years 1414-18 and 1434-42 which
include tithes from fishing expeditions. There were nearly 50 regular
fishermen in Scarborough and they fell into three categories. The largest
group mainly fished in 'batellae' and 'cobles'. 'Batellae' simply means 'small
boats' and 'cobles' were presumably the ancestors of the boat-type of that
name still used on the north-east coast (McKee, 1983: 88, 114). A quarter of
the fishermen sometimes employed 'farcosts', a boat-type of unknown
features (Burwash, 1947: 123), with examples recorded as being of from five
to 40 tons in the fifteenth century. Small boats were used to catch plaice in
winter, lobsters and cod during Lent, and skate in summer. The average
annual income of these fishermen was about £7. The second group of
fishermen fished for herrings in winter and were engaged on 'the North Sea
fare', catching cod and haddock on the spawning grounds east of Scotland
and the Dogger Bank, in spring. All of these also fished in farcosts, but only
two briefly used cobles or batellae. Their average gross annual income was
more than ten times greater than the first group, at £77. The third group
consisted of 16 fishermen who, in addition to fishing herring, went to Iceland
for cod in summer. Their average gross annual income was £85.
Richard Hakluyt, writing in the sixteenth century, claimed that the men of
Blakeney fished off Iceland in the reign of Edward III. There is no evidence
for this. It may be that the first English fishermen to sail to Iceland were
those whose ship arrived there in 1412 (Marcus, 1956: 313). Icelandic sagas
record they were followed in the summer of 1413 by 30 or more fishing
doggers from England. It is possible that some fragments of English doggers
lie on the sea-bed around Iceland: in 1419 no fewer than 25 English ships
were wrecked in a gale on Maundy Thursday. 'All the men were lost, but the
goods and splinters of the ships were cast up everywhere' (Marcus, 1956:
315).
Vessels called 'doggers' were used for the Iceland trade. They seem to have
originated in the Low Countries but by the fifteenth century the dogger was
the ubiquitous east coast fishing vessel (Marcus, 1954: 294). The term seems
to have been applied to a variety of medium-sized craft. Fifteenth-century
documents referring to East Anglian ports record them as ranging from 30 to
80 tons with crews of 20 to 30 men (Heath, 1969: 60). From 1483 the
Fishing 143
annual fishing fleet sailed to Iceland in convoy and the Iceland trade
probably reached its peak in the early sixteenth century (Marcus, 1954: 296).
The 'crayer' was another type of vessel mentioned in fifteenth-century
documents as comparable in size to doggers or a little smaller (Burwash,
1947: 120-24). Its name suggests it may originally have been used for
catching marine crustaceans but in the fifteenth century crayers seem to have
been used for carrying cargoes rather than for fishing.
It seems that around 1416-25 herring ceased to spawn in the Baltic and
the failure of the fishery there may have contributed to the decline of the
Hanse. The North Sea became the new focus of the herring industry
although Heath (1969: 61) has questioned whether the east coast fishery
grew as a result. Dutch fishing vessels worked way down the coast,
their
starting at Shetland in June or July, reaching Great Yarmouth in the autumn
and home from the Thames estuary after Christmas (Cutting, 1955:
sailing
63). Herring busses evolved in the Netherlands in the fifteenth century as
deep-sea fishing vessels with capacity for salting fish on board. Tradition has
it that the first busses were built at Hoorn and Enkhuizen in 1416 and by
1450 the whole of province of Zeeland had them. They were large,
seaworthy ships usually of 80 to 100 tuns burden, with a characteristically
high and narrow poop and very full in the body and the waterline. They had
a beam of 15 to 17 feet with a keel length of only about 50 feet. Their crews
of about 14 or 15 men included skilled picklers and coopers to make the
barrels tight. Busses were usually armed and frequently convoyed by
warships (Cutting, 1955: 66).
Fish processing
Fish were preserved by four different methods in the Middle Ages: drying,
smoking, salting and pickling. In conditions of cold and low humidity in
Norway and Iceland, white fish (cod, haddock, pollock and ling) were dried
out of doors with little or no pre-treatment with salt. The stockfish produced
in this way were traded throughout Europe. Jones (1992) has suggested that
they may be recognised in bone assemblages by the disproportionate ratio of
head to body bones. Smoking was either done cold, as for example with
bloaters, or 'hot' in a kiln. Recent practice in smoking on a domestic scale
can probably be projected back into the Middle Ages - Arbroath smokies
were smoked in pit kilns and in Sanday, Orkney, the poorer inhabitants lived
on and hung in the chimneys to smoke (Fenton, 1978: 528). Red
fish, salted
vinegar. Herring were gutted, washed in sea water, roused in salt and packed
in barrels. Towards the end of the fourteenth century Dutch herring cured in
this way were considered superior to those of any other country. They were
sorted and packed in casks, head to tail, with salt between the layers, which
were then arranged alternately across each other. Regulations stated that
they had to contain one part of salt to three parts of fish. The barrels were
made airtight and branded with the date of catch (Cutting, 1955: 62-3).
Pilchards were salt-cured and packed in barrels like herrings. As they are
fattier than herring, they were pressed in barrels to remove some of the oil
before topping up with brine and lidding (Cutting, 1955: 79).
Preserved fish were much cheaper than fresh fish and could be distributed
further. In 1338 Edward III obtained 400 lasts of herring (about half a
million fish) from Great Yarmouth, for the army in Flanders (Cutting, 1955:
35). The port books of Southampton for 1427-30 show that there was a
from Covehithe, Southwold,
large coastwise trade in salted herrings, chiefly
Dunwich, Lowestoft, Orwell and Walberswick in Suffolk, and Great
Yarmouth, Cromer and Cley in Norfolk. Other sources mentioned are
Newcastle, Easton near Portland and Penzance. The total recorded import to
Southampton in 1430 was 2,590 1/2 barrels, or about 500 tons (Cutting,
1955: 69).
Fishing settlements
By the thirteenth century, Great Yarmouth had become one of the major
herring markets of Europe. There are many records of monastic houses
sending ships there. These included Durham priory and several Cistercian
monasteries such as Waverley, Boxley, Robertsbridge and Beaulieu. The
monastic houses acquired property in or near the town as a base for their
operations (Bond, 1988: 75-6).
The villages of Porthoustock and Porthallow in Cornwall were used as
fishing bases by Beaulieu Abbey. In about 1240 the abbey purchased a
building in Porthoustock with an open space extending ten feet by 20 feet
outside for drying fish. In 1317 the Abbey acquired a plot at nearby
'Porthalon' with a slipway where the abbot's men could draw up their boats
(Bond, 1988: 76). Porthoustock today has a broad, well-sheltered, gently
sloping and firm gravel beach, onto which fishing boats are still drawn up
(figure 8.9). Steep, well-wooded slopes on each side would have provided
abundant fuel and a small river runs to the sea there. Porthallow has similar
topography but with a narrower beach and steeper slopes flanking a wider
valley with more space available for settlement. Neither of these places is
easily accessible by land and they must have relied largely on seaborne
communications. The Beaulieu Abbey fishermen seem to have used the ports
Fishing 145
Figure 8.9 The fishing cove at Porthoustock, Cornwall (photo: Steve Hartgroves,
Cornwall Archaeological Unit).
Fighting at sea
Figure 9.1 Crusaders attacking Damietta, Egypt, in 1248 (Matthew Paris MS 16,
f.55v.By permission of the Master and Fellows of Corpus Christi College,
Cambridge).
grappling irons. 'Gadds', pointed bars, of iron were hurled down from the
fighting tops. At the Battle of Sluys in 1340 some of the French ships had
boats lashed to their masts with a supply of stones for dropping onto their
opponents. It was advantage and the larger the ship,
crucial to have a height
the more fighting men could final phase was to board the
be carried. The
opposing vessel and overpower the defenders in hand-to-hand fighting. The
objective was to capture the enemy's ships rather than to sink them.
At Sluys more than 200 French ships formed three lines across the mouth
of the River Swyn. They were chained together until they drifted east
towards Cadzand and began to foul each other and then the chains were
hurriedly cast off. Ships were positioned according to their size with the big
Christopher, which had been captured from the English, in the front line
with other large ships. The English front line included the largest ships from
Yarmouth and the Cinque Ports and Edward Ill's flagship, the cog Thomas.
All the ships were fortified with timber-works, 'so that they looked like a
row of castles', according to the chronicler Froissart.
Trumpets were used for signalling in sea battles and are shown on some
town seals, such as that of Winchelsea of the thirteenth century (figure 3.1).
At the battle known as 'Les Espagnols Sur Mer', waged off Winchelsea in
1350 against a Castilian fleet fighting on behalf of the French king, the
148 Medieval ships and shipping
> *|\c tiir erf ticnr fencntrftr tti ftS- tcm^lYcf fotei
Figure 9.2 Fighting at sea, from a c.1270 edition of Vegetius' De Re Militari (Ms
Marlay Add. I, f.86r. Fitzwilliam Museum, Cambridge).
Ships in warfare 149
nobles in their ships were entertained before the battle by minstrels. When
the Spanish ships were sighted the trumpets gave the signal to launch the
attack.
Guns were first used in open field in the early fourteenth
warfare in the
century. Experiments with their use on board ship soon followed and may
have preceded their installation in purpose-built land fortifications, which in
Britain began in the Solent area in the 1360s. Early guns lacked accuracy,
range and power and their introduction to ships did not immediately change
the way in which sea battles were fought. Ships were equipped with a
relatively large number of small guns, many of them swivel-mounted, and a
few larger pieces. The large guns in particular took a long time to reload,
seriously limiting their effectiveness in an age when tactics were based on a
rapid closure and hand-to-hand fighting. Even an opening 'fusillade' would
have been difficult to aim and the idea of broadside fire had yet to develop.
At least until the end of the sixteenth century, including the Spanish Armada
engagements, guns were effective only as anti-personnel weapons and for
disabling ships, not for sinking them.
What evidence of sea battles might be preserved in the archaeological
record? We should not expect to find many shipwrecks. Occasional sinkings
resulted from collision, including Espagnols Sur Mer' (Lewis and
one at 'Les
Runyan, 1990: 125-6). The town seal of Hastings of the early fourteenth
century appears to depict a Cinque Ports ship slicing through another.
However, of the 213 French ships which took part in the Battle of Sluys, 190
were captured, 23 escaped and none sank (Sumption, 1990: 327). The bed of
the River Swyn must be littered with weaponry from the battle, including
some specifically maritime artefacts such as gadds, grapnels and signalling
trumpets. Debris from the ships burnt at the battle of Damme in 1213
(Brooks, 1933: 201) might also be preserved. As a result of reclamation the
sites of the harbours of Sluys and Damme, among the most important in
northern Europe for trade as well as for the sea battles fought there, are now
under dry land.
Most vessels involved in sea warfare were not purpose-built fighting ships.
The crown owned varying numbers of ships throughout the medieval period
but these were only ever a small proportion of those ships involved. Apart
from an arrangement with a group of ports in south-east England, the
Cinque Ports, which were granted certain privileges in return for providing
shipping services (see Brooks, 1929b), naval power was based on the king's
right to 'arrest' merchant ships and their crews - that is, to requisition them
in return for payment of compensation. Merchant ships were frequently used
as transports, taking men and equipment across the Channel and also to
fight in Scotland. English merchant ships were not built for speed and few of
them were of great size. This meant that the king had to supply his own fast
oared vessels and 'great ships'.
At the beginning of the medieval period there was already differentiation
in ship-types between warships and cargo vessels - shown by the Skuldelev
150 Medieval ships and shipping
Oared vessels
William the Conqueror's invasion fleet appears from the Bayeux tapestry to
have consisted of vessels propelled by oars as well as by sail. There were at
least 700 ships, mostly supplied by his leading vassals (Brooks, 1929a: 18).
In 1196 Richard I built a fleet of naves cursoriae in order to defend the Seine
(Brooks, 1929a: 24). As these were suitable for fast travel on rivers and the
sea we can assume they were oared. In 1213 naves cursoriae were used
against the French at Damme, when the English anchored some distance off
and used the oared vessels to attack the French ships at anchor.
Galleys were important naval vessels in the thirteenth and fourteenth
centuries and they are frequently mentioned in documentary sources. There
is, however, a surprising absence of pictures of them. In 1205 King John had
vessel had the sharp ends and fine run which would be expected of a galley.
Galleys were used for convoy duty, for stopping and searching ships and
for enforcing customs regulations. They also had a role in raising fleets for
the king; in 1207 galleys were ordered to arrest all ships they met at sea
(Brooks, 1929a: 44). The galleys kept at Chester and Bristol in the 1240s
were used for offensive purposes and for victualling castles in Wales and
Ireland (Brooks, 1933: 152). Two galleys equipped with crossbowmen took
part in the winter blockade of La Rochelle in 1242 (Lewis and Runyan,
1990: 133-4).
In 1294, probably in response to Philip the Fair's establishment of the clos
des galees at Rouen (see page 78), Edward I ordered 26 towns between
Newcastle and Plymouth to share the building of 20 galleys for the war with
France and their construction has already been described in Chapter 1. These
galleys were to be vessels of 120 oars each, except one at London which was
to have 140 oars. The Newcastle account itemises a keel in two pieces, with
a combined length of 108 feet (32.92 metres). Allowing for scarves we can
assume a keel length of 100 feet (30.48 metres) with stems and sterns
bringing the total overall length of the galley to about 140 feet (42.67
metres). Postulating a racey length-to-beam ratio of 7:1 would give a beam
of 20 feet (6.1 metres). The only other dimensions given which might be
useful in establishing what these vessels were like are those for the planks.
Up to the seventh week, 595 planks had been bought for the Southampton
galley with a total length of 12,444 feet (3,793 metres), or 6,222 feet (1,896
metres) for each side of the hull. The accounts do not say how wide the
planks were but even if they were only eight inches (20.3cm) there would
still have been ample timber for making bottom boards as well as planking
the hull.
The number of oars specified for the 1294 galleys raises problems.
Although they were supposed to have 120 oars, they nearly all seem to have
been equipped with fewer. The York galley had 97 oars, those of Ipswich
and Dunwich 100 each. With oarsmen seated one behind the other there is
not enough room to fit a single bank of 60 oars into one side of a ship with
a keel length of about 100 feet. The absolute minimum oar spacing with this
arrangement is 28 inches (71cm) but to fit in 120 oars it is necessary to have
one oar for every 20 inches (50.8cm) of keel on each side of the galley. The
Newcastle galley account records oars of two sizes: 22 to 23 feet (about 6.86
metres) and 16 to 17 feet long (about 5.03 metres). These might have been
used in pairs at the same level, working on oar pivots quite close together.
Figure 9.3 suggests how this might have been achieved, with alternate
spacings of 12 inches (30.48cm) and 28 inches (71cm). These spacings
represent an attempt to cram the maximum possible oar-power into the
minimum possible space. They would require a short chop rowing action
with little body movement and whether the oars could have been operated
efficiently is open to question.
Whether for reasons of ease of supply or better design, when Edward II
wanted galleys in 1317 he bought five from Genoa. More were hired from
Genoa in 1336, though in the same year Edward EI had a galley built in
England, which was based at King's Lynn, and another in 1347 at
Winchelsea (Tinniswood, 1949: 277). Galleys were used intensively after the
152 Medieval ships and shipping
time. The was of 100 tons and had 48 oars and is also
larger balinger
termed a 'barge' in the building account. The smaller was of 80 tons and had
38 oars (Carpenter Turner, 1954: 70). In 1418 one merchant and two royal
balingers together captured two ships, probably Spanish, carrying iron and
wool (Carpenter Turner, 1954: 63).
Even when they were not involved in official combat, merchant vessels
increasingly had to be able to defend themselves against attack at sea. For
example, when war was declared in 1243, masters of all ships were licensed
and encouraged to annoy the king's enemies. Ships of neutrals carrying
enemy goods were also considered fair game. In the thirteenth century the
use of banners of arms to distinguish port, nationality and ownership became
general.
Town seals of medieval ports provide important information about the
modifications made to ships for fighting at sea (figure 9.4). The earliest
representations of castles show them as flimsy fighting stages inserted inside
the ends of the hull. For example, on the seal of Hythe of the late twelfth or
early thirteenth century (Brindley, 1938: fig. 5) the fighting stages do not
project above the stem tops. The seal of Pevensey of 1230 or earlier
(Brindley, 1938: fig. 12), very similar to that of Winchelsea (figure 3.1) of the
early fourteenth century, also shows openwork stages of slightly increased
height. None of these ships has a fighting top. A fighting top shown on an
is
earlier seal, that of Dunwich, from the late twelfth century (figure 9.4a). The
top square and placed centrally on the mast. Square fighting tops attached
is
to the forward side of the mast are shown on the late thirteenth- or early
fourteenth-century seals of Great Yarmouth (Brindley, 1938: fig. 21),
Sandwich (Brindley, 1938: fig. 15) and Faversham (figure 9.4b). The change
of position is probably a result of the addition of ratlines to the rigging. The
Faversham ship appears to have a line of crenellation on the quarter, above
the rudder, as well as on the stern castle.
The ship on the seal of Dover of 1305 (figure 9.4c) also has a square
fighting top on the forward side of the mast. Although it is contemporary
with the seal of Winchelsea (figure 3.1) it has castles which are clearly
becoming integrated with the hull structure. The castles are longer and are
carried out over the stems and given support by the stem posts. The ship
shown on the thirteenth-century seal of the town and Admiralty Court of
Poole (figure 9.4d) is quite unlike the ships of the Cinque Ports and
Yarmouth referred to so far. The sides of the hull are higher and its castles
and top are large. It looks like a ship purpose-built for war rather than a
fortified merchant vessel.
For the expedition to Ireland in 1210, King John hired five Friesland cogs
(Brooks, 1929a: 29) and a cog was also used by the English at the battle off
Dover in 1217 (Brooks, 1933: 218-19). Cogs are depicted in a seafight in
the Decretals of Gregory IX, an English manuscript of the early fourteenth
century (figure 9.5) and had become important in warfare by the beginning
of the Hundred Years War.
154 Medieval ships and shipping
Figure 9.4 Town seals of: (a) Dunwich, (b) Faversham, (c) Dover, (d) Poole
(National Maritime Museum).
1347, which consisted of 738 ships with 15,000 mariners. It took 32,000
troops to the siege of Calais. There were frequent complaints throughout the
Hundred Years War from shipowners all round the coast about the effect on
trade of the arrest of ships and the impressment of crews. They considered
the payment of three farthings a ton to be inadequate compensation.
Royal ships
The identification of the large wreck which lies in the River Hamble in
Hampshire as Henry V's Grace Dieu is based on its large size - with a keel
at least38 metres (127 feet) long - and its location (Anderson, 1934: 161;
Prynne, 1968: 120). Only the very bottom part of the hull remains and has
been the subject of archaeological investigations in the 1930s (Anderson,
1934 and 1938; Prynne, 1938a and 1938b) and by the Archaeological
Research Centre of the National Maritime Museum in the 1980s (Clarke et
al., 1993). The Grace Dieu is notoriously difficult to investigate as a 'dry'
sitebecause the wreck is uncovered only at the equinoctal spring tides and
then for only brief periods (figure 9.6). The viscous Hamble mud impedes
excavation and obscures features. Some recording has been
structural
achieved by divers and probable that future work on the site will be
it is
carried out under water (McGrail, 1993b). The structure of the surviving
part of the hull has been discussed in Chapter 2. Grace Dieu was the
English answer to the carracks, such as those from Genoa used by the
French at Harfleur and in the Channel raids. The statesman in command of
the annual Florentine merchant galley fleet to England and Flanders was
given a tour of the ship in 1430, while it was moored in the Hamble, and
he recorded that he had never seen so large and beautiful a construction
(Friel, 1993: 17). Grace Dieu's inventory of 1420 records three guns with
of more than 135 tons, had two iron guns, three-dozen headed darts, one
dozen headless darts, 98 gadds of iron and a grappling iron and chain. In
1413 Thomas de la Toure, a single-masted ship of a little under 200 tons,
had four guns with three chambers each and one firepan. The Holigost is
listed in 1415 as having two guns and 81bs of gunpowder. By 1420 this is
increased to seven guns and 12 chambers. The Grand Marie de la Toure a
Ships in warfare 157
158 Medieval ships and shipping
Figure 9.7 A battle at sea in the fifteenth century, after the introduction of artillery to
ships (Julius E IV art 6 f 18v. By permission of the British Library).
single-masted ship of 116-40 tons, had in 1416 three guns with ten
chambers, four-dozen darts, 15 lances, 21 bows, 15 sheaves of arrows, 30
iron gadds and one grapnel.
It is interesting that the inventories of carracks captured at Harfleur in
1416 and in other actions at about the same time list no more than two or
three guns each. The Christophre, a Genoese carrack of 600 tons, had two
guns with four chambers, while the George, another Genoese carrack of 600
tons, captured at Harfleur, had in 1420 three guns with six chambers as well
as three scaling ladders and six hooks. Another carrack captured at Harfleur,
Ships in warfare 159
the Marie Sandwich of about 500 tons, was listed as having no guns but
only a seizing grapnel and 16 fathoms of iron chain, weighing 3 cwt 31 lb.
There are several pictorial representations of warships at the beginning of
the fifteenth century. The seal of Tenterden in Kent (close to Smallhythe,
where royal ships were built) of the first half of the fifteenth century (Ewe,
1972: fig. 201), that of Admiral Beaufort of 1418-26 (figure 2.1a), and the
fifteenth-century seal of the Sub-Admiralty of England (Brindley, 1938: fig.
33) show ships of similar rigging and proportions. By 1400 the fighting tops
were tub-shaped and placed centrally on the mast above all the rigging. The
run of the hull planking varies and includes various hybrids of the hulk. The
relief carving on a bench-end from the church of St Nicholas at King's Lynn
(figure 3.6), probably made in 1415-20, shows a large, two-masted clinker-
built ship equipped for war.
The Hastings manuscript, of the late fifteenth- or early sixteenth-century,
shows clinker-built warships with wales (figure 10.7). The stems have
become almost vertical as the fore castles were no longer used as seige
towers but contained heavy guns which needed buoyancy beneath them. The
stern castle is fully integrated with the lower hull and has a pair of ports for
big guns. By the 1470s carvel-building had been adopted for great ships,
three masts were common and guns were an essential part of the armament.
The building account for the Regent, a 1,000 ton four-masted royal ship
of 1487, specifies that carvel nails were used. A slightly smaller ship, the
Sovereign, of 800 tons, was built in the following year. The Sovereign may
have been first built in clinker technique then converted to carvel during the
rebuild in 1509, if the remains of a hull found at Roff's Wharf, Woolwich
are really from this ship. The Woolwich ship was discovered in 1912 during
excavations for the foundations of a new engine house for a power station.
Its identification as a warship was suggested by its large size and relatively
light framing. An Admiralty Committee reported in 1914 that the keel length
was well over 100 feet (30.5 metres), the beam of the ship was about 40 feet
(12 metres) and that therefore the burden must have been considerably over
1,000 tons. The discovery of some stone shot within the wreck does not
necessarily strengthen the argument, since merchant ships as well as warships
would have carried guns. The wreck's position in the riverbank suggested
that it must be earlier than the seventeenth century. The ship appears to have
been laid up in an inlet of the south bank of the Thames, abandoned and
partly broken up. It was found at a depth of 18 feet (4.5 metres) below the
contemporary ground surface and about six feet below the high-water mark.
The contexts of associated finds were not adequately recorded to provide
dating evidence (Salisbury, 1961: 82).
Anderson (1959) thought the wreck might be that of Henry VM's Henry
Grace a Dieu which was destroyed by fire at Woolwich in 1553. An earlier
date was suggested by the fact that the frames appear to have the traces of
joggles on their undersides, indicating that the ship had originally been built
with clinker planking (Salisbury, 1961: 86). Salisbury suggests that this
would explain the shallowness of the frames, which were 14 inches (36cm)
wide but only eight inches (20cm) deep, with five inch (13cm) spaces in
between. The framing was stiffened by riders 18 inches by 16 inches (46cm
X 41cm) spaced about four feet six inches (37cm) apart. The ceiling planking
160 Medieval ships and shipping
was three inches (7.6cm) thick and the outer hull planking was five inches
(13cm) thick. The nature of the fastening was not described but the
superintending architect commented that the whole thing was very securely
pinned together, making the removal a very difficult operation.
The Woolwich ship excited considerable interest at the time of its
Figure 9.8 French fifteenth-century hulks with gunports through the planking at the
bows (photo: Bibliotheque Nationale, Paris. Ms Fr 2829, f.32,v).
Stability was seriously affected by the presence of heavy guns on the upper
deck and in the castles. In a sea trial in 1512, it was reported that the Christ
could not sail as the result of being 'overladen with ordnance' (Laughton,
1960: 284). The solution to this was to mount guns lower down in the hull,
firing through gunports. There is a tradition that in 1501 a French
shipbuilder, Descharges, was the first to cut holes in ships' sides for gunports
and to fit them with hinged lids. Fundamentally, the level of the lower deck
had to be raised by up to a metre in order to keep the gunports safely above
the waterline. The wreck of the Mary Rose testifies to the fact that the
importance of this modification was not fully appreciated at the beginning of
the sixteenth century and stability problems persisted much longer.
The oldest gun so far found in British waters may be that from the
Studland Bay wreck, which probably sank in the 1520s. Presumably Spanish,
it was more than two metres long and constructed from iron staves formed
into a cylinder. It was found with its breech chamber wedged in position and
a six-inch stone shot in place. Iron bands appeared to have been used to
fasten gun to its carriage. On the same site another, larger breech
the
chamber was found, indicating that an even bigger gun was on board, as
well as part of a smaller four-inch, breech-loading swivel gun and its
162 Medieval ships and shipping
chamber. Finds of two inch and three inch shot indicate that smaller calibre
guns were also carried (Ladle, 1993: 22-3).
Two early guns were found on a dispersed wreck site at Walney at the
entrance to Morecambe Bay in the nineteenth century. They are also of bar
and hoop construction with breech chambers (Archibald, 1844). Laughton
(1960: 266) calculated that the larger gun was for stone shot of at least
140 lbs and that the powder chamber capacity was about 3Vi lbs. These
guns are in the Royal Artillery collection at the Rotunda Museum,
Woolwich.
and was only ever spasmodically successful (Sumption, 1990: 227; White,
1930). A survey of royal coastal fortifications was made in 1335 and
improvements were made at the Tower of London, Carisbrooke, Dover,
Pevensey and Porchester (Sumption, 1990: 250).
Throughout the Hundred Years War fleets of fast-oared vessels burnt ships
at their berths and anchorages. The English burned the coastal villages of
Normandy in a deliberate policy of damaging the communities which
sustained French seapower in the Channel. In return the French mounted
frequent raids on England, notably at Southampton in 1338, the Isle of
Wight in 1369, Rye in 1377 and Winchelsea in 1380. Against this threat the
River Hamble, which was used as an anchorage for the royal ships, was
fortified with a wooden bulwark and a spiked wooden defence pale while
Grace Dieu was being built, beginning in 1416 (Carpenter Turner, 1954:
57).
The introduction of artillery opened up new possibilities for defending the
coast. Guns were mounted in fortifications in the 1360s, particularly in the
Solent area and Kent. Edward Ill's castle at Queenborough, Kent, was built
to a circular plan in the 1360s to deploy both guns and trebuchets (Saunders,
1989: 18-19). Quarr Abbey was probably equipped with guns in about
1365 and in about the mid-1370s keyhole gunports were inserted in the
walls along the western waterfront at Southampton. These early gunports
were intended for close defence.
The royal castles at Portchester and Carisbrooke were provided with guns
in the 1380s, as was Cooling Castle, near the mouth of the Thames in Kent.
In 1405 Plymouth successfully used artillery fire to repulse a raid. English
ships there were protected by a barrier of barges (Oppenheim, 1968: 18-19).
In 1420-22 a round masonry gun-tower was built at the mouth of
Portsmouth harbour to protect the king's ships. A timber tower was built on
the Gosport side and a chain boom defence was stretched between them
across the harbour mouth. A chain boom defence was also installed at
Fowey in 1457. Bulwarks built of earth revetted with timber were
constructed round several harbours. In 1451 a two-storey high bulwark
within a water-filled moat was built at Sandwich. It proved inadequate in the
French raid of 1457 and had to be repaired and supplemented by another,
this time of brick, at Fishergate.
The 'strong tower and bulwark', built Dartmouth from 1481 and
at
replacing an earlier castle, was designed for large guns.It was sited near the
water's edge and close to sea-level. It demonstrates that even before 1500
fortifications were being built with sufficient firepower to be an effective
defence against hostile shipping.
10 Pilotage and navigation
The ability of mariners to conduct their ships safely from one port to another
depends on knowledge and skill. Knowledge of natural forces and cyclical
events is still fundamental to sailing today and can be inferred back into
prehistory. Documented voyages, such as that of Richard the Lionheart, who
sailed with a northern European Crusader fleet around Iberia and through
the Mediterranean to Palestine in 1191, indicate a thorough knowledge of the
maritime geography of Europe at an early date, as well as testifying to the
seaworthiness of the ships used. Documentary sources show that scientific
approaches to navigational problems were being developed by scholars.
Bede's eighth-century treatise on the tides and the early thirteenth-century
tide tables forLondon Bridge, compiled by the monks of St Albans (Taylor,
1971: 136), reveal attempts to understand the theoretical principles under-
lying empirical knowledge.
Thirteenth-century Icelandic sagas include sailing directions for the north
Atlantic (Marcus, 1980: 112), showing that oral lore was long established.
Sailing directions were written down at least as early as the mid-fifteenth
century. These 'rutters' (from the French routier) are invaluable records of
the fields of information and which were considered most useful
specific data
at the time. More than that, they point to a revolution made possible by
literacy, which allowed the dissemination of information to supplement oral
communication and each navigator's personal experience. The art and
science of chart-making, which presents this information in graphic form, has
been well described by both Waters (1958) and Taylor (1971). Documentary
sources also refer to navigational instruments and the skills required for their
use: for example, describing early versions of the magnetic compass and
listing equipment in ships' inventories.
There are two main classes of material evidence which can contribute to
the study of medieval navigation: navigational instruments and man-made
sailing marks. The date of introduction of new instruments is often obscure
and can only be clarified by archaeological finds. Shore features, such as
church towers, are commonly regarded as part of the landscape but the way
they figured in the seascape also needs to be appreciated. The siting of
prominent structures may have been influenced by navigational consider-
ations.
Shipwrecks themselves cannot be taken as evidence of navigational error,
since it is perfectly possible to be wrecked while in no doubt about the ship's
position and heading. Losses are as likely to be caused by failure of the ship
or its equipment, frequently as a result of adverse weather.
An and practice of pilotage and navigation is given in
outline of the theory
this chapter to provide background for the function and design of the
instruments and the location of sailing marks; and also to explain, to some
extent, why ships sailed the routes they did. Combining information about
medieval sailing routes and methods of coastal navigation contributes to an
understanding of what was expected of ships, particularly their ability to
Pilotage and navigation 165
maintain their position against a head wind and so escape being blown onto
a lee shore. It must be emphasised that the idea that medieval ships 'hugged
the coast' is a misconception. There was coastwise sailing but the risk of
being driven onto a lee shore meant that ships stood well out to sea to avoid
danger. Ushant in Brittany was and still is particularly lethal, with numerous
rocks far off shore and frequent fogs. Twentieth-century sailing directions
say that 'Ushant must not be sighted', for if a ship comes within ten miles it
may become subject to a powerful onshore set and be swept onto the rocks
(Taylor, 1971: 4).
The business of passage-making is today regarded as being split into two
branches: pilotage for coastal voyages, when the vessel's position can, at least
and navigation undertaken out
intermittently, be fixed in relation to the land;
of sight of known land, where the on the earth's surface has
vessel's position
to be established by celestial observation. There is of course some overlap
between the two. The core of pilotage had been known, in rudimentary
form, in the prehistoric period whereas the techniques, knowledge and
instruments belonging to ocean navigation were innovations during the
medieval period, developed in the course of voyages of exploration.
Pilotage
parts of more distant high features will stand up above the horizon and can
be seen from further off. A feature would have to reach at least 35 metres
above sea level for our observer at a height of 15 metres to be able to see it
at a distance of 20 nautical miles.
The mast-head height above sea level of Skuldelev 3, the smaller eleventh-
century trading ship from Roskilde Fjord, was only about ten metres. The
top of the mast of the fourteenth-century Bremen cog was about 20 metres
above the sea and the mainmast of the fifteenth-century great ship Grace
Dieu was said to be 200 feet (61 metres), high enough to make it possible to
see the sea horizon 16 nautical miles away. This is shown schematically in
figure 10.2.
Ships were fitted with tops or 'crow's nests' for look-outs, not just for
fighting. Although it was necessary to sight the coast from time to time to
establish position, ships did not hug the coast as thiswould mean entering
the shipwreck zone. Risks intensified when the wind was blowing onshore.
The term 'landfall' means sighting or falling in with the land, not landing on
it.
By the time the rutters were compiled, the magnetic compass was in
general use for determining direction. In earlier times, including the first part
of the period under discussion, northern seafarers had managed ro find their
way without compasses, principally by knowledge of the relative positions
and movement of heavenly bodies across the sky. The sun rises from an
easterly direction and achieves its highest altitude at noon when it is due
south, then sinks in a westerly direction. At night, north is indicated by the
Pole Star which maintains a constant position in the sky, unlike the moon
and the other stars. The Pole Star was known as the scip steorra to the
Anglo-Saxons and its use is mentioned in Icelandic sagas. Bearings obtained
from heavenly bodies were approximate and allowances would have to be
made according to the season of the year and the time of day or night. An
artefact of about AD 1000, found at a settlement site in Greenland, has been
interpreted as a bearing dial for trans-Atlantic sailing (Vebzek and Thirslund,
1992).
Direction can also be estimated in relation to wind and swell. In northern
waters winds from different directions may be recognised by the weather
accompanying them; thus a warm wet wind blows from the south-west and a
cold wet wind from the north-west (McGrail, 1987: 281). Medieval seamen
had to understand wind and weather patterns in order to take advantage of
Pilotage and navigation 167
favourable winds for voyage making and to avoid being taken by surprise by
storms. Icelandic sagas repeatedly refer to observations of the clouds, wind,
sea and the behaviour of seabirds (Marcus, 1980: 101). Vikings are said to
have released birds from their ships to indicate the direction of the nearest
land. Other indicators are cloud formations, the flight of birds at dawn and
dusk, and the presence of seaweed in the water.
The invention of the magnetic compass and its introduction to ships is
obscure but it was in use among northern and Mediterranean seamen in
the twelfth century. Magnetic iron ore was mined in Elba and carried by
ships of Amalfi. Tradition has it that Amalfi seamen were the first to use
compasses at sea (Waters, 1958: 21). The construction of compasses will be
described later but in their early forms they were difficult and unreliable to
use at sea. The earliest descriptions of a compass are by an English monk,
Alexander Neckham, writing in the 1180s. He said that it was used during
cloudy weather when the sailors could not see the sun or stars (Taylor,
1971: 95). Once the initial problems had been overcome, the compass
allowed greater precision in direction-finding and by the thirteenth century,
rhumb (direction) lines on the oldest surviving chart show that the
seaman's horizon had been divided into thirty-two compass-points (Waters,
1958: 21).
In order to fix a vessel's position it is necessary to know not only what
course is being sailed but also how fast the vessel is travelling. A mariner
would become familiar with his vessel's speed capabilities on the different
points of sailing (such as tacking into the wind or running before it) and in
different wind strengths. On coastal voyages he could check his estimates
against known distances on the shore. When out of sight of land a piece of
wood could be thrown into the sea and an estimate could be made of the
speed at which it was left behind. (As we shall see, the relative position of
the wood and the ship would also indicate the ship's leeway angle.) The date
at which a length of line was attached to the 'log' before it was thrown over,
so that the ship's speed could be measured rather than just estimated, is not
known, but may be after 1500 rather than before.
Pilotage and navigation 169
compass bearing of the moon at the moment of high water (Waters, 1958:
31; Frake, 1985).
Knowing the depth of the water under the vessel and the nature of the
sea-bed were very important factors in making a safe passage. It was
essential to avoid running aground and knowing the composition and
contours of the sea-bed could help to establish the vessel's position. The sea-
bed off western Europe slopes down to a depth of about 100 fathoms at the
edge of the continental shelf and then drops steeply to the ocean bed.
Measuring the depth of water under the ship by means of a line with a
weight at its end was termed 'taking soundings' and when a ship was 'in
soundings' it was over the continental shelf, in water less than 100 fathoms
deep. Off Ireland, Spain and Portugal the edge of the continental shelf is
only ten or 20 miles offshore but it is much further from the Lizard
peninsula of Cornwall. On a voyage from England to Spain a ship passed
out of soundings when some 100 miles south-west of Ushant and did not
enter them again until perilously close to the coast of north-west Spain
(Waters, 1958: 18-19). On the return voyage, the run from Spain to the
British Isles was made directly and navigators would take soundings to
locate the steep edge of the continental shelf about 100 miles west of
Penmarch Point (Taylor, 1971: 135).
The Lansdowne Manuscript sailing directions contain detailed information
about soundings and the nature of the sea-bed (Waters, 1967: 194-5).
Sounding leads had a concavity in their under-surfaces which were filled with
tallow, to which a sample of the sea-bed sediment would adhere. Northern
seamen were expert in knowing where they were by this means. For
example, the ground to the south-west of Scilly consists of a matrix of red
sand with white shelly inclusions and near the Lizard the bottom was said to
be of 'ragged' pebbles the size of beans.
170 Medieval ships and shipping
Navigation
Sailing marks
The term 'sailing mark' is used here to denote any man-made object or
structure placed so as to be visible to mariners and to serve as an aid to
navigation. They range from simple sticks in the mud to lighthouses.
Branches have been universally used to show the edges of mudbanks and
shallows, with larger poles in exposed places. Much interesting information
about the changing course and depth of channels could be obtained if an
accurate and inexpensive method of dating roundwood stakes were to
become available. Stakes are also used as leading marks in transit lines. Two
objects are said to be 'in transit' when one is directly in front of the other.
The back marker needs to be large and prominent, such as a church tower.
The transit line between the leading mark and the back marker may indicate
a safe channel. To aid recognition, different topmarks such as baskets or
barrels could be fixed on the poles. In medieval northern Europe trading
places were marked by having a mask, or Grimskalle, on a pole as a
seamark at the harbour entrance (Cederlund, 1989: 90).
In Viking Age Norway stone cairns were erected as seamarks (Morcken,
1969: 7). Similar features were probably common around the coast of
Britain. Investigation of a mound at Tywn Llewelyn, Glamorgan, showed
that was composed largely of natural rock, but heightened by building a
it
cairn on top. It probably marked the channel of the River Thaw during high
tide (Wilson and Hurst, 1957: 170).
Britain does not seem to have adopted the eleventh-century Norwegian
Pilotage and navigation 171
practice of erecting tall stone crosses as sailing marks. The tallest of them
may have been up to 15 metres high (Morcken, 1969: 32-3). The siting of
churches in positions where they could be useful for navigation perhaps
suggests a maritime economic factor in ecclesiastical foundations. It is surely
more than a coincidence that the church at Bosham, the seat of EarlGodwin,
which appears on the Bayeux tapestry as a record that Harold from
sailed
here in 1064, lies directly in line with the last 1.5 kilometres of the narrow
navigable channel in Bosham Creek (figure 10.3). Many other churches
served as sailing marks. Along the Suffolk coast church towers such as those
at Covehithe and Kessingland were the only tall structures. They are still
Figure 10.3 The siting of the church at the head of the creek at Bosham, marking the
navigable channel (Crown copyright. Reproduced from Admiralty chart 3418 with the
permission of the Hydrographer of the Navy).
Pilotage and navigation 173
M LDlEVA L LIGHTS
A COMflUMlD BY DOC UM£ N TA R Y OR
STRUCTURAL EVIDENCE
-a TRADITIONAL
I i S
Figure 10.4 Map of medieval lights (from Hague and Christie, 1975).
Medieval ships and shipping
//// VV,^^
Figure 10.5 St Catherines lighthouse, Isle of Wight (from Stone, 1891. Photo: Isle of
Wight Cultural Services Department).
Navigational equipment
As already noted, sounding leads were attached to lengths of line and used to
measure the depth of water under a vessel and to retrieve samples of the sea-
bed. They were used by the Romans but it is not known when they were first
used by the people of northern Europe. They were certainly standard
equipment long before they were first documented in royal ship inventories
in the fifteenth century. Sounding leads were apparently always made of lead
and not of iron or bronze. Lead has several advantages. It does not have the
corrosion problems of iron and was not as expensive as bronze. Its softness
was also useful as a rocky ground could be recognized by fresh dents and
cuts in the lead. Its most important property, however, is its density. A
sounding device has to sink fast; otherwise, despite all efforts to remain
stationary, the ship will have moved on and the reading will not be a true
vertical depth (see figure 10.6). The records of nineteenth-century Admiralty
experiments show that a sounding lead might commonly take 45 seconds to
reach 100 fathoms and that a cylindrical form presents the least surface for
friction. The line needs to be as thin as possible to minimise the braking
effect produced by its buoyancy and resistance (Davis, 1867: 2-8).
No medieval sounding leads have yet been identified, though one is
depicted in the fifteenth-century Hastings Manuscript (figure 10.7). The
earliest examples known from post-Roman Britain are those found with the
sixteenth-century Rother barge (Rice, 1824) and the Mary Rose. That from
the Rother barge is recorded as a sketch in the margin of an engraving of the
scene of the excavation. It was octagonal, about eight inches (0.20 metres)
high and of slightly larger diameter at the base than at the top. It had a
concavity in the bottom about one inch (25 mm) deep. The top had a ring,
worked in the solid, with an internal diameter of approximately a quarter
inch (6mm) for the attachment of the line.
The sounding leads for locating the 100-fathom sea-bed contour must
have had rather more than 600 feet (183 metres) of line attached. In the
seventeenth century there were two sorts of lead and line: a heavy, deep-sea
lead on a thinner line and a lighter lead (weighing 7lbs or 3.2kg and
measuring one foot or 30cm long) for depths of less than 20 fathoms (36.5
metres) (Waters, 1958: 19-20).
Although documentary sources tell us that the magnetic compass was in
use on ships in northern Europe in the twelfth century, the earliest survivals
176 Medieval ships and shipping
lathoms
Figure 10.6 Taking soundings, the effect of drift (National Maritime Museum).
we have are those from the Mary Rose, which sank in 1545, and they show
similarities to the earliest illustration of an English sea compass, which dates
to 1562 (Waters, 1958: 26; Rule, 1982: 118-21). The compass described by
Alexander Neckham in the 1180s was made by magnetising a needle by
rubbing it on a piece of magnetic ore (a lodestone). The needle was then put
through a reed at right-angles so that it floated on a bowl of water and
indicated the four cardinal points (Waters, 1958: 22). An instrument which
had a bowl of water as one of its main components would be of limited
usefulness at sea. A French treatise of 1269 records that a type of compass
had been developed which was dry, not wet. The compass needle was
mounted within the bowl on a vertical axis with a pivot at each end. The
edge of the bowl was fitted with a graduated ring.
Compasses with a bare needle and a compass-card or wind-rose, either
underneath it or marked round the rim, were prone to errors of parallax. To
Pilotage and navigation 177
Figure 10.7 Swinging the lead, from the Hastings Manuscript of about 1500 (The
Pierpont Morgan Library, New York. M.775, f.!38v).
178 Medieval ships and shipping
how to construct such a 'Little Ship of Venice' (Kragten, 1989). The mast
pivots and has a latitude scale calibrated for Exeter, London, Oxford,
Northampton and York. When the height of the sun was observed through a
pair of sighting holes on the stem and stern, a plumb line suspended from the
appropriate latitude on the mast intersected the marks for the hours of the
day engraved on the hull. The instrument would be accurate to within about
a quarter of an hour.
Traverse boards were used to record the course steered and the length of
time, measured by sandglasses, that the ship had been sailing on each
bearing. The lid of the Red Bay binnacle (Grenier, 1988: 79) had a
chequered pattern on it which may have been used for this purpose. Other
examples surviving from the sixteenth century have the compass rose marked
180 Medieval ships and shipping
on them and little holes for pegs (Waters, 1958: 36). Again, their date of
introduction is unknown but as they were used when sailing without the
benefit of charts an early origin is to be expected.
The making of sea charts was
strongly influenced by the Arabs and
throughout most of the medieval period manufacture was concentrated in the
Mediterranean. The oldest surviving maritime chart, the Carta Pisana which
dates to about 1275, plots the Mediterranean coast in some detail but shows
the west coast of Europe beyond Cape St Vincent only sketchily. This
contrasts with the Genoese charts of the first quarter of the fourteenth
century which are accurate and detailed as far as southern and western
Ireland and parts of Scotland (Taylor, 1971: 112). The Great Catalan Atlas
of 1375 shows further progress in cartography, in both scope and accuracy.
The Catalans in the Balearics and the Venetians were the leading chart-
makers until the sixteenth century. It appears that northern seamen did not
use 'sea-cardes' until they ventured into ocean navigation (Waters, 1958: 14).
The oldest surviving English maritime chart dates to the 1530s and is of the
Thames Estuary (Taylor, 1971: 193).
Charts are extremely unlikely to be preserved on shipwreck sites but their
former presence may be inferred if instruments for use with charts are found.
For example, the navigational instrument assemblage from the Mary Rose
includes dividers and a slate protractor (Rule, 1982: 121).
Globes were to become important in developing geographic understanding
and in the discussion and record of voyages and discoveries. The earliest
surviving terrestrial globe was commissioned by the city council of
Nuremburg in 1490. John Cabot (1450-98), who discovered Newfoundland,
used a globe to mark his voyages (Scammell, 1981: 73).
A simple though inaccurate method of estimating the speed of a vessel is
to drop a floating object into the sea and watch how quickly it is left behind.
The speed of separation can be measured rather than guessed if a line of
known length is tied to the float and allowed to run out from the ship. The
log and line is first mentioned in 1574 in a treatise called the Regiment of the
Sea and was apparently in use long before. Measurement was made of the
length of line paid out during one minute, timed by a sandglass (Taylor,
1971: 201).
The quadrant and the astrolabe (figure 10.9) were both instruments for
measuring the altitude of sun or stars. The earliest recorded mention of how
to take an observation, from 1456-57, refers to the quadrant rather than the
astrolabe. The earliest known use of the astrolabe by a mariner is 1481
(Scammell, 1981: 46) and the earliest surviving marine astrolabe dates to
1540 (Stimson, 1988). Quadrants were quarter-circles made of wood or
brass, with a plumb-bob suspended from the apex. The arc was graduated
from 0 degrees to 90 degrees and fitted with two sighting vanes along one
edge. The plumb-bob meant that they were not practical for shipboard use
and pilots went ashore to take observations. The astrolabe does not need a
plumb-bob. It is a perforated disc of brass suspended from a ring, made
heavy to counteract disturbance by the movement of the ship and the wind.
It was usually graduated in the two upper quadrants from 0 degrees to 90
degrees and fitted with an alidade consisting of a centrally pivoting bar with
two sighting vanes, each with a small and large sighting hole. The cross-staff
Pilotage and navigation 181
Bilge, turn of. The transition between the bottom and the side of a hull.
Bowline. Line from the forward edge of a square sail, used to keep it taut
when sailing to windward. May be led to the bowsprit.
Chain wale or channel. A timber fastened on the outside of the hull of a ship
amidships to provide a secure attachment point for the chains at the lower
end of the shrouds.
Clench nail (or clinker nail) (commonly 'rivet'). Nail with large head, used
for fastening clinker planking, clenched by hammering the end over a rove.
Cog. Type-name for a medieval ship built with a flush-laid flat bottom,
straight stems and clinker-built sides.
Deadwood. Solid timbering at the stern of a ship between the keel and the
stern post.
Draft. The vertical distance from the waterline of a vessel to the bottom of
the keel.
Fender. A fitting or moveable object for preventing chafing when ships are
lying against each other or against a quay.
Floor. The first and lowest transverse framing element, which crosses the
keel.
Glossary 185
Forefoot. The area of the junction between the stem and the fore-end of the
keel.
Futtock. A framing timber which does not cross the keel, nor reach the
sheer.
Hawsehole. Hole in the bows of a vessel through which the anchor cable
passes.
Hogged. The condition of a vessel in which the bow and stern have
drooped.
Hulk. Type-name for a medieval ship with a rounded hull-form and strakes
terminating at the sheer rather than on the stem or stern posts.
Joggle. A step cut on the underside of framing to allow a close fit to the
clinker planking.
Keel. Central longitudinal member, normally the lowest member in the hull,
scarfed to the stem and stern posts or to intermediate timbers.
Keelson. Longitudinal member fitted over the floors above the keel to
increase strength and distribute stress.
Lands. Those parts of the surfaces of clinker strakes in contact with adjacent
strakes.
Leeway. The effect of wind deflecting a vessel sideways from the course
steered.
Limber hole. A
notch aligned fore-and-aft cut on the underside of a floor to
allow the passage of bilge water.
Luff. 1. The leading edge of a sail. 2. A fitting for controlling the leading
edge of a sail. 3. To turn the bow of a vessel closer towards the wind.
Mast crutch. A Y-shaped support for the mast when it is unstepped and laid
along the boat.
Mast step. A fitting used to locate the heel of the mast, sometimes integral
with the keelson.
Mizzen. The mast and sail at the after end of a ship. Confusingly, the Italian
mezzana and French misaine both refer to the foremast.
Mortice. A hole cut into a timber to receive the shaped end of another
timber or a free tenon.
Oar. A pole with a flattened blade, used as a lever to pull a vessel through
the water.The part between the blade and the oar pivot is termed the shaft
and the inboard part is termed the loom.
Paddle. A pole with a blade used without a pivot to pull a boat through the
water.
Pintle. A metal pin fastened onto the rudder, which fits into a gudgeon.
Port side. That side of a vessel which is on one's left hand side when facing
the bow.
Rabbet (or rebate). A groove or slot cut into a timber to receive the edge or
end of another timber.
Reef. To reduce the sail area by taking in or rolling up a part and securing
it.
188 Medieval ships and shipping
Reef point. A short length of thin line attached to a sail, used to secure the
sail when reefed.
Ribband. A long flexible length of timber (in one or more pieces) used in
carvel-building, attached to the stem and stern posts and the principal frames
and helping to define the shape of the hull.
Rove. A thin piece of metal, usually quadrilateral, forced over the point of a
clinker nail before it is clenched.
Rudder stock. The upper portion of a rudder, to the head of which a tiller is
attached.
Scarf. A joint between the ends of two timbers of similar section, uniting
them into a continuous piece.
Seam. The narrow gaps between strakes in a hull and planks in a deck,
which need to be made watertight.
Shaft. The portion of an oar between the pivot and the blade.
Sheet. A rope fastened to a lower corner of a sail and used to control the
plane of the sail in relation to the wind direction.
Glossary 189
Ship. A large seagoing vessel, superior in size, complexity and status to those
contemporary vessels which were referred to as boats.
Shroud. A
rope leading from the masthead to the side of the vessel to give
transverse support to the mast. In later medieval ships chains, attached to the
shrouds by means of blocks called deadeyes, linked the lower end of the
shrouds to the hull.
Side timber. A framing element fitted to the planking in the area between the
bilge and the sheer. May be isolated or part of a continuous frame.
Skeg. Additional timber(s) fastened under the stern to act as a partial keel.
Starboard side. That side of a vessel which is on one's right hand side when
facing the bow.
Stem. 1. The timber which closes the hull of a vessel at one or both ends,
jointed to the central longitudinal member and onto which strakes are
fastened. 2. The forward end of a vessel, as opposed to the stern.
Stern post. Main structural element at the aft end of a vessel, joined to the
central longitudinal member and onto which strakes or the transom are
fastened.
Strake. A run of planking, normally from one end of the vessel to the other,
commonly made up of several planks scarfed together. Strakes are numbered
from the keel upward.
190 Medieval ships and shipping
Stringer. A member running fore-and-aft along the inboard face of the hull
to increase longitudinal strength.
Sweep. A large oar, generally used with both hands in a large vessel which
would not normally be rowed.
Tenon. Either a projection from a timber cut to fit into a mortice, or a free
tenon which joins two timbers by fitting into a mortice in each of them.
Thole. A vertical wooden pin projecting above the sheer and serving as an
oar-pivot.
Tiller. A bar, one end of which is fitted into the rudder stock head and
which is used to turn the rudder.
Wooldings. Rope bindings used to hold timbers together, as for example the
mast or the overlapping ends of timbers forming a yard.
parts of a composite
A list of ships, boats and their fittings from the period 1050 to 1500 found
in Britain and the Channel Islands
Key:
5. Gilbert, 1964.
KENTMERE, Cumbria
1. Extended logboat, 4.25 metres long, with 5 strakes on each side.
2. 650 ± 120 bp (c.1300 ad) D-71 (radiocarbon).
3. Found in 1955 in the bed of a former lake, during mineral extraction.
4. The logboat with the greatest number of added strakes yet found in
Europe.
5. Wilson, 1966; McGrail, 1974.
LANEHAM, Nottinghamshire.
1. Small boat with oak clinker planking fastened by iron nails without
roves, caulked with sheep's wool.
2. 480 ± 70 bp (c.1470) HAR-5021 (radiocarbon).
3. Exposed in the bank of the River Trent in 1982 and 10 to 12 years
previously; only partly uncovered.
4. Needs further investigation.
5. Unpublished.
LONDON
Note: A corpus of boatfinds from archaeological excavations in London has
been prepared by Peter Marsden and is expected to be published in 1994 or
1995
-: BILLINGSGATE
1. Clinker planking fastened with wooden pegs, reused in waterfront.
2. 12th century (stratigraphy).
3. Archaeological excavation in 1982.
194 Medieval ships and shipping
4.
5. Youngs et ai, 1983: 192.
-: BLACKFRIARS 3
1. 16 metres of a large, flat-bottomed river vessel, including keel and
keelson, one stem, strakes and framing.
2. 15th century (contents).
3. Archaeological excavation in 1976.
4. Most complete 15th-century found in Britain.
vessel so far
5. Marsden, 1977: 130-32; 1979: 87-91; 1981: 10-16.
-: BLACKFRIARS 4
1. Clinker wreck.
2. 15th century, because of similarities to Blackfriars 3.
3. Seen in the side of a deep hole in 1970; not excavated.
4. Sunk with cargo of Kentish building stone.
5. Marsden, 1979: 91.
-: BLACKFRIARS 5
1. Frame timber reused in waterfront revetment.
2. 15th century (context).
3. 1970.
4.
5. Marsden, 1979: 83.
-: BRIDEWELL
1. Planking.
2. 15th century - stratified under Henry VIII's palace.
3.
4.
5. Marsden, 1979: 83.
-: CUSTOM HOUSE
1 . Planking from both sides of a vessel and framing, reused in waterfront
revetment.
2. 13th century (context).
3. Archaeological excavation in 1973.
4.
5. Marsden, 1979: 86-7; Webster and Cherry, 1974: 202.
MELTON, Humberside
1. Framing timber with joggles for four strakes and treenails.
2. Undated.
3. Found on north Humber foreshore in about 1982.
4. Stray find reported to Hull City Museum.
5. Unpublished.
RrVER MERSEY
1. Nine logboats.
2. 9th to 12th century ad (radiocarbon).
3. Found between 1889 and 1971.
4. Group exhibits some common characteristics.
5. McGrail and Switsur, 1979.
NEWPORT, Gwent
1. Part of a clinker vessel.
2. 1000 ± 80 bp (c.950 ad) HAR-3203 (radiocarbon).
3. Found in 1878 by workmen digging a timber pond at Newport Dock.
4. No drawings of the vessel and only a small fragment has been preserved.
5. Hutchinson, 1984; Morgan, 1878.
5. Unpublished.
SANDWICH, Kent
1. Remains of a ship perhaps originally 33 metres long; parts of rudder,
frames and planking lifted.
2. 15th century (topography).
3. During pipe laying in 1973.
Catalogue 197
2. Undated.
3. Found in 1848 'near American Wharf.
4. No records or remains.
5. Prynne, 1973: 229.
SOUTHWOLD, Suffolk
1. Two side rudders: I is 3.91 metres long; II is 4.36 metres long.
2. I 1080 ± 90 bp (c.870 ad); II 1020 ± 90 bp (c.930 ad) (radiocarbon).
3. I was dredged up by a fisherman in about 1980; II was found on the
beach in 1986, after a storm.
4. Rare items, unusual form of attachment.
5. Hutchinson, 1986.
WALBERSWICK, Suffolk
1. Two frame timbers with joggles for clinker planks.
198 Medieval ships and shipping
2. Undated.
3. Washed up on beach.
4. Stray finds reported to Suffolk County Archaeologist.
5. Unpublished.
WEYBRIDGE, Surrey
1. Small boat fastened with iron clench nails.
2. 410 ±
60 (c.1540 ad) HAR-4996 (radiocarbon).
3. 1931 partly excavated from riverbank.
4. Possibly late medieval; some remains preserved in Weybridge Museum;
potential for excavating rest of boat.
5. Unpublished.
WOOLWICH, Kent
1. Remains of a large ship, at least 120 feet (36 metres) long and 45 feet
(13.5 metres) in the beam.
2. 16th century or earlier (topography).
3. Found in 1912 during construction of a power station.
4. Frames showed that ship had been originally been clinker-built, then
rebuilt with flush-laid planking; ship may be the Sovereign, built 1488
and rebuilt 1509.
5. Anderson, 1959; Philp and Garrod, 1983; Salisbury, 1961.
Sjohistoriska Samfundet.
Allan, J.P. 1984. Medieval and Post-Medieval Finds from Exeter, 1971-1980. Exeter,
Exeter City Council and University of Exeter.
Andersen, Erik 1986. 'Steering experience with square-rigged vessels' in Crumlin-
Pedersen, O. and Vinner, M. (eds) 1986, 208-18.
Andersen, P.K. 1983. Kollerupkoggen. Thisted.
Anderson, R.C. 1928. 'English galleys in 1295' Mariner's Mirror 14, 220-41.
Anderson, R.C. 1934. 'The Burseldon ship' Mariner's Mirror 20, 158-170.
Anderson, R.C. 1938, 'The Burseldon ship' Mariner's Mirror 24, 112-3
Anderson, R.C. 1945. 'Jal's "Memoire No. 5" and the manuscript "Fabbricca di
Galere'" Mariner's Mirror 31, 160-67.
Anderson, R.C. 1959. 'The story of the Woolwich ship' Mariner's Mirror 45, 94-9.
Anderson, R.C. 1976. Oared Fighting Ships. Kings Langley, Argus.
Anderson, Romola and R.C. 1926. The Sailing Ship. London, Harrap.
Archibald, CD. 1844. 'Ancient cannon' Nautical Magazine 13, 653-7, 744-54.
Arenhold, L. 1911. 'Ships earlier than 1500 AD' Mariner's Mirror 1, 298-301.
Aston, Michael (ed.) 1988. Medieval Fish, Fisheries and Fishponds in England.
Oxford, BAR.
Aston, M. and Dennison, E. 1988. 'Fishponds in Somerset' in Aston, Michael (ed.)
1988, 391-403.
Ayres, B.S. 1981. 'Hull', in Milne, G. and Hobley, B. (eds) 1981, 126-9.
Balard, Michel 1991. 'La "Revolution Nautique" a Genes de XHIe-debut du
(fin
Bridbury, A.R. 1955. England and the Salt Trade in the Later Middle Ages. Oxford,
Clarendon.
Brindley, H.H. 1912. 'Reefing gear' Mariner's Mirror 2, 129-34.
Brindley, H.H. 1927. 'Mediaeval rudders' Mariner's Mirror 13, 85-8.
Brindley, H.H. 1938. Catalogue of Seals in the National Maritime Museum.
Greenwich, National Maritime Museum.
Brooks, F.W. 1928. 'Naval armament in the 13th century' Mariner's Mirror 14,
115-31.
Brooks, F.W. 1929a. 'The King's ships and galleys mainly under John and Henry III'
Christensen, A.E. 1987. 'A mediaeval ship model' Int J Naut Archaeol 16.1, 69-70.
Christensen, Arne Emil 1989. 'Hanseatic and nordic ships in medieval trade. Were the
cogs better vessels?' in Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1989,
17-23.
Clarke, Helen 1979. 'The archaeology, history and architecture of the medieval ports
of the east coast of England, with special reference to King's Lynn, Norfolk' in
McGrail, S. (ed.) 1979, 155-65.
Bibliography 201
Clarke, Helen 1981. 'The medieval waterfront of King's Lynn' in Milne, G. and
Hobley, B. (eds) 1981, 132-5.
Clarke, Helen 1983. 'The historical background to North Sea trade, c.1200-1500' in
Davey, P. and Hodges, R. (eds) 1983, 17-25.
Clarke, Helen 1984. The Archaeology of Medieval England. London, British
Museum.
Clarke, H. and Carter, A. 1977. Excavations in King's Lynn 1963-1970. Soc
Medieval Archaeol Monograph Series, 7.
Clarke, R., Dean, M., Hutchinson, G., McGrail, S. and Squirrell, J. 1993. 'Recent
work on the R. Hamble wreck near Burseldon, Hampshire' lnt J Naut Archaeol
22.1, 21-44.
Clowes, W. Laird 1897. The Royal Navy: a history vol. 1. London, S. Low, Marston
& Co.
Coad, J.G. and Streeten, A.D.F. 1982. 'Excavations at Castle Acre Castle, Norfolk,
1972-77' Archaeol J 139, 138-301.
Cobb, Henry S. 1961. The Local Port Book of Southampton for 1439-40.
Southampton, Southampton University,
de Courcy Ireland, J. 1989a. 'Irish maritime trade and Irish ships at the close of the
Middle Ages' in Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1989,
115-120.
de Courcy Ireland, J. 1989b. 'A survey of early Irish maritime trade and ships' in
McCaughan, M. and Appleby, J. The Irish Sea: aspects of maritime history 21-5.
Belfast, Queen's University and Ulster Folk and Transport Museum.
Crumlin-Pedersen, Ole 1972. 'The Vikings and the Hanseatic Merchants' in Bass, G.
(ed.) 1972, 182-204.
Crumlin-Pedersen, Ole 1976. '"Guldskibbet" ved Vejby Strand' Skalk 6, 9-15.
Crumlin-Pedersen, Ole 1979. 'Danish cog finds' in McGrail, S. (ed.) 1979, 17-34.
Crumlin-Pedersen, Ole 1985. 'Wrecks as a source for ships and sea routes' in
Haarmann, J. (ed.) 1985, 67-73.
Crumlin-Pedersen, Ole 1989. 'Wood technology and forest resources in the light of
medieval shipfinds' in Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1989,
25-42.
Crumlin-Pedersen, Ole (ed.) 1991. Aspects of Maritime Scandinavia AD 200-1200.
Roskilde, Viking Ship Museum.
Crumlin-Pedersen, Ole 1991. 'Ship types and sizes AD 800-1400' in Crumlin-
Pedersen, Ole (ed.) 1991, 69-82.
Crumlin-Pedersen, Ole and Vinner, Max (eds) 1986. Sailing into the Past. Roskilde,
Viking Ship Museum.
Culver, Henry B. 1929. 'A contemporary 15th century ship model' Mariner's Mirror
15, 213-21.
Cutting, Charles L. 1955. Fish Saving: a history of fish processing from ancient to
modern times. London, Leonard Hill.
Daniels, R.1991. 'Medieval Hartlepool: evidence of and from the waterfront' in
Good, G.L., Jones, R.H. and Ponsford, M.W. (eds) 1991, 43-50.
Darby, H.C. 1940. The Draining of the Fens, Cambridge, Cambridge University
Press.
Darby, H.C. (ed.) 1948. An Historical Geography of England before AD 1800.
Cambridge, Cambridge University.
Davey, P. and Hodges, R. (eds) 1983. Ceramics and trade: the production and
distribution of later medieval pottery in north-west Europe. Sheffield, University of
Sheffield.
Davis, J.E. 1867. Notes on Deep-Sea Sounding. London, Admiralty Hydrographic
Office.
202 Medieval ships and shipping
Ellmers, Detlev 1985b. 'The cog of Bremen and related finds' in Haarmann, J. (ed.)
1985, 188-203.
Ellmers, Detlev 1989. 'Development and usage of harbour cranes' in Villain-Gandossi,
C, Busuttil, S. and Adam, P. (eds) 1989, 43-69.
Ewe, Herbert 1972. Schiffe auf Seigeln. Rostock, VEB Hinstorff.
Farrell, A.W. 1979. 'The use of iconographic material in medieval ship archaeology'
in McGrail, S. (ed.) 1989, 227-46.
Fenton, Alexander 1978. The Northern Isles: Orkney and Shetland. Edinburgh, John
Donald.
Fenwick, Valerie (ed.) 1978. The Graveney Boat. Oxford, BAR.
Filgueiras, Octavio Lixa 1991. 'Gelmirez and the reconversion of the W. Peninsular
shipbuilding tradition (Xlth-XIIth centuries)' in Reinders, R. and Paul, K. (eds)
1991, 32-41.
Fliedner, S. 1964. Die Bremer Kogge. Bremen, Focke Museums.
Fox, R. 1981. 'Portsmouth' in Milne, G. and Hobley, B. (eds) 1981, 147-8.
Frake, Charles O. 1985. 'Cognitive maps of time and tide among medieval seafarers'
Man (NS) 20, 254-70.
Friel, I. 1983a. 'England and the advent of the three-masted ship' in Proceedings of
the 4th International Congress of Maritime Museums, 1981 130-138. Paris,
Musees de Marine.
la
Friel, Ian 1983b. 'Documentary sources and the medieval ship: some aspects of the
Greenhill, B. 1980. 'Vessel of the Baltic -The Hansa cog and the Viking tradition'
Country Life July 31, 402-4.
Grenier, Robert 1988. 'Basque whalers in the New World: the Red Bay wrecks' in
Bass, G. (ed.) Ships and Shipwrecks of the Americas 69-84. London, Thames and
Hudson.
de Groot, P.H. 1984. 'Mataro Model van Catalaanse Nao: analyse van
draagvermogen en stabiliteit' BvS Rapport 966. Rotterdam.
Guerout, M., Rieth, E. and Gassend, J-M. 1989. 'Le navire Genois de Villefranche:
un naufrage de 1516?' Archaeonautica 9. Paris.
Guilleux la Roerie, L. 1957. 'More about the ship of the Renaissance' Mariner's
Mirror 43, 179-93.
Haarmann, Joachim (ed.) 1985. 5th International Congress of Maritime Museums
Proceedings 1984. Hamburg.
Hague, Douglas B. and Christie, Rosemary 1975. Lighthouses: their architecture,
history and archaeology. Llandysul, Gomer Press.
Hall, R.A. 1991. 'The waterfronts of York' in Good, G.L., Jones, R.H. and Ponsford,
M.W. 1991, 177-84.
(eds)
Halliday, F.E. 1953. Richard Carew of Anthony: the Survey of Cornwall. London,
Andrew Melrose.
Hattendorf, John B., Knight, R.J.B., Pearsall, A.W.H., Rodger, N.A.M. and Till,
Geoffrey (eds) 1993. British Naval Documents 1204-1960. Scolar Press. Navy
Records Society, vol. 131.
Heath, P. 1969. 'North Sea Fishing in the Fifteenth Century: The Scarborough Fleet'
Northern History 3-4, 53-69.
Heinsius, P. 1956. Das Schiff der Hanstschen Fruhzeit. Weimar.
Herteig, Asbjorn 1959. 'The excavation of "Bryggen", the old Hanseatic wharf in
Bergen' Medieval Archaeology 3, 177-186.
Herteig, A.E. (ed.) Conference on Waterfront Archaeology in NorthEuropean Towns
No. 2, Bergen 1983. Bergen, Historisk Museum.
Hill, David 1981. An Atlas of Anglo-Saxon England. Oxford, Basil Blackwell.
Hindle, Brian Paul 1976. 'The road network of medieval England and Wales' Journal
of Historical Geography 2.3, 207-21.
Hobley, B. 1981. 'The London waterfront - the exception or the rule?' in Milne, G.
and Hobley, B. 1981, 1-9.
Hobley, B. and Schofield, J. 1977. 'Excavations in the City of London: first interim
report' Antiq J 56, 31-66.
Homer, Ronald F. 1991. 'Tin, lead and pewter' in Blair, J. and Ramsey, N. (eds)
1991, 57-80.
Horsey, LP. 1991. 'Poole: the medieval waterfont and its usage' in Good, G.L., Jones,
R.H. and Ponsford, M.W. (eds) 1991, 51-3.
Howard, Frank 1979. Sailing Ships of War, 1400-1860. Greenwich, Conway
Maritime Press.
Hull, P.L. 1971.The Caption of Seisin of the Duchy of Cornwall. Torquay, Devon
and Cornwall Record Society, New Series 17.
Hurst, D. Gillian 1969. 'Post-Medieval Britain in 1968' Post Med Archaeol 3,
193-212.
Hurst, D. Gillian 1970. 'Post-Medieval Britain in 1969' Posr Med Archaeol 4,
174-88.
Hurst, John 1983. 'The trade in medieval pottery around the North Sea' in Davey, P.
and Hodges, R. 1983, 257-60.
(eds)
Hurst, J.G., Neal, D.S. and van Beuningen, H.J.E. (eds) 1986. Pottery Produced and
Traded in North-West Europe 1350-1650. Rotterdam papers VI.
Hutchinson, Gillian 1983. 'Boatfind at the Caldecotte Lake site' Archaeology in
204 Medieval ships and shipping
Marsden, Peter 1979. 'The medieval ships of London' in McGrail, S. (ed.) 1979,
83-92.
Marsden, P. 1981. 'Early shipping and the waterfronts of London' in Milne, G. and
Hobley, B. (eds) 1981, 10-16.
Marsden, Peter 1992. 'Roman and medieval shipping of south east England' in
Carver, M., Heal, V. and Sutcliffe, R. (eds) 1992, 125-30.
Martin-Bueno, Manuel 1992. 'A 15th century Aragonese ship in Sardinian waters' in
Carver, M., Heal, V. and Sutcliffe, R. (eds) 1992, 55-60.
Martin-Bueno, Manuel 1993. La nave de Cavoli y la arqueologia subacuatica en
Cerdena. Zaragoza, Universidad de Zaragosa.
206 Medieval ships and shipping
van der Merwe, P. 1983. 'Towards a three-masted ship' in Proceedings of the 4th
International Congress of Maritime Museums 1981 121-9. Paris, Musees de la
Marine.
Miller, Louise 1977. 'New Fresh Wharf: 2, The Saxon and early medieval waterfonts'
London Archaeologist 47-53.
3.2,
Milne, Gustav 1982. 'Recording timberwork on the London waterfront' in McGrail,
S. (ed.) 1982, 7-23.
Olsen, Olaf and Crumlin-Pedersen, Ole 1978. Five Viking ships from Roskilde Fjord.
Copenhagen, The National Museum.
O'Neill, T.J. 1987. Merchants and Mariners in Medieval Ireland. Dublin, Irish
Academic Press.
Oosting, Rob 1987. 'De opgraving van het vlak van een kogge bij Rutten' in
Reinders, R. (ed.) Raakvlakken tussen scheepsarcheologie, maritieme geschiedenis
en scheepsbouwkunde. Lelystad, Rijksdienst voor de IJsselmeerpolders,
Flevobericht 280.
Oppenheim, M. 1896. Naval accounts and inventories of the reign of Henry VII.
Navy Records Society.
Oppenheim, M.M. 1968. The Maritime History of Devon. Exeter, University of
Exeter.
Osier, A. and Barrow, A. 1993. Tall Ships Two Rivers: Six centuries of sail on the
Rivers Tyne and Wear. Newcastle upon Tyne, Keepdate.
Palou i Miquel, H., PujolHamelink, M., Raurich Santalo, J. and Reith, E. 1992.
i i
'Two medieval shipwrecks in Catalonia: Culip VI and Sorres X' in Carver, M.,
Reinders, R. 1979. 'Medieval ships: recent finds in the Netherlands' in McGrail (ed.)
1979, 35-44.
Reinders, R. 1982. Shipwrecks of the Zuiderzee. Lelystad, Rijksdienst voor de
IJsselmeerpolders, Flevobericht 197.
Reinders, R. 1985. Cog finds from the IJsselmeerpolders. Lelystad, Rijksdienst voor
de IJsselmeerpolders, Flevobericht 248.
Reinders, Reinder and Paul, Kees (eds) 1991. Carvel Construction Technique. Fifth
International Symposium on Boat and Ship Archaeology, Amsterdam 1988.
Oxford, Oxbow.
Rice, William McPherson 1824. 'Account of an ancient vesel recently found under the
old bed of the River Rother, in Kent' Archaeologia 20, 553-65.
Richardson, Katherine M. 1959. 'Excavations in Hungate, York' Archaeol J 116,
51-114.
Rieck, Flemming 1991. 'Aspects of coastal defence in Denmark' in Crumlin-Pedersen,
Ole (ed.) 1991, 83-96.
Rieth, Eric 1989. 'Le clos des galees de Rouen, lieu de construction navale a clin et a
carvel (1293-1419)' in Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1989,
71-77.
Rieth, Eric 1991a. 'L'epave du debut de XVIeme siecle de Villefranche-sur-Mer
(France): un premier bilan de l'etude architecturale' in Reinders, R. and Paul, K.
(eds) 1991, 47-55.
Rieth, Eric 1991b. 'Systeme de conception des navires medievaux de la Mediterranee.
Un long Moyen Age' in Villain-Gandossi, C., Busuttil, S. and Adam, P. (eds) 1991,
67-77.
Rogerson, A. 1976. 'Excavations on Fullers Hill, Great Yarmouth' East Anglian
Archaeology Report No. 2 Norfolk 131-245. Norfolk Archaeological Unit.
Rose, Susan 1982. The Navy of the Lancastrian Kings Navy Records Society.
Ruddock, A. A. 1942. 'The method of handling the cargoes of medieval merchant
galleys' Bull Inst Historical Research 19, 140-48.
Rule, Margaret 1982. The Mary Rose. London, Conway Maritime.
Runyan, Timothy J. 1991. 'The relationship of northern and southern seafaring
traditions in late medieval Europe' in Villain-Gandossi, C., Busuttil, S. and Adam,
P. (eds) 1991, 197-209.
Ryder, M.L. 1969. 'Remains of fish and other aquatic animals' in Brothwell, D. and
Higgs, E. Science in Archaeology: a survey of progress and research (2nd edn)
376-94.
Salisbury, C.R. 1991. 'Primitive British fishweirs' in Good, G.L., Jones, R.H. and
Ponsford, M.W. (eds) 1991, 76-87.
Salisbury, W. 1961. 'The Woolwich ship' Mariner's Mirror 47, 81-90.
Salzman, Louis F. 1923. English Industries of the Middle Ages (2nd edn). Oxford,
Clarendon.
Salzman, L.F. 1931. E.nglish Trade in the Middle Ages. Oxford, Clarendon.
Salzman, L.F. 1952. Building in England down to 1S40. Oxford, Clarendon.
Saul, A. 1979. 'Great Yarmouth and the Hundred Years War in the fourteenth
century' Bull Inst Historical Research 52, 105-15.
Saul, A. 1981. 'The Herring Industry at Great Yarmouth, C.1280-C.1400' Norfolk
Archaeol 38. i, 33-43.
Saunders, A.D. 1976. 'The defences of Southampton in the later Middle Ages' in
Burgess, L.A. (ed.) The Southampton Terrier of 1454 20-31. London, HMSO.
Saunders, Andrew 1989. Fortress Britain. Liphook, Beaufort.
Saunders, Andrew 1993. Tynemouth Priory and Castle. London, English Heritage.
Scammell, G.V. 1981. The World Encompassed: the first European maritime empires
c. 800-1 650. London, Methuen.
Bibliography 209
Scandurra, Enrico 1972. 'The maritime republics: medieval and Renaissance ships in
Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1989. Medieval Ships and the
Birth of Technological Societies. Vol. I: Northern Europe. Malta, Foundation for
International Studies.
Villain-Gandossi, C, Busuttil, S. and Adam, P. (eds) 1991. Medieval Ships and the
Birth of Technological Societies. Vol. II: The Mediterranean area and European
integration. Malta, Said International.
Vlek, Robert 1987. The Medieval Utrecht Boat. Oxford, BAR.
Walker, Philip 1982. 'The tools available to the medieval woodworker' in McGrail, S.
1948, 266-97.
Wilson, David M. 1966. 'A medieval boat from Kentmere, Westmorland' Medieval
Archaeol 10, 81-8.
Wilson, D.M. and Hurst, J.G. 1957. 'Medieval Britain in 1956' Medieval Archaeol 1,
147-71.
Winter, Hcinrich 1956. Die Katalanische Nao von 1450. Magdeburg, Robert Loef.
Young, G.A.B. 1987. 'Excavations at Southgate, Hartlepool, Cleveland, 1981-2'
Durham Archaeol J 3, 15-55.
Youngs, S.M., Clark, J. and Barry, T. 1983. 'Medieval Britain and Ireland in 1982'
Medieval Archaeol 27, 161-229.
Youngs, S.M., Clark, J., Gaimster, D.R.M. and Barry, T. 1988. 'Medieval Britain and
Ireland in 1987' Medieval Archaeol 32, 225-314.
Zupko, R.E. 1977. British Weights and Measures. Wisconsin, University of Wisconsin
Press.
Index
References to illustrations are shown in italic, and are to pages, not figure
numbers.
Southampton, 57, 117, 144, 152, 163 timber, 21, 32, 57, 72, 139, 154
galley (1294), 23, 24, 48, 151 see also named species
port, 112, 114, 114, 115, 115, 150, 162 store at Poole 25
seal, 10, 15, 27 in waterfront structures, 107-9
trade, 78, 80-1, 84-5, 88, 93, 98 tolls, 112, 171
Southchurch Hall boat, 123, 125 tops (crow's nests), 166
Southend, Essex, 123 Topsham, 104
Southgate Area A, Hartlepool, 110 Touraine, 77
Southwold, Suffolk, 144 Tournai, 11, 12, 50
rudders 51, 52 Tower of London, 125, 163
Sovereign (1488), 46, 56, 159, 160 trade, 88-103
Spain, 32, 43, 79, 169 see also cargoes; specific commodities
trade, 75, 87, 129 79-81
Atlantic Iberia,
spears, 130, 139 Baltic, 72-5
speed, 168, 169, 180 Denmark, 72
spurdog, 131 France, 73, 77-9
stability, 161 Hanse, 72-5
Stanley Ferry, Yorkshire, 122 Ireland, 65
Staveren, 72, 75 Italy, 84-7
stays, 56 Low Countries, 75-7
Steelyard, King's Lynn, 115 Mediterranean Spain, 81-4
Steelyard, London, 113, 115 Norway, 69-70
steering, 50-5, 61, 118 transports, 149, 150, 154
stern castles see castles, shipboard traverse boards, 179-80
stockfish, 143 treadwheel, 112
Stonar, 104 trebuchets, 146, 163
stone, 96, 119, 121, 121 Trent, River, 117, 122, 139
stone shot, 159, 161, 162 Trig Lane, London, 105, 130
stone-throwing engines, 146 Trimte de la Toure (1398-1413), 156
stowage, 89, 93, 102 tunnage, 92-3
Stralsund, 72 tuns, 89-93
Studland Bay wreck, 49, 81, 90, 95, 161 turbot, 131
sturgeon, 129 turf boat, 127
styves, 112 Turkey, 36
178
Suffolk, 44, 51, 137, 144, 171, two-masted ships, 43, 60-1
Sun and Topping's Wharf, London, 110-11 Tyne, River, keels, 5, 119, 120, 127
sundials, 178 Tyne and Wear, 137
superstructure, 47 Tynemouth Priory, 171
Sussex, 137 Tywn Llewelyn, Glamorgan, 170
Sutton Hoo ship, 4
Sverri, King,69 Ulster, Earl of (1296-1328), 154
Sweden, 15, 73 Ushant, 165
Swyn, River, 147, 149 Utrecht, 12, 13
wadmal, 96 whipstaff, 55
Walberswick, 144 whiting, 130, 144
Wales, 67, 87, 136, 151 William I, the Conqueror, King of England,
Walney, Morecambe Bay, 162 150
Walter de Godeton, 171 Winchelsea, 150, 162, 163
Wantage, 130 galley (1347), 151
Wantsum Channel, 171 town seal, 48, 48, 80, 147, 153
warehouses, 113-15 winches, 130
warships, 5, 146, 147, 148, 149-50 Winchester, 11, 12, 50
Warwick Roll, 156 windlasses, 47-9, 112
Waterford, 67, 171 wine trade, 65, 78, 85, 87, 89, 97-8, 114
waterfronts, 105, 107-8, 110, 150 wippe, 112
Watergate, London, 111 Wisbech, 104
Watergate, Southampton, 115 Wismar, 72
waterschip, 142 Wittlesey Mere, 121
watertightness, 30, 46, 47 Wood Quay, Dublin, 65
Waverley, 144 Woodstock, 121
weapons, 146, 151, 156, 158, 163 Wool Quay, London, 114
see also artillery; guns wool trade, 84-5, 89, 113
weather, 117, 166, 168 wooldings, 56
weaver knot, 133 Woolwich ship, 46, 56, 159-60
weighing beams, 113
weights, 137-9, 138 yards, 55, 56
Welland, River, 121 Yarmouth, 143, 147, 153
Weser, River, 16 Yassi Ada ship, 36
West Hithe, Southampton, 115 York, 107, 110, 112, 117, 179
Wexford, 65, 69 galley (1294), 23, 24, 151
whale, 129 Yorkshire, 122
wharfage, 112 Youghal, 67
wharves, 105, 112
whelks, 129 Zedelghem, Belgium, 11, 50
whetstones, 139 Zeeland, 143