|
The basic elements of ancient Irish dress for people in the upper classes were the léine and the brat.
These lasted, with variations over time, from the earliest recorded times down to the 16th century.
There is some speculation that the adoption of the léine is due to Roman influence;
but I think it is more likely that the Romans adopted the more fitted clothing of the 'barbarians' toward the later years of the Roman Empire
(as those 'barbarians' gained more power in the Empire).
Tunic (Léine)
The léine (pronounced /lay'-nuh/) in early Ireland in early depictions (between 5th and 12th century CE)
is a long smock-like garment made of linen, not too widely cut, reaching to slightly above the ankles and decorated around the neck, wrists,
and lower hem with embroidery.
McClintock says it resembles a djelabbeh. It might have sleeves or be sleeveless.
The léine can be drawn up through the belt to knee-level
(which causes it to bunch in such a way that carvings of men wearing their léines this way are sometimes mistaken for wearing a kilt).
(H.F. McClintock, Old Irish Dress, p. 2)
The léine may sometimes have opened in the front to the waist (see below), but most pictures show a neckline and don't indicate such an opening.
The léine's neckline can be round, square, or v-shaped. Sometimes a léine is described by the term culpatach, meaning hooded;
this could have meant that it had a collar (culpait) large enough to be used as a hood. (McClintock, Old Irish Dress, p. 13).
Both women and men wore the léine, but for
women, it was a little longer. The full-length léine
is nearly always shown being worn with a brat, not by
itself, and is never shown worn with trews or the inar.
The léine can sometimes be shorter than
ankle-length; a shorter léine, however, seems to
be a mark of lower status, as the wearer probably is
involved in physical labor. Some effort was made to
assure that the léine wasn't too short.
(Dunleavy, p 17)
Laborers or peasants are sometimes seen in what superficially appears to be a short kilt, which has some
embroidery around the lower hem. However, this most likely represents a léine, with the upper part
thrown off to allow for coolness and freedom of movement while working. This would indicate that the neck-line of
the léine is big enough to allow the wearer to put his whole body through it, so that it hangs around
the waist. One figure on the cross shows an opening big enough to do this. (Dunleavy, p. 4)
The léine as seen in the Book of Kells has a high neckline, too narrow for the wearer to throw off the
top of the garment for work. Sleeves are narrow and close to the arm. The long, flowing sleeves of léinte
from the 16th century are a later development.
Decoration
The léine is usually described as being gel, or bright. This probably
indicates light-colored linen. Some of the léinte shown in the Book of Kells are of various colors, including
light blue or green, which are obtainable with woad, with an under-dye of weld for the green. Linen doesn't take
dye very well, and most colors applied would come out light, rather than the intense, dark colors we are able
to achieve with modern chemical dyes; the exceptions are the pigments obtained from indigin (from woad) and murex
purple. The Book of Kells seems to indicate embroidery or woven borders at the neck, wrists and hem.
A good description of how to construct a similar tunic
can be found at the following site of How
to Make a Viking Tunic -- look at the Birka tunic with a round neckline and gores let into the side seams.
I'll be posting pictures and instructions eventually.
In the meantime, my information for The Rogart Shirt would probably be fairly accurate
Women (left) and men (right) from the Breac Maedhóc, a bronze house-shrine from the 11th or 12th century:
Cloak (Brat)
The brat (pronounced /braht/) was a rectangular
woolen cloak worn over the shoulders like a shawl and/or
fastened with a brooch on the chest or the right
shoulder. The brat seems most commonly to have
been rectangular, and rather voluminous, so that it could
be folded several times around the wearer, with longer
length indicating greater status. Sometimes the brat
is described as 'five-folded' (Gantz, p. 157), but we
don't know exactly what this means. They are sometimes
portrayed as having some sort of hood, or as being folded
and/or pinned in such a way that part of the brat could
be drawn up over the head as a hood.
Several other forms of the brat seem to have
been used, though it's hard to tell from the pictorial
evidence -- one form seems to have holes through which
one can put one's arms without unfastening the cloak.
Some are shown that look like modern capes -- a
half-circle, with the bottom edge parallel to the ground,
with or without a hood. (Dunleavy, p. 3) Both large and
small mantles are portrayed. The shorter brats,
however, are usually worn with trews. Women are usually
portrayed wearing the full-length brat.
Decoration
Unlike linen, wool takes dye very well, and the brat
is often described as being colored. Usually the brat
is one color with a fringe (corrthar) or border of
another color. These borders or fringes could have been
either woven into the brat, as was common with
fabric woven on a warp-weighted loom, or made separately,
and could include silver and golden threads. It is
possible that embellishments included appliqué and
tapestry-woven patterns. (McClintock, Old Irish Dress,
P. 15) Bright colors were common, with purple, crimson
and green being mentioned most often. Other colors listed
are blue, black, yellow, speckled (which,
from the Latin, can mean checked or
tartan), gray, dun, variegated and striped. (McClintock, Old
Irish Dress, p. 14) McClintock downplays the
possibility of tartans being used, but scraps of checked
cloth have been found from ancient Scotland and elsewhere
in Europe, so a simple check is certainly not impossible.
The brat is also sometimes described as being
'fleecy': the woman who enters Da Derga's Hostel in the
tale of that name is described as wearing a brat
that was fleecy and striped. (Gantz, p. 76) Some brats
from later periods have been found that had a pile woven
into the fabric, so that they looked rather like a rug.
It is also likely that the nap of the fabric was drawn
out with teasels, so that the fabric was very fuzzy; this
fiber could then be either left long or sheared short, so
that it looked like modern woolen blankets. The
depictions in the Book of Kells and other manuscripts,
however, do not show mantles with obvious tufting.
No Kilts!
One of the myths making its way through the Celtic
community is that the Irish used to wear a kilt. There is
no evidence to support this. Several sculptures have been
cited to support the existence of kilts; however, most
authorities (including H.F. McClintock) on the subject
say that the garments portrayed are léinte, gathered
around the waist (see both my comments above in the
segment on léinte, and Scottish Clothing, ca. 1100-1800 AD.
The kilt arose in Scotland around 1600 C.E., when Scots
started belting their brat around their waist. This was
remarked on by observers, who said they could tell the
Scots from the Irish soldiers in Ulster because of this
habit of belting their cloaks.
Below: The disputed
panel from the Cross of Muiredach
Jacket (Inar)
Soldiers are portrayed as wearing a close-fitting
sleeved or sleeveless jacket, waist-length, fastened in
the chest with a brooch. The jacket is worn with a pair
of trews, not over a léine. The high waists of
16th c. jackets seem to be a later development. One
soldier portrayed has sleeves to the middle of his
forearm and trews that come to a few inches below the
knee. Another has sleeves that come to his wrists.
(Dunleavy, pp. 21, 22)
Trews (Bróc)
Trews in Ireland are usually shown on soldiers, who
are wearing them with a short jacket. The trews are
usually close-fitting, sometimes shown to end above the
knees, sometimes to a few inches below the knee, and
sometimes cover the whole leg. They are sometimes marked
with vertical lines which may represent decoration or a
striped weave in the cloth. An illustration in the Book
of Kells shows a soldier wearing a fitted green jacket
with close-fitting sleeves and a round neck, and bright
blue trews. The trews end just above the ankle-bone and
have a strap going under the foot, making them like
modern stirrup pants. There is a line right under the
knees that may represent a garter. (McClintock, Old
Irish Dress, p. 5; Dunleavy, p. 21, 22) This is
corroborated by an account of the arrival of Harald
Gille, who later became king of Norway, who came to
Norway from Ireland claiming to be a son of King Magnus
Barefoot by an Irish mother. His clothes are described
thus: "he had on a shirt and trousers which were
bound with ribands under his foot-soles, a short cloak,
an Irish hat on his head and a spear hat under his
hand." (McClintock, Old Irish Dress, p. 5)
Other forms of trews: on the Cross of Muiredach (10th
c.), soldiers are shown wearing what appear to be striped
trews that are rather short -- they only reach to
mid-thigh at most. (Dunleavy, p. 21)
Soldier from the Book of Kells, ca. 800 CE (jacket is green, trews blue):
Persons who would have worn trews would have included
charioteers, the king's bodyguard, food bearers, door
keepers, and scouts. Kings and other notable persons are
usually shown wearing the long léine.
Belt (Crios)
The crios usually refers to a leather or woven belt.
These are probably either tabby weave (as is the criosana still woven in Aran today), or tablet-woven. It was woven
not only to keep the léine in place but to carry object and utensils in the usual medieval fashion.
Shoes:
Contrary to popular opinion, going shoeless is not a universal Celtic trait. The Rule of Ailbe of Emly
directed that "no matter how ascetic a person became he should never go barefoot." (Dunleavy, p. 20)
See Diarmuit Ui Dhuinn's Footwear of the Middle Ages site for information on Irish shoes for this period.
Colors of Clothing:
Brehon law laid out the colors of the clothing that people were allowed to wear -- see my Textile Page.
Also see:
Social History of Ancient Ireland
(excerpts from P. W. Joyce) -- use with caution. His information on the spurious Irish 'kilt' has been thoroughly refuted.
Molly ni Dana's Home Page - another essay on Irish clothing, and an essay on shoes.
|