University of Virginia Library

V. 2

PREHISTORIC, PROTOHISTORIC
& EARLY MEDIEVAL PROTOTYPES
OF GUEST & SERVICE BUILDINGS
OF THE PLAN OF ST. GALL

V.2.1

LITERARY EVIDENCE

THE NORTH GERMANIC HOUSE OF THE
SAGA PERIOD

In 1889 the Icelandic literary historian and philogist
Valtyr Gudmundsson[50] was able to demonstrate, on the
basis of a careful and painstaking analysis of words and
passages in the Nordic Sagas referring to the layout and
construction of houses, that the Germanic standard house
of the Saga Period (ninth to thirteenth centuries) in Norway,
Sweden, Denmark, and Iceland was a three-aisled
timber structure with an open fireplace (eldr, "fire" or
arinn, "hearth") in the middle of its center aisle (golf);
that this house received its light through an opening in
the roof (ljóri, "light inlet"), which also served as a smoke
outlet (hence also referred to as reykháfr or reykberi, "smoke
hole,") and which could be closed and opened by means of
ropes or poles; that the roof (ráf or ræfr) of this house
was supported by a free-standing inner frame of timber,
composed of two longitudinal rows of uprights (súlur,
stafir, stođir, stolpar, and sometimes more specifically referred
to as innstafir or innstolpar, "inner posts," in contradistinction
to the útstafir, the corresponding "outer" or
"wall posts"), which were connected lengthwise by means
of roof plates (ásar or langvidir, "long beams") and crosswise
by means of tie beams (vagl, vaglbiti or þvertrè,
"crosstree"). Gudmundsson summarized his findings
visually in a plan and a perspective view of the interior of
the Saga house, drawn up for him by E. Rondahl (figs. 284
A-B).[51] He demonstrated that this house type could be used
for a variety of purposes, without changing any of its basic
dispositions. It served, as circumstances demanded, as a
general living house (stofa or stufa), as a dining or festal
hall (drykkjuskáli, veizluskáli), as kitchen or fire house
(elda skáli), as sleeping hall (skáli or hviluskáli), even, as a
hay or cattle barn (hlađa or fjóshlađa). In the early days—
and subsequently in the lower social strata—all these functions
were performed simultaneously under one roof; later,
as increasing wealth and social prestige permitted, they
were progressively relegated to separate buildings.

Gudmundsson could establish that in the general living
house as well as in the festal or banqueting hall the floors
of the aisles (langpallar) and of the cross bay at the upper end
of the house were, in general, raised above the level of the
center floor and covered with wooden planks. Long benches
(langbekkir) and tables (borđ) were set up in the aisles
parallel to the two long walls of the house and also crosswise
along the gable wall at the rear of the house. This raised
section at the innermost part of the house was referred to as
the crossbench (þverpallr).

The chieftain or owner of the house sat on his high seat
(œdra öndvegi, "first seat of honor") in the middle of one
of the two long benches (i miđju bekk) while his principal
guest of honor occupied the second best high seat (úœdra
öndvegi) in the middle of the opposite bench. The women sat
on the crossbench at the rear of the house. The fire
crackled in the middle of the center floor. The entrance lay,
in general, in the center of one of the two gable sides of the
house, and was often separated from the rest of the house by
an entrance hall (forstofa, forskáli) which occupied the foremost
bay of the house, forming a counterpart to the women's
cross bench at the opposite gable. Often this entrance bay
was separated from the main space by a cross partition
(þverpili), which had in its center a second or inner door.
The walls of the Saga house (veggir, or, more specifically,
langveggir, "long wall," and gaflveggir, "gable walls") were,
as a rule, constructed of earth or turf (torfi), and insulated
inside with a wooden paneling (veggþili). The rafters rose
from wall plates (syllr or staflægjur) and converged at the
top in a ridge beam (mœniáss) which was carried by short
king posts (dvergr, "dwarf post") rising from the center of
the tie beams.

From numerous incidental references to the house, made
in the dramatic accounts of battles waged when a householder
and his family were attacked in their sleep and
forced to rise to defend themselves, Gudmundsson could
infer that the sleeping house (skáli) was divided lengthwise,
like the stofa, into a center aisle and two lateral
aisles and received its warmth from an open fire that burned
in the middle of the center floor. As in the stofa, the floors
of the lateral aisles were raised above the level of the center
aisle and covered with wooden planks. But instead of
supporting tables, the aisle floors of the skáli (called set)
were covered with a bedding of straw and subdivided
crosswise into individual sleeping compartments by means
of rugs or "hangings" (sængarklæđi) suspended from cross
beams. Each compartment was sufficiently large to accommodate
two people (sengefeller, "bedfellows") lying parallel
to the walls of the building, one outside (fyrir ofan) or near

[ILLUSTRATION]

284.A THE NORTH GERMANIC HOUSE
OF THE SAGA PERIOD, 9TH-13TH CENTURIES

[after Rosenberg, 1894, 257, fig. 9]

The house is entered through one of its gable walls. Its center floor
(a,a) is made of stamped clay; b,b,b are open fireplaces framed by
stones. The aisles are slightly raised (c,c) and covered with wooden
boards. They accommodate the benches and tables for the men (f,g).
The terminal bay, with the benches and tables for the women (e.g)
is treated in the same manner; h is a table from which food and
drinks are served; i,i are footstools in front of the high seats; j is a
secret door for escape through an underground channel should the
principal entrance be blocked by enemies. The walls are built in a
mixture of earth, rubble and turves.

the wall paneling (vid þili), the other inside (fyrir framan)
or near the sleeper beam (vid stokk, i.e., the floor beam that
forms the edge of the slightly raised level of the aisles). The
bedsteads of the master and his wife were often separated
from the adjoining bedsteads by means of a wooden wall
partition, so as to form a bed closet (rekkja) that could be
locked and was then called a lokhvílu (lockable closet). One
or two further closets of this type were frequently installed
for persons next in rank or for guests of honor.

In like manner the aisles of the cattle barns were subdivided
into individual cross compartments for the stabling
of the livestock.

As one reviews this evidence one cannot fail to be struck
by the amazing similarity of the North-Germanic Saga
house, spatially and functionally, with that of the guest and
service structures of the Plan of St. Gall. Both have as
nucleus an open center space accessible to all, which gives
admittance to a peripheral suite of outer spaces surrounding
the center space on three or all four sides. In both, the
hearth lies in the middle of the center space and has in
the roof above it a shielded opening that serves as a smoke
outlet. In both this layout is used, either separately or
combined, without requiring the slightest alterations of its
basic dispositions, as shelter for the people, as shelter for
their livestock, and as shelter for the harvest.

There are, to be sure, some distinctive differences. The
Saga house has its entrance, in general, in the middle of one
of its gable walls; that of the St. Gall House is, in the
majority of cases, in the middle of one of its long walls. Yet
three of the St. Gall houses belong to the former type.[52]
Another difference is to be found in the fact that in such
buildings as the House for Distinguished Guests and the
Hospice for Pilgrims and Paupers, the tables and benches
are ranged around the periphery of the center space (figs.
392 and 396). In the Saga house they are set up in the
aisles and on the cross bench. Only on extraordinary
occasions, namely, when the throng of visitors was so
great that they could not all be accommodated in the aisles
and on the cross bench, were special rows of chairs set up
in the nave of the hall. A typical case in point is the fateful
wedding banquet given in the winter of 1253 in Gizur
Thorvaldsson's home at Flugumyr (figs. 328A-B). The
number of guests attending this party amounted to well
over a hundred men (á œdra hundradi). Since Gizur's dining
hall was only 26 ells long and 12 ells broad (stofan var
sex álna ok tottugu löng, en tólf alna breiđ), the host gave
orders that in addition to all the seats that could be placed
in the aisles (the seating capacity of the aisles and of the
cross bench had already been doubled by the setting up of
an outer row of forechairs), two further rows of stools
should be set up in the center aisle. The latter were borrowed
from the church. "And lengthwise all along the
two benches there were forechairs and all along the center
aisle church stools were set up on which people sat in two
rows."[53] Finally, the outer spaces of the St. Gall house
appear to have been more rigidly separated from the center
space than was the case in the house of the Sagas. But of
this there will be more to say in a later chapter.

[ILLUSTRATION]

284.B THE NORTH GERMANIC HOUSE OF THE SAGA PERIOD, 9TH-13TH CENTURIES

PERSPECTIVE VIEW OF INTERIOR, MADE FOR V. GUDMUNDSSON by E. RONDAHL [after Rosenberg, 1894, 257, fig. 10]

The hall was warmed by a fire burning in the middle of the center floor. An opening in the roof above the hearth served as smoke escape and
admitted light and air to the interior. In the aisles on either side of the fire: to the left the high seat of the owner, to the right the seat for his
most distinguished guest. Between them: the high seat pillars, part of the roof-supporting frame of timbers, but decoratively carved and sacred
to the gods.

THE EARLY MEDIEVAL HOUSE IN THE TERRITORY
OF THE ALAMANNI, THE BAJUVARIANS, AND
THE FRANKS, IN THE
LIGHT OF CONTEMPORARY LEGAL CODES

Gudmundsson's findings are restricted to the Germanic
territories of Sweden, Denmark, Norway, and Iceland,[54]
and cannot, of course, be automatically applied to continental

[ILLUSTRATION]

THE NORTH GERMAN HOUSE OF
THE SAGA PERIOD, 9th-13th CENT.

285.B EXTERIOR VIEW

285.A INTERIOR VIEW

[author's interpretation modified from Walter Schultz]

These sketches attempt to render the appearance of a typical house
of the Saga Period and to demonstrate in particular that the house
received its air and light not through windows in the walls (which
would have made the dwelling too vulnerable to hostile intrusion)
but through an opening in the roof, which also served as smoke
escape and was itself surmounted by a small roof raised slightly above
the level of the main roof. The walls were built of turves and often
even the roof itself was covered with them (cf. fig. 292), blending
house and landscape in a continuous carpeting of grass.

Europe where conditions may have been more
complex. Fortunately the information that Gudmundsson
could extract from the Sagas can be supplemented by some
extremely informative Continental sources. Significant documentary
evidence concerning house construction in the
territory of the Alamanni, the Bajuvarians, and the Franks
is scattered through a number of early medieval legal codes
that regulate, among other matters, the compensation to
be adjudged for damage wrought upon a dwelling and its
sundry appendages.[55]

LEX ALAMANNORUM

The earliest document of this nature is the so-called Lex
Alamannorum, an Alamannic code of law laid down between
716 and 719 by an assembly of thirty-four dukes, thirty-three
bishops, and sixty-five counts, under the presidency
of Duke Lantfrid I (d. 730).[56] Article 82 of this code, which
fixes the compensation for arson, bears out what Tacitus
had stated some seven centuries earlier about the loosely
scattered character of Germanic settlements. This article
establishes the following sanctions:[57] If at night a man sets
somebody's house (domus) or hall (sala) on fire and is
caught and found guilty, he is bound not only to restore
whatever he has destroyed by fire but, in addition, to pay a
fine of 40 shillings. If he lays fire to any other houses in the
yard (in curte), viz., the barn (scuria or granica) or the
storehouse (cellarium), he must likewise compensate for the
inflicted damage and settle with an additional fine of 21
shillings.

Fines are specified in the same manner for damage and
destruction to all the other service structures, the bathhouse
(stuba), the sheepfold (ovilis), and the pigsty (porcaricia),
as well as the houses and barns for the serfs (servi
domus, scura servi, spicaria servi). From this it follows that

the Alamannic farmstead of the beginning of the eighth
century consisted of a group of separate buildings in a
common yard (curtis); its principal structures were the
house (domus) and the hall (sala). Whether domus and sala
are synonymous or indicate a separation of the principal
unit into a living house (domus) and an eating hall (sala)
remains uncertain.

From another passage in the same Lex Alamannorum
we gather incidentally some insight into the inner architectural
layout of such a dwelling. Article 94 states that if a
mother dies in childbirth, leaving a child who expires
after having lived for one hour and opened his eyes during
this time so that he could see the ridge (culmen) and the
four walls (iv parietes) of the house, the maternal inheritance
will fall to the father.[58] This stipulation presupposes a
building open to the roof with internal subdivisions, whatever
they might have been, that did not obstruct the
simultaneous visibility, from the mother's bedstead, of the
four walls of the house and of the ridge of its roof.[59]

LEX BAJUVARIORUM

Information of a considerably more specific nature can be
obtained from the Lex Bajuvariorum. This code of law,
which is slightly later than the Alamannic Law, on which
it draws in part, was compiled between 740 and 748 at a
time when the territory of the Bajuvarians was already
under the firm control of the Franks.[60] Article 10 deals with
arson and the compensation imposed for the destruction of
buildings or their component structural parts by fire or any
other means. The information contained in this article is
so vital to the history of early medieval house construction
that it deserves to be quoted in full:

[ILLUSTRATION]

BRONZE SCANDANAVIAN ORNAMENT. LATE IRON AGE (6TH CENTURY)
UPPLAND. Length 10cm. Redrawn from Marten Stenberger.

ARTICLE 10

De incendio domorum et eorum conpositione.

1. Si quis super aliquem in nocte ignem inposuerit et incenderit
liberi [vel servi] domum, inprimis secundum qualitatem personae omnia
aedificia conponat atque restituat, et quicquid ibi arserit, restituat
unaquaeque subiectilia. Et quanti liberi nudi evaserint de ipso incendio,
unumquemque cum sua hrevevunti conponat; de feminis vero
dupletur. Tunc domui culmen cum XL solidis conponat.

2. De scoria vero liberis, si conclusa parietibus et pessulis cum clave
munita fuerit, cum XII solidis culmen conponat; si autem septa non
fuerit, sed talis quod Baiuvarii scof dicunt, absque parietibus, cum VI
solidis conponat.

De illo granario, quod parc appellant, cum IV solidis conponant.

De mita vero, si illam detegerit vel incenderit, cum III solidis
conponat.

De minore vero, quod scopar appellant, cum I solido conponat.

Et universa parilia restituatur.

3. De minorum aedificiorum, si quis desertaverit aut culmen eiecerit,
quod saepe contingit, aut incendio tradiderit, uniuscuiusque quod firstfalli
dicunt, quae per se constructi sunt, id est balnearius, pistoria, coquina
vel cetera huiusmodi, cum III solidis conponat et restituat dissipata vel
incensa.

4. Si autem ignem posuerit in domum, ita ut flamma eructuat, et non
perarserit et a familiis liberata fuerit, unumquemque de liberis cum sua
hrevavunti conponat, eo quod illos inunwan, quod dicunt, in disperationem
vitae fecerit, et non conponat amplius nisi tantum quantum
ignis consumpserit.

Ducalis vero disciplina integer permaneat. Et si negare voluerit de
istis, cum campione se defendat aut cum XII sacramentales iuret.

De servorum vero firstfalli uniuscuiusque ut manus recisa conponat.

5. Modo qui [de] domorum incensione sermo perfinitum censemus,
incongruum non est, ut de dissipatione domui aedificiorum conpositione
non edisseremus.

6. Si quis relicti vel quolibet causa, per presumptionem vel inimicitiam
nec et incurie aut certe ebitione, liberi culmen eiecerit, domini domui XL
solidos conponat.

7. Si eam columnam a qua culmen sustentatur, quam firstsul vocant,
cum XII solidis conponat.

8. Si interioris aedificii illam columnam eiecerit quam winchilsul
vocant, cum VI solidis conponat.

9. Ceterae vero huius ordinis conponantur cum III solidis.

10. Exterioris vero ordinis columna angularis cum III solidis
conponat.

11. Illas alias columnas huius ordinis singulas cum singulis solidis
conponat.

12. Trabes vero singuli cum III solidis conponat.

13. Exteriores vero quas spanga vocamus eo, quod ordinem continent
parietum, cum III solidis conponat.

14. Cetera vero, id est asseres, laterculi, axes ve quicquid in aedificio
construitur, singula cum singulis solidis conponat.

Et si una persona haec omnia commiserit in alterius aedificio,
amplius non cogatur solvere quam culminis deiectione vel ea quae
maiora huius commiserit criminis; minora huius personae non secuntur,
nisi tantum restituendi secundem legum.[61]

[ILLUSTRATION]

286. HOUSE OF THE LEX BAJUVARIORUM

8th CENTURY

RECONSTRUCTION BY KARL GUSTAV STEPHANI

[I, 1902, 326; redrawn]

Stephani correctly interpreted WINCHILSUL ("columna angularis")
and "caterae columnae huius ordinis" as, respectively, cornerpost
(A-A) and posts standing in the walls connecting them (B-B). Not
mentioned in the text and therefore purely arbitrary are entrances
to porch and main house (D, E), hearth (F), and porch itself (L-L).

The "columna a qua culmen sustentatur quam firstsul vocant"
Stephani correctly interpreted as ridge post (C), but incorrectly
presumed that it referred to a single massive timber erected in the
center of the living space. He also overlooked the reference in the
text to an inner order of posts ("ordo interioris aedificii").

The scale of Stephani's plan as it is drawn seems illogical. Evidence
from site excavations supports the contention that a ridge post such
as he postulates could not have been larger than ca. 18″ × 18″.

This dimension, applied to his drawing as a scale indicator, would
make his proposed living space about 14 feet square, or drastically
less than that needed to accommodate a Bajuvarian freeman's
family and servants. As a diagram, the plan is wholly deceptive with
little of constructive value.

ARTICLE 10

On arson and the compensation payable therefor.[62]

1. If someone sets fire at night to somebody's property, and
ignites the house of a free man (or of a serf) he is bound, first of all,
to pay a fine according to the rank of the person and make restitution
for all of the buildings; and whatever he sets on fire there,
furnishings and equipment, he will have to restore. And with all
free men who have escaped from said fire without their clothes on,
he will have to settle according to their wound money; in the case of
women, however, this will have to be doubled. Moreover, for the
roof of the house, he will have to settle with a fine of 40 shillings.

2. And in the case of the barn[63] of a free man, if it is enclosed
with walls and provided with a lockable bar, he will have to settle
for the roof with a fine of 12 shillings; if, however, it is unenclosed,
what the Bajuvarians call a scof,[64] i.e., a shed without any walls, he
will have to settle with a fine of 6 shillings.

In the case of such a granary, however, as they call a parc,[65] he
will have to settle with a fine of 4 shillings.

But in the case of a mita,[66] if he un-roofs it or sets it on fire, he
will have to settle with a fine of 3 shillings.

But in the case of a smaller one, which they call a scopar,[67] he
will have to settle with a payment of 1 shilling.

And everything he will have to restore in like.

3. In the case of smaller buildings, if someone devastates them,
or tears their roofs down, as often happens, or surrenders them to
fire, which they call firstfalli,[68] he will have to settle for every one
which is separately built, such as a bathhouse, a bakehouse, a
kitchen house, or any other structure of this sort, with a fine of 3
shillings and will have to restore whatever is destroyed or burned
down.

4. However, if he sets fire to a house so that it bursts into flame
yet the house does not burn down and is saved by the members of
the household, he will have to settle the wound money for each of
the free people, because he inunwan[69] them, as they say, i.e., put
them in fear of their life, and beyond that he will not have to make
any further compensations in excess of that which has been consumed
by the fire.

The fines forfeited to the duke, however, remain unaffected.
And if he wishes to contest any of these he will have to defend
himself with a champion or must take an oath supported by 12
oath helpers.

As far as the serfs are concerned the destruction of a house
(firstfalli) will have to be settled in like manner as the cutting off of
a hand.

5. And now, since we deem our ruling on the burning of buildings
completed, it is not inappropriate that we explore in greater
detail the fines imposed upon the destruction of the living quarters
of a household.

6. If someone with criminal or any other intent, through arrogance
or hostility, through negligence or a certain lack of understanding,
tears down the roof of a free man, he will have to settle
with a fine of 40 shillings.

7. If [he tears down] that post by which the ridge is held in place
and which they call firstsul ("ridge post"), he will have to settle
with a fine of 12 shillings.

8. If he tears down in the interior of the building that post which
they call winchilsul ("corner post"),[70] he will have to settle with a
fine of 6 shillings.

9. For the other posts of this order, however, he will have to
settle with a fine of 3 shillings.

10. But for the corner posts of the outer order he will have to
settle with a fine of 3 shillings.

11. For all the other posts of this order he will have to settle, for
each individually, with 1 shilling.

12. For the tie beams,[71] however, he will have to settle each with
a fine of 3 shillings.

13. For the outer beams, however, which we call spanga[72] [literally,
"clamp"] because they hold together the order of the walls,
he will have to settle with a fine of 3 shillings.

14. For everything else, however, that is the boards,[73] the
shingles,[74] and the bracing-struts,[75] or whatever else is used in the
construction of a building, he will have to settle with 1 shilling each.

And if a person has inflicted all of this damage to the building
of another person, he shall not be compelled to pay more than what
is due for the destruction of the roof and whatever crimes he has
committed greater than this. Minor infractions of this person are
not to be prosecuted with the exception of those for which restitution
has to be made according to law.

*

The article then goes on to define the compensation set
for damage to the yard, the braided wattle enclosures, the
pastures, roads, and pathways.

Of all surviving literary sources on early medieval architecture
this article of the Lex Bajuvariorum offers the
fullest and most detailed information on the nature of
contemporary domestic building. In the first place it
confirms what had already been demonstrated by the Lex
Alamannorum, namely, the fact that the West Germanic
farmhouse of the eighth century consisted of an aggregate
of separate structures, which included a living house (domus),
a bathhouse (balnearius), a bakehouse (pistoria), and a
kitchen house (coquina), plus an entire group of agricultural
service structures, such as the various barns and stables
(scoria, granarium quod parc appellant, etc.). But more
importantly, in paragraphs 6-14 we are furnished with an
item by item account of the component members of the
roof-supporting frame of timber. Their functions are defined
by their names, listed often both in Latin and in their
vernacular Old High German form; and their varying
size and structural importance are reflected in the weight
of the fine that is placed upon their damage or destruction.
Listed in the sequence of their constructional importance
they are:

1. Culmen or first: the "ridge" or "ridge beam" to
which the head of the rafters is fastened. Its demolition
entails the collapse of the entire roof; hence, the largest
fine is set for its destruction (40 shillings). In the Lex
Bajuvariorum the term is alternatingly used in the specific
sense of "ridge" or "ridge pole" or as pars pro toto for the
entire roof of the house.

2. Firstsul: "the post by which the ridge is carried" (eam
columnam a qua culmen sustentatur). The structural importance
of this column finds recognition in the fact that the fine
imposed upon its demolition is set at 12 shillings, almost a
third of the fine imposed for the destruction of the whole
house.

[ILLUSTRATION]

HOUSE OF THE LEX BAJUVARIORUM. 8TH CENTURY

287.B

287.A

287.D

287.C

RECONSTRUCTION BY OTTO GRUBER (1926, 24, fig. 13)

The principal characteristic of the house type on which Gruber modeled his reconstruction of the Bajuvarian standard house is that its roof is supported
by three parallel rows of posts, the center row carrying a ridge beam, the outer rows roof plates or purlins. Gruber calls this a "ground floor house for man
and livestock" (ebenerdiges Wohnstallhaus). The earliest extant examples date from the end of the 15th century. They are found on both the Swiss and
German sides of Lake Constance, the Aargau, the Kanton of Bern in the southern parts of the Black Forest, and less frequently in the Saar river basin
and the Eifel Mountains.

In late and post-medieval times the interior of this house was divided, in accordance with the transverse alignment of its posts, into a series of compartments
used for hay or harvest storage (Schopf), usually under a hipped portion of the roof; for livestock (Stall); as central access area to other compartments
(Tenne) and in winter also for wagon storage; as kitchen (Küch); and as a withdrawal area often subdivided by an axial wall into two private rooms
(Stuben), under the roof at the end of the house opposite the Schopf.

In general structural organization this house may well derive from that of the Lex Bajuvariorum, but whether the latter may have been divided into
compartments cannot be decided on textual evidence. (For extant examples see O. Gruber, 1926, and a posthumous study by him, Bauernhäuser am
Bodensee, edited by K. Gruber, Lindau and Konstanz, 1961.)

3. Winchilsul: this member is explicitly said to stand in
the interior of the building (interioris aedificii). It is part of
a columnar order whose individual posts (assessed at 3
shillings) rate only half of its own value (6 shillings).
The context leaves no doubt that winchilsul was the Old
High German designation for the four corner posts of the
freestanding inner frame of timbers which carried the roof
plates and separated the house internally into a center
space and a peripheral suite of aisles. The corner posts
were obviously of a heavier make than "the remaining
posts of this order" (ceterae huius ordinis), since they were
rated twice their value. But rising only midway up to the
roof, they rate in turn only half the value of the ridge-supporting
firstsul.

4. Columna angularis exterioris ordinis: "the corner
column of the outer order of posts." Its penal value amounts
to 3 shillings, in contradistinction to the "other members
of this order" (aliae columnae huius ordinis) which are
assessed at 1 shilling each.

The relative value assessed to all of these members
suggests that the outer wall posts had only half the strength
of the posts of the inner frame.

5. Trabes: the horizontal long and cross pieces ("tie
beams" and "roof plates"), which frame the principal
uprights together. The relation of paragraph 12 to paragraph
13 leaves no doubt that trabes is used as a generic
designation for all those horizontal timbers which connect
the uprights lengthwise and crosswise. Paragraph 12 deals
with the trabes of the inner order, i.e., the "tie beams"
and "roof plates" which connect the principal inner posts
that separate the nave from the aisles of the hall. Their
penal value (3 shillings) is identical with that of the supports
on which they rest, save for the heavier corner posts
(winchilsul) which rate twice that value. Paragraph 13
deals with the trabes of the outer order (exterioris vero) and
refers to them with the Old High German designation:

6. Spanga, "clamp," so-called "because they hold the
walls together." The fine assessed for the destruction of
these timbers, in modern architectural terminology referred
to as "wall plates," is identical with that of the corresponding
pieces of the inner order (3 shillings).

7. Asseres, laterculi, and axes, the "rafters," the "shingles,"
and the "bracing struts". Their penal value is 1
shilling each.

We are not the first, of course, to try our hand at a
reconstruction of the Bajuvarian standard house based on
this meticulous enumeration of its component structural
members. A first attempt of this kind, consisting of a plan
only, was made in 1902 by Karl Gustav Stephani (fig.
286);[76] a second, consisting of a plan and various sections
and elevations, in 1926 by Otto Gruber (fig. 287);[77] and a
third, in the form of an isometric perspective, in 1951 by
Torsten Gebhard (fig. 288).[78]

Stephani's interpretation (fig. 286) of the house as a one-room
structure with a porch on one of its narrow ends
misses the basic message of the text, which makes a clear
distinction between an "inner" and an "outer order of
posts" and within each of these between their "regular
members" and their "corner posts." This suggests a house
that is composed of a center space and a peripheral belt of
outer spaces. Even more untenable is Stephani's explanation
of firstsul as a ridge-supporting center post. I presume
that it was the fact that this term is used in the singular
which induced Stephani to interpret the passage to mean
that the ridge of the Bajuvarian house was supported by a
single post that stood in the very center of the building.
Such an arrangement is constructively incongruous and
must be refuted on both linguistic and architectural
grounds. Linguistically, one finds, the singular form appears
again in the very next paragraph, and there it refers
to a structural member (winchilsul, "corner post") which
by definition cannot have possibly existed in a singular
form, since a house with one corner post would be a logical
absurdity. Constructionally, a ridge beam may be supported
by a center post, but a center post alone could not
possibly hold it in place; its stability required additional
supports at each end of the beam. It must have been
Stephani's faulty exegesis of the text that induced Dehio
to remark with regard to the Lex Bajuvariorum that "the
attempt to reconstruct the Bajuvarian standard house is
unconvincing."[79] The criticism is fully justified when
applied to Stephani, but it would be wrong if it implied, as
the context suggests, that the source did not lend itself to a
convincing reconstruction.

Gruber's reconstruction (fig. 287) comes considerably
closer to the truth; but his internal subdivision of the house
into areas used as stables, barns, and living quarters are derived
from post-medieval house forms (Old Upper Suebian
farmhouse and Hotzen house) and are, therefore, purely
conjectural. Decidedly wrong in Gruber's reconstruction
is the application of the term winchilsul to all the members
of the "inner order" (designated with the Arabic figure 2
in his plan), because the text distinguishes clearly between
the "corner posts" (winchilsul) and the "other columns
of this order" (ceterae vero huius ordinis).

By far the most convincing of all the existing reconstructions
is that of Thorsten Gebhard (fig. 288). As a point of
minor criticism it might be noted that there is nothing in
the Lex Bajuvariorum which would suggest that the center
space was boarded off against the outer space by the solid
wooden paneling shown in Gebhard's reconstruction;
while, conversely, this reconstruction fails to show a feature
that is explicitly mentioned in the text, namely, the "remaining
posts of the inner order" (ceterae huius ordinis
[columnae]). Gebhard is probably right when he assumes
that the Bajuvarian standard house had its principal
entrance in the middle of one of its long sides, but again

[ILLUSTRATION]

HOUSE OF THE LEX BAJUVARIORUM

288. AXONOMETRIC VIEW

8TH CENTURY

[reconstruction by Thorsten Gebhard, 1951, 234, fig. 3]

In overall appearance this reconstruction of the house of the Lex Bajuvariorum
is fairly convincing. But like Stephani, Gebhard fails to account
for the inner order of posts ("columnae interioris aedificii") which, the text
states, stood between the cornerposts (winchilsul).

The horizontal timbers connecting the heads of these posts could not have
carried the roof load over very wide spans without additional posting as
described in the text; unsupported, they would surely have sagged or broken.
The same holds true for the ridge purlin. Nor is there any indication in the
text that the center space of the house was separated from the peripheral
spaces by a solid wall partition, as Gebhard shows.

*

The orientation of the large group of buildings at Zwenkau-Harth (fig. 288.X.a) is
conspicuous in the treatment to be found in Quitta, 1958, and was possibly to gain advantageous
solar exposure, or protection from the wind.

[ILLUSTRATION]

ZWENKAU-HARTH NEAR LEIPZIG, GERMANY

288.X.b SECTION

3RD MILLENNIUM B.C.

[after Quitta, Neue Ausgrabungen in Deutchland, 1958, 69 and 75]

Houses divided lengthwise into four aisles by three axial rows of posts,
carrying ridge beam and purlins, were among main characteristics of the
architecture of the Banded Pottery People (Bandkeramiker) who introduced
agriculture and animal husbandry into Central Europe between 50003000
B.C., and who, owing to their sedentary life of seeding and harvesting,
became the first European village builders.

A distinctive construction feature of their houses is the transverse alignment
of the roof-supporting posts that divide the house crosswise internally
into a sequence of compartments—a trait perplexingly similar to the partitioning
of the late and post-medieval houses studied by Gruber (fig. 287).
The house of the Lex Bajuvariorum, as well as its late medieval derivatives,
may have its first roots in this Neolithic house tradition, but the precise
manner in which these concepts might have been transmitted over three
milennia is not known. (For possible Bronze and Iron Age links, see fig.
289.A)

this is a purely conjectural feature. In my reconstruction
(fig. 289) I have limited myself to showing only those members
which are explicitly mentioned in the Lex. The Lex
does not tell us anything about the position of the hearth,
but the location of the hearth is not in question. In structures
of this type the hearth was always in the middle, or
somewhere else along the axis of the center space, at
maximal distance from the incendiary timbers of the walls
and the roof.

Dehio, then, greatly underrated the importance of the
Lex Bajuvariorum for the history of early medieval house
construction. This code not only lends itself to a structural
reconstruction of the Bajuvarian standard house, but it does
so with singular explicitness, and from the information thus
obtained we can draw general conclusions that are of
importance for the broader issues of our study. Foremost
among these is the recognition that during the eighth
century a European house type existed with a general
design that closely resembled the North Germanic house of
the Saga period. Like the latter, it is a skeletal timber
structure and is covered by a large pitched roof, whose
rafters converge in a ridge pole.

There are some distinctive differences, to be sure. In the
Saga house, as has been pointed out, the ridge pole was carried
by short king posts (dvergr) that rose from the center of
the cross beams. In the house of the Bajuvarians the ridge
was supported by posts that rose from the ground. The
Saga house was three-aisled like the Germanic all-purpose
house discussed below, pp. 45ff. The house of the Lex
Bajuvariorum is four aisled, bearing striking, yet so far
inexplicable, resemblance to a house type common in
Central Europe in the 3rd millenium B.C. (see caption,
288X).

CAROLINGIAN CROWN ESTATES AND THEIR
HOUSES, IN THE LIGHT OF
CONTEMPORARY ADMINISTRATIVE ORDINANCES
AND PROPERTY DESCRIPTIONS

In contradistinction to the Alammanic and Bajuvarian law,
the law of the Franks (Lex Salica)[80] is a disappointingly
unrewarding source of architectural information. It does
not include a special chapter on arson, nor does it otherwise
define the fines imposed upon the demolition of the whole
or any part of the Frankish house. But this deficiency is
compensated for, to some extent, by the survival of two
administrative ordinances of the Frankish court which give
us some insight into the architectural layout of a royal
crown estate, the Capitulare de villis and the so-called
Brevium exempla.

CAPITULARE DE VILLIS

The Capitulare de villis,[81] an ordinance formerly assumed
to have been drawn up in 794 or 795 by the young Louis
the Pious in order to correct certain abuses that had
arisen in the administration of the royal estates of Aquitania,
is now believed to have been issued by Charlemagne
shortly before 800 as a directive to the entire empire
(except Italy) in part to curtail mismanagement, in part to
set a program for the future. Among the seventy-odd
articles of which it is comprised, there are some that refer
to architecture. They read like a description of some of the
guest and service structures of the Plan of St. Gall, and
exhibit with vivid distinctness the basic similarity of the
architectural layout of a secular and a monastic Carolingian
manor. In fact, being laid down for the specific purpose of
defining what buildings are considered to be indispensable
components of a royal estate, they form literary counterparts
to the agricultural service structures of the Plan of
St. Gall. While providing us with a comprehensive picture
of the diversity of buildings associated with Carolingian
crown estates, they unfortunately do not tell us anything
about their design or construction.

I am extracting from these articles whatever appears to
have a bearing on architecture, without regard to the order
in which this material appears in the original.

Article 27 prescribes: "At all times our houses [casae
nostrae] shall be provided with fireplaces and fire[?]guards
[foca et wactas habeant] so that they do not suffer any
damage."[82]

Article 42 specifies the household equipment of the
royal supply room (camera). It stipulates that it be provided
at all times with its full complement of bedding,
tableware, cutlery, cooking equipment, and all other kind
of utensils, so that one will never be in need of sending for
them or borrowing them from outside. It contains nothing
further that would shed any light on the layout of the
royal mansion itself.[83]

Article 41 provides, "that the buildings in our estates
[intra curtes nostras], and the surrounding fences [sepes] be
well guarded and that the stables [stabulae], the kitchens
[coquinae], the bakehouse [pistrina], and the presses [torcularia]
be planned with care, so that our men [ministeriales
nostri] can perform their functions properly and with
cleanliness."[84]

[ILLUSTRATION]

289.A HOUSE OF THE LEX BAJUVARIORUM. 8TH CENTURY

PERSPECTIVE WITH ROOFING REMOVED, SHOWING STRUCTURAL SCHEME

AUTHOR'S INTERPRETATION

The relative severity of the penalties imposed by the Lex Bajuvariorum to compensate a householder for willful damage done to his dwelling
(see fig. 289.B) is clearly related to the size and structural importance of the particular timber involved. The preoccupation of the text with
penalties for "pulling down" house timbers presumes that in general the overall framework of the typical house was sufficiently light, and its
key timbers sufficiently accessible, to make this mode of revenge an attractive nuisance.

Timbered early medieval houses with a central row of posts supporting the ridge parlins have, since this chapter was written, appeared in
excavations in Manching and Kirchheim, near Munich (see Schubert, Germania, L (1972), 110ff, and Dannheimer, IBID., L1 (1973), 168ff.
For sporadic Bronze and Iron Age antecedents see Zippelius, 1953, 19, fig. 2; Reinerth, I, 1940, 16, fig. 4b; Pl. 6 opposite p. 26; 28, fig. 7;
139, figs. 60-62; 198, fig. 85.)

I am not aware of the existence of any Central European Bronze and Iron Age houses with three parallel rows of roof-supporting posts. The
connection of the house of the Lex Bajuvariorum with those of the Banded Pottery People suggested in fig. 289.X must therefore be treated
with caution.

In West and North Germanic territory, houses with a row of center posts for carrying ridge purlins are a great rarity. Notable exceptions are
the two Iron Age houses of Wijchen, shown below, figs. 300 and 301.

[ILLUSTRATION]

289.B HOUSE OF THE LEX BAJUVARIORUM. 8TH CENTURY

PLAN. STRUCTURAL MEMBERS IDENTIFIED, WITH FINES LEVIED TO COMPENSATE DAMAGE

AUTHOR'S INTERPRETATION

Article 23 prescribes: "Our superintendents shall see to
it that each of our estates be provided with its dairy
[vaccaritia], its piggery [porcaritia], its facilities for raising
sheep [berbicaritia], its facilities for raising goats [capraritias],
and its facilities for raising billy goats [hircaritias];
and of all this they shall have as much as they can handle;
and none of our estates shall be without these installations."[85]

Article 46 prescribes, "that the enclosures for animals
commonly referred to as brogli lucos nostros, quos vulgus
brogilos vocat be well guarded, and always kept in good
repair, and that one should not wait until it is necessary to
rebuild them anew; and the same applies to all of the buildings."[86]

Article 50 prescribes, that each superintendent determine
the number of chickens that should be kept in each stable
(stabulo) and the number of caretakers to be stationed with
them. (In Article 19 it had already been established "that
not less than 100 chickens and 30 geese shall be kept in the
barns of our main estates [ad scuras nostras in villis capitaneis]
and not less than 50 chickens and 12 chickens and
12 geese in our outlying settlements [ad mansioles].")[87]

Article 45 prescribes, "that each of our superintendents
see to it that he have skillful craftsmen [artifices] in his
district [in suo ministerio], that is: blacksmiths [fabros ferrarios],
goldsmiths [aurifices], silversmiths [argentarios], shoemakers
[sutores], lathe workers [tornatores], carpenters [carpentarios],
shieldmakers [scutarios], fishermen [piscatores],

[ILLUSTRATION]

KÄNNE (STAVGARD), PARISH OF BURS, GOTLAND,
SWEDEN

GERMANIC LONGHOUSE, 3RD-5TH CENTURY

PLAN [after Stenberger, II, 1955, iii, fig. 357]

The house was built in two stages. Its northern half (the original dwelling) had
a floor of stamped clay. The inner walls were lined with heavy granite boulders.
The roof was covered with turves that fell into the house as its supporting
timber frame collapsed, smothering the fire that destroyed it.

The floor of the southern half of the house was paved with fine gravel. Its roof
was of lighter construction and its walls less solidly built than the northern half.
Entrances were in the gable walls.

falconers [aucipites id est ancellatores], soapmakers [saponarios],
brewers [siceratores], that is, those who know how to
make beer [cerevisam], apple cider [pomatium], pear cider
[piratium], and any other kind of drink; the bakers [pistores],
who make pastry for our table, the netmakers
[retiatores] who know the art of making nets for the hunt,
as well as for fishing and for the catching of birds; and all
such other craftsmen [reliquos ministeriales] which it would
be too long to enumerate."[88]

BREVIUM EXEMPLA

The Brevium exempla ad describendas res ecclesiasticas et
fiscules consist of three specimen descriptions of property,
more or less fiscal in character, and were presumably
drawn up for the guidance of the royal agents who assessed
the produce of the domain.[89] The first description is of the
possessions of the see of Augsburg on an island in Staffelsee
in Bavaria, the second is part of a register of the possessions
of the Abbey of Weissenburg in Alsace, and the third is
the survey of five royal fiscs directly belonging to the crown.
Two of these are listed by name, viz., the estates of
Asnapium (Anappes in France, dép. Nord, arr. Lille,
cant. Lannoy), and the estate of Treola (no longer identifiable,
probably in Alamannia); three others are left anonymous
(perhaps the hamlets of Vitry, Cysoing, and the
Soumain near Anappes). The date of the Brevium exempla
is uncertain, but the prevailing view is that they were
written about 812.

Considerably less interesting from a general historical
point of view than the Capitulare de villis, the Brevium
exempla have the virtue of being more detailed and factual
in their reference to architectural conditions. Here we are
given a precise account not only of the number and type of
buildings found on each of the five aforementioned
estates, but also of the construction materials, and in the
case of the royal mansions, even the number and type of
rooms. The following passages from the Brevium exempla
describe portions of the crown estates of Anappes and its
outlying settlements, Treola, and three holdings ("anonymous
estates") not cited by name.

The crown estate of Anappes and its outlying
settlements

Invenimus in Asnapio fisco dominico salam regalem ex lapide factam
optime, cameras III; solariis totam casam circumdatam, cum pisilibus
XI; infra cellarium I; porticus II, alias casas infra curtem ex ligno
factas XVII cum totidem cameris et ceteris appendiciis bene compositis;
stabolum I, coquinam I, pistrinum I, spicaria II, scuras III. Curtem
tunimo strenue munitam, cum porta lapidea, et desuper solarium ad
dispensandum. Curticulam silimiter tunimo interclausam, ordinabiliter
dispositam, diversique generis plantatum arborum.[90]

[ILLUSTRATION]

LOJSTA, GOTLAND, SWEDEN

291.C

291.B

291.A

GERMANIC HOUSE

3RD-5TH CENTURY

RECONSTRUCTION BY G. BOETHIUS
AND J. NIHLEN

[photos: Statens Historiska Museet, Stockholm]

A. Foundation of house after excavation.

A magnificent and one of the first excavated
examples of an aisled Germanic house of the
Migration Period. Its walls were made of
earth carefully lined with stones. The roof
was supported by two rows of wooden posts
rising from flat stones all of which were still
in place. These supports must have been
framed at their heads into stable trusses by
means of cross beams and long beams. The
entrance was in the western gable wall; the
hearth in the middle of the center floor
toward the inner end of the hall.

B and C. Reconstruction of the dwelling.

Reconstructed at full scale on the original
site in 1932, the dwelling follows drawings
submitted by the excavators. Although now
questioned in the rendering of certain
details, this reconstruction nevertheless
gives a very accurate impression of the
unitary quality of the interior space
unmarred by the fact that its roof-supporting
frame divides into a multiplicity
of bays. The roof may not have been
covered with thatch but with turves. The
walls were originally a little higher, and
the entrance wall was probably not straight
but hipped at the eastern end of the roof.

[ILLUSTRATION]

292.A ÞÓRSÁRDALUR VALLEY, ICELAND. HALL STÖNG

PLAN OF HOUSE [after A. Roussel in Stenberger, 1943, 78, fig. 137]

I. Fore room, Jorskáli

II. Sleeping house, skáli, divided by transverse partition into room for men,
karlskáli, and room for women, kvennaskáli

III. Living house, stofa

IV. Dairy, mjólkrbûr

V. Room for cold storage, kjátlari

The house had only one entrance and no windows; it received light and air through a lantern-surmounted opening in the roof. Its turf walls were raised on a stone
foundation two courses high; the roof likewise was covered with turves. The center floor of the main house (II) was of stamped clay and contained a fireplace. Two rows
of posts divided this space into three aisles, the two side aisles being raised and boarded, and partitioned transversely into men's and women's sleeping quarters. A
square area boarded off at the inner end of the south aisle probably formed a sleeping alcove for the farmer and his wife.

The living room (III) contained a hearth for cooking, a stone box 50cm deep. The dairy (IV) was accessible only from inside the house and contained three round
impressions in the floor, presumably from large vats. Its walls were lined with lava stones to a height of 1.1m. A room presumably for cold storage (V) was accessible
only from the fore room (I).

The photograph (fig. 292.B) taken from the door of the living room shows the excavation of the main hall, and reveals with great clarity how the aisles and floor of
the fore room were raised above the level of the center floor. The banked earth of these side aisles was retained by staked boards. Large flat stones at 2-meter intervals
provided footing for the roof posts. Smaller stones set along the walls, pieces of wood still attached, show the house was wainscotted. Absence of personal effects indicates
the residents were forewarned of the eruption of Mt. Hekla, in 1300, that destroyed the house and converted the fertile valley into a wasteland of lava and ash.

The reconstruction (fig. 291.C) portrays the ingenious simplicity with which man could, in a harsh Atlantic climate, make a dwelling not only secure against attack, but
warm and homely as well. The compact top-growth of Iceland terrain is well suited to turf-cutting. For timber the chieftains of the Saga Period relied on wood
imported from Norway, or on driftwood swept in by Atlantic storms from distant North American coasts. The only locally available building material was a dwarf
birch whose fine branches were used as matting for the roof turves.

[ILLUSTRATION]

292.C INTERIOR VIEW OF HOUSE. REDRAWN FROM ROUSSEL, 1943, 211, fig. 144

[ILLUSTRATION]

292.B FOUNDATIONS OF HALL AFTER EXCAVATION. PHOTO COURTESY OF A. ROUSSEL

[ILLUSTRATION]

293. EZINGE (GRONINGEN), THE NETHERLANDS

FOUNDATIONS,

HOUSE A OF WARF-LAYER VI, 4th CENTURY B.C.

[photo by courtesy of A. E. Van Giffen]

The remains of this flatland-level farmhouse show that its interior was divided into a broad center space and two aisles, each roughly half the
width of the nave, by two rows of roof-supporting wooden posts of young, unscantled oak. Their stumps, cut a few feet above the original floor
level when the house was dismantled to make a new settlement on higher ground, were well preserved to a depth of several feet. Braided
wattlework walls formed an enclosure slightly inside the perimeter of outer posts and independent of them. The corners of the house were rounded,
suggesting that the roof was hipped over its narrow ends. A cross partition divided the interior into a dwelling area containing a fireplace, and
a much larger byre for livestock. The building was entered on one of its long sides. In a rectangular yard extending to the north, nine rows of
posts formed supports for a wooden platform presumably used to store fodder and other produce.

This settlement was dismantled after about a hundred years, because the rising waters of the North Sea made it unsafe to live on this horizon.
As centuries passed and the inundation level continued to rise, the site developed as a dome-shaped mound on successively higher, broader levels,
formed by earth, turves, and manure thrown up by the dwellers. The growth of the settlement is traceable through six layers over seven
centuries. The mound attained a diameter of 450m and a center height of 5.5m. The terrain elevation seen at the right is an undisturbed
portion of the present surface of the mound, now occupied by the church and houses of modern Ezinge.

We found on the royal estate of Anappes the royal hall built in
stone, in the best manner, three chambers, the entire house surrounded
by solaria; with eleven heatable rooms[91] and below one cellar;
two porches; seventeen other houses within the main yard,[92] built
in timber, with the same number of chambers, and other appendices,
all well constructed; one stable, one kitchen, one bakehouse, two
grain barns, three other barns. The main yard well protected with
a fence,[93] with a masonry gate, and above this, a solarium. The
smaller yard likewise enclosed with a fence built in the usual
fashion and planted with various types of trees.

The document subsequently lists the dead and live stock
at Anappes down to the smallest detail, and then turns to
the inventory of the outlying settlements:

In the estate of Gruson[94] we came upon the outlying settlements.
There are three barns, and the yard is surrounded by a fence. . . .

On another estate we found the outlying settlements and the
yard protected with a fence, and inside three barns. . . .

On a third estate [literally, on "that estate"] we found the
outlying settlement to be comprised of two barns, one granary, one
garden and the yard well protected with a fence. . . .

The crown estate of Treola

Invenimus in Treola fisco dominico casam dominicatem ex lapide optime
factam cameras II cum totidem caminatis, porticum I, cellarium I, torcolarium
I, mansiones virorum ex ligno factas III, solarium cum pisile
I; alia tecta ex maceria III, spicarium I, scuras II, curtem muro
circumdatam cum porta ex lapide facta. . . .[95]

We found on the crown estate of Treola the royal mansion built
excellently in stone, two chambers with the same number of heatable
rooms, one porch, one cellar, one press-shed, three houses for
men, built in timber, a solar with one heatable room, three other
houses [literally, "roofs"] in masonry, one granary, two barns,
the yard surrounded with a wall and [provided] with a stone-built
gate. . . .

The first anonymous estate

Repperimus in illo fisco dominico domum regalem, exterius ex lapide et
interius ex ligno bene constructam; cameras II, solaria II. Alias casas,
infra curtem ex ligno factas VIII: pisile cum camera I, ordinabiliter
constructum; stabolum I. Coquina et pistrinum in unum tenentur.
Spicaria quinque, granecas III. Curtem tunimo circumdatam, desuperque
spinis munitam cum porta lignea. Habet desuper solarium. Curticulam
similiter tunimo interclusam. . . .[96]

We found on that crown estate the royal house, externally built in
stone and inside well constructed in timber; two chambers, two
solars. Within the main yard eight other houses built in timber; a
heatable room with one chamber built in the usual fashion, one
stable. Kitchen and bakehouse built together, five grain barns,
three granaries. The court surrounded with a fence, above provided
with spines, with a wooden gate. It has above a solar. The smaller
yard likewise enclosed by a fence. . . .

[ILLUSTRATION]

294. EZINGE (GRONINGEN), THE NETHERLANDS

PLAN [after Van Giffen, 1936, Beilage I, fig. 5]

HOUSE A OF WARF LAYER VI, 4th CENTURY B.C.

Plan of the house and storage platform, the remains of which are
shown in the preceding figure. The excavated area is identical with
that shown in figure 296, which shows the next stage of the
settlement.

[ILLUSTRATION]

295. EZINGE (GRONINGEN), THE NETHERLANDS

EXTERIOR VIEW OF SETTLEMENT, 4th CENTURY B.C.

[redrawn from reconstruction by H. Reinerth, 1940, 88, fig. 25]

The discovery of this Iron Age village in 1931-34 was a great landmark in the history of premedieval house construction in transalpine Europe.
The find showed that a house well portrayed by Albrecht Dürer (fig. 335) and Peter Bruegel the Elder (fig. 336) was already fully developed
and in common use for close to 2,000 years.

Later excavations brought the even more startling discovery that this same house type was a standard construction form as early as 1250 B.C.,
and perhaps even in the 14th century B.C. (fig. 323). In the lowlands of Holland and Northern Germany, the same house is used even today
with only minor modifications, for the same purposes for which it was originally conceived (Frisian Los-hus, Lower Saxon Wohnstallhaus).
Its life span is at least 3,300 years, and does not yet appear to have entered its terminal phase.

The most distinctive trait of this type of structure is that it offers, with only a minimum of materials, an ingeniously simple method of covering
large spaces beneath a vast roof carried by a frame of light timbers; these divide the interior of the house lengthwise into nave and two aisles
(figs. 297, 298) and crosswise into a multitude of separable yet transparent bays.

The building type owes its longevity to its ability simultaneously to offer spatial
unity and spatial divisibility. In pre- and protohistorical times almost exclusively
confined to dwelling, sheltering of animals, and harvest storage, the structure entered,
in response to growing complexities of medieval life and social organization, a
virtually explosive phase of functional variety, and came to fill many diverse needs.
On the highest of society, it appeared as residential and administrative seat for
feudal lords and their retainers (figs. 339, 340, and 344-348), including the king
himself. It was used as church (Horn, 1958, 4, figs. 3-8) and Horn, 1962); as
hospital for the sick and infirm (figs. 341-343); as meeting and council hall for the
guilds. And from the 12th century onward in response to the rise of international
trade it became, in Paris and countless smaller towns of France, the standard form
for urban market halls, under whose sheltering roofs the local peasants and traders
from distant places could rent stalls from which to sell produce and goods (Horn
1958, 15ff; Horn and Born, 1961, Horn, 1963).

The second anonymous estate

Invenimus in illo fisco dominico casam regalem cum cameris II totidemque
caminatis, cellarium I, porticus II, curticulam interclusam
cum tunimo strenue munitam; infra cameras II, cum totidem pisilibus,
mansiones feminarum III, capellam ex lapide bene constructam; alias
intra curtem casas ligneas II, spicaria IV, horrea II, stabolum I,
coquinam I, pistrinum I; curtem sepe munitam cum portis ligneis II et
desuper solaria.[97]

We found on that crown estate the royal house with two chambers
and the same number of heatable rooms, one cellar, two porches,
the smaller yard enclosed by a well-built fence; inside, two chambers
with the same number of heatable rooms, three houses for women,
a chapel well constructed in stone, two other timber houses in the
court, four grain barns, two hay barns, one stable, one kitchen, one
bakehouse. The main yard protected with a fence with two wooden
gates and solaria above.

The third anonymous estate

Repperimus in illo fisco dominico domum regalem ex ligno ordinabiliter
constructam, cameram I, cellarium I, stabolum I, mansiones III,
spicaria II, coqinam I, pistrinum I, scuras III, Curtem tunimo circumdatam
et desuper sepe munita. . . Portas ligneas II. . . .[98]

We found in that crown estate the royal house constructed in timber
in the usual fashion, one chamber, one cellar, one stable, three
dwellings, two grain barns, one kitchen, one bakehouse, three
barns. The yard surrounded with a wall, protected above by a
fence . . . two wooden gates. . . .

While failing to reveal anything about the architectural
design of the enumerated structures, the Brevium exempla
are of particular value because they offer concrete information
about the relative use of stone and timber in the architecture
of a Carolingian crown estate. The account of the
sala regalis at Anappes—as we had occasion to point out in
an earlier chapter—[99] with its open solariums, heatable
rooms, and two galleried porches reads like a description
of the Abbot's House on the Plan of St. Gall. Like the
latter, it was composed of several stories and built in
stone. The Brevium exempla, however, make it equally
clear that stone was not considered to be the ordinary
material. With the exception of the chapel of the second
anonymous estate and the two gate houses at Anappes and
Treola, stone appears to be the exclusive prerogative of the
royal mansion, and the superlative form of the epithets
associated with the use of this material (ex lapdie optime
factam) is indicative of the high esteem in which this
material was held. However, only two out of five royal
mansions recorded in the document were entirely stone
structures, those of Anappes and Treola. The domus regalis
of the first anonymous estate had its outer walls constructed
in stone, but everything else inside is in timber (exterius
ex lapide et interius ex ligno bene constructam.) The domus
regalis of the third anonymous estate was built entirely in
wood (ex ligno ordinabiliter constructam). Timber, we will
also have to assume, was used where no specific reference
to any material is made for the casa regalis of the second
anonymous estate. All the other structures of the five
estates either are explicitly said to be built in timber or

must be assumed to be built in timber because of the
absence of any statement to the contrary. And the timbered
buildings formed, of course, an overwhelming majority.

SUMMARY AND CONCLUSIONS

If we review the evidence obtained from the analysis of the
legal and administrative documents discussed on the preceding
pages, we find ourselves confronted with results of a
widely varying nature. The most illuminating of the considered
sources is doubtlessly the Lex Bajuvariorum. It has
furnished us with a body of specific and detailed architectural
information that enables us to reconstruct the Bajuvarian
standard house of the beginning of the eighth
century. The Lex Alamannorum conveyed a clear idea of the

[ILLUSTRATION]

297. EZINGE (GRONINGEN), THE NETHERLANDS

INTERIOR, HOUSE B, CLUSTER SETTLEMENT, Warf-layer V, 4th-3rd centuries B.C.

[author's reconstruction redrawn by Walter Schwarz]

House B of Warf-layer V played a dominant role in our attempt to identify the constructional features of the guest and service buildings of
the Plan of St. Gall (see below, 77ff). Like the majority of the latter, it is entered broadside through a long wall, and in layout consists of a
spacious inner hall with open fireplace in the axis of the house, and a peripheral suite of outer spaces accessible only from the center floor and
used for more specialized functions such as sleeping, or the stabling of livestock.

This is a reconstruction of the interior of House B, which appears at the bottom right of the plan of Warf-layer V, fig. 296 (and at a larger
scale in fig. 327). The drawing first published in Horn, 1958, 7, fig. 13, was made before the excavator realized that the animals stood with
their heads not inward, but toward the outer walls of the dwelling (cf. below, p. 53 n. 64). The braided wattle mats running along the posts on
either side of the center aisle were found to be manure mats, not fodder mats as previously supposed. Since the artist is no longer alive, and
since his handsome drawing portrays quite persuasively the general character of the space in the dwelling, we decided against trying to retouch
the drawing; the animals remain incorrectly positioned.

general layout of a West-Germanic farmstead of this
period with its principal living unit, the domus or sala, and
its variety of special service structures scattered throughout
the yard and the fields. But they told us little, if anything,
about the architectural design of these structures. The
Capitulare de villis gave us an insight into the administrative
complexity of a Frankish crown estate. The Brevium
exempla, finally, provided us with a precise statistical
account of the number and type of buildings to be found
on five such Carolingian crown estates, and illustrated how
on this highest level of Frankish society a new material,
stone, began to intrude into the northern tradition of building
in timber. They told us a good deal about the number
and type of rooms of which the individual buildings were
composed—but they told nothing about the constructional
features of these rooms, or the houses of which they were a
part.

Thus we would still remain thoroughly ignorant about
the architectural layout and design of a Carolingian residence
and its agricultural service structures were it not for
the light that has recently been thrown on this question by
our colleagues in the field of archaeology.

V.2.2

ARCHAEOLOGICAL EVIDENCE

Until the end of the second decade of this century the
literary sources discussed on the preceding pages were all
that students of early medieval house construction had to
lean on when discussing the question of Northern parallels
for the guest and service structures of the Plan of St. Gall.
To be sure, some isolated excavations had already been
made in Sweden and Iceland,[100] but this fact was not
widely known; and the procedure for unearthing houses
whose structural members, in many cases, could be identified
only by a shadowy patch of soil discoloration left in
the ground as they rotted away, had as yet not developed
into that highly accomplished technique so successfully
practiced today. But in 1928-1931 this situation began to
change when John and Nils Nihlen laid bare on the island
of Gotland two three-aisles houses of the Migration Period
which looked like physical embodiments of Gudmundsson's
Saga house. In 1930-1934 the Dutch anthropologist Albert
Egges van Giffen initiated a new era of northern house
research with the excavation of an Iron Age dwelling
mound in a hamlet called Ezinge (Groningen) in Holland,
which revealed that a very similar type of dwelling was in
use as early as the fourth century B.C. in the territory of the
Frisians, a West Germanic tribe. In the two following
decades the information gathered from these excavations
was broadened by an increasing number of further discoveries.
At the date of this writing we are able to trace, on
the basis of several hundred excavated dwellings, the
development of the timbered three-aisled house in the
Germanic territories of transalpine Europe from its beginnings
in the Middle Bronze Age through the Iron Age, into
the Early Middle Ages, and through the Middle Ages to
its modern survival form

KÄNNE, BURS, GOTLAND, SWEDEN

The first example of this long line of excavations, as just
remarked, was a three-aisled dwelling, excavated in 1928 by
John and Nils Nihlen, in a place called Känne, in the
parish of Burs, in East Gotland (fig. 290).[101] It was 33 feet
wide (10 m.) and had the extraordinary length—not as
yet matched by any dwelling subsequently unearthed—of
203 feet (62 m.) A recent review of the site has disclosed
that the hall was constructed in two successive phases, and
in its original state was only half as long.[102] Its roof was
supported by two rows of freestanding inner posts, rising
in pairs, at intervals of 9 to 13 feet (3 to 4 m.). Each of
these uprights was firmly secured in the ground by a
ring-shaped wrapping of stones. Over fifty charred beams
and numerous fragments of wood were found on the floor;
among these were the remains of two large beams which
were jointed into each other at right angles. The walls
consisted of solid banks of earth heavily interspersed with
small stones and were faced, outwardly and inwardly, with
a strong lining of heavier stones. The roof must have been
covered with sods of turf, as no other material would have
smothered so effectively the fire that destroyed the house
yet preserved so much of the timbered frame of the roof.
The hall received its warmth from two hearths which lay in
the middle of the center aisle, one of them 33 feet (10 m.)
long. "Longfires" of this kind are well attested from the
Sagas, where they are referred to as langeldar or máleldar.[103]
The general character of the accessories found in the house
pointed to about the year A.D. 200 as the approximate
period of construction.

LÖJSTA, GOTLAND, SWEDEN

The second house, a structure of more normal proportions,
85 feet by 33½ feet (26 m. × 10.5 m.), was excavated in
the summer of 1929 in the vicinity of castle Lojsta in
Gotland (fig. 291A-C).[104] It was the same construction type
except that here the roof-supporting posts were not sunk

[ILLUSTRATION]

298.B EZINGE (GRONINGEN), THE NETHERLANDS. CATTLE BARN OF Warf-LAYER IV, 2nd CENTURY B.C.

[author's reconstruction, drawn by Walter Schwarz]

[ILLUSTRATION]

298.A PLAN
REDRAWN FROM VAN GIFFEN

1:150

[ILLUSTRATION]

299. EZINGE (GRONINGEN), THE NETHERLANDS. CATTLE BARN OF Warf-LAYER IV, 2nd CENTURY B.C.

[excavation photo by courtesy of A. E. Van Giffen]

This building, like those unearthed above and beneath it, owes its magnificent state of preservation to the fact that each settlement stratum in
which houses were buried in the course of successive inundations was sealed by sterile layers of sand and clay deposited after flooding, sealing
their content against the infiltration of air and thus protecting it from decay. The roof-supporting posts of oak, the braided walls and cross
partitions (wattled saplings of birch) were preserved to a height of 4 feet. The manure mats were found to be in such good condition that they
could be walked upon without breaking. The building was 29 feet wide (7.20m) and over 75 feet long (23m) but was never excavated to its full
length. Its construction was identical with that of the houses found in the earlier Warf layers (figs. 293-297).

The systematic division of the aisles into stalls, together with the absence of any fireplaces, suggests that it was used for the stabling of livestock
exclusively. Since every stall had room for two head of cattle, this barn must have been able to hold at least 48 animals, striking evidence of
the economic wealth of these early shoreland farmers. Livestock entered and left the building through doors in the two narrow ends—a feature
found in many other early Iron Age houses (figs. 304, 310, 312, 315, 316), and today in the Lower Saxon Wohnstallhaus and the Frisian
Los-hus, modern descendants of this building type.

[ILLUSTRATION]

PRE- & PROTOHISTORIC CARPENTRY JOINTS

300.A

300.B

300.C

300.D

[after Zippelius, 1954, figs. 1, 2, & 5]

A. Forked posts (Neolithic)

B. Post with slit head

C. Mortice and tenon joint in post and plate assemblage
(Neolithic)

D. Mortice and tenon joint in post and ground sill assemblage
(Bronze Age)

in the ground but rested on slabs of stone. All of these
stones were still in their original position (fig. 291A). The
posts themselves had disappeared. Rising freely from stones
as they did, they could only retain their vertical position by
being framed together at the top by means of cross beams
and long beams. Slight irregularities in the longitudinal
alignment of the posts suggested that the cross beams lay
underneath the long beams. The excavators felt so sure of
their interpretation of these conditions that they undertook
to reconstruct the entire hall on its original site (figs. 291B
and C). Some of the details of this reconstruction have since
been questioned, but the doubts amount basically to no
more than that in the original house the walls were probably
a little higher than they are shown at present.[105] The
pottery found in the house suggests as period of construction
the third century A.D. In the fifth century, for unknown
reasons, the hall appears to have been abandoned.

In the two decades that followed probably more than
sixty houses of the Lojsta type were unearthed on the
islands of Gotland and Öland, on the mainland of Sweden,
in Norway and in Denmark,[106] and, last but not least, in
Iceland, the country whose literary tradition introduced us
to this type of dwelling.

STÖNG, ÞÓRSÁRDALUR VALLEY, ICELAND

Iceland was the subject of an expedition undertaken between
1934 and 1939 by a joint excavation team of Danish,
Swedish, Norwegian, Finnish, and Icelandic archaeologists.[107]
I show as a typical example of the results of this
expedition, the plan and excavation photos of the dwelling
of a farm called Stöng in þórsárdalur Valley (fig. 292A-C)
which was settled during the landnám period at the end of
the ninth century and covered by the ashes of nearby
Mount Hekla in an eruption in the year 1300. The dwelling
unit of this farmstead consisted of a long house 98 feet
long, divided into foreroom, sleeping house, living house
(forstofa, area I; skáli, area II; stofa, area III), milkhouse
(area IV), and cooler (area V). The sleeping hall was
54 feet long and 19 feet wide. Its aisles were raised so
as to form continuous "benches"—the langpallar of
Gudmundsson's Saga house. Inserted into the curbs
of these benches about every 6 feet were large blocks
which served as base stones for the wooden uprights that
once supported the roof of the hall (fig. 292B). The fireplace
lay in the middle of the center floor. Two shallow stone
foundations which bisected the aisles crosswise suggest that
the sleeping hall was subdivided by means of wooden cross
partitions into a sleeping house for men (karlskáli) and
another for women (kvennaskáli)—a distinction also well
known from the Sagas. And judging from the presence of
two rows of stones ranged carefully along the base line of
the two long walls, the hall must have been wainscotted
its entire length (the veggþili or langþili of the Sagas).
There was a "crossbench" (þverpallr) on the entrance
side of the hall, raised like the aisles and screened off by a
wooden cross wall. I am drawing attention to this house not

only because it is the keystone of cumulative archaeological
evidence that established the correctness of Gudmundsson's
literary work, but also because this dwelling may date from
the same century in which the Plan of St. Gall was drawn.
During the Iceland expedition of 1934-39 a total of eight
such houses was unearthed. But by the time these excavations
were conducted, discoveries of even greater significance
were in progress on the Continent.

EZINGE, PROV. GRONINGEN, THE NETHERLANDS

The low-lying coastlands of the Netherlands and northern
Germany are dotted with man-made circular earthen works
on which the cattle-raising Iron Age settlers of this territory
erected their dwellings in order to protect themselves from
the heavy tides that flooded the surrounding flatlands
during the storms that lashed the shores of the North Sea
in the winter and spring. These dwelling mounds, called
Warfen or Wurten in German,[108] terpen in Dutch,[109] are
the product of the struggle of man against a geophysical
event of major importance which started some ten thousand
years ago, has as yet not subsided, and is even today only
temporarily checked by an elaborate system of dikes. Since
the retreat of the last great glacial cap of ice the shorelands
of Holland and northwestern Germany have gradually
sunk away in the course of a geological action in which long
periods of sinking alternated with shorter and less effective
periods of uplift.[110] The last of these cycles of sinking
started in the centuries immediately preceding the birth
of Christ and is still in progress. Prior to its inception the

[ILLUSTRATION]

302. WIJCHEN (GELDERLAND), THE NETHERLANDS

PLAN [after Bloemen, 1933, 6, fig. 7]

Alternation of heavy posts with saplings in the outer walls of both
houses reveals that the braided wattle walls did not form an
independent envelope, as with the Ezinge houses, but stood in line
with the outer posts.

[ILLUSTRATION]

303.A FOCHTELOO (FRIESLAND), THE NETHERLANDS

HOUSES OF A WEALTHY WEST GERMANIC FARMER AND HIS FOLLOWERS, 1ST-4TH CENTURIES A.D.

303.A VIEW FROM THE AIR LOOKING NORTHWARD. RECONSTRUCTION BY A. E. VAN GIFFEN, 1954, fig. 85 [drawing based on a sketch by L. Posterna]

303.B PLAN OF SETTLEMENT SHOWN IN AIRVIEW

[ILLUSTRATION]

303.B

This large dwelling was associated with a hamlet of three
similar houses approximately the same width, but only half
its length. It was excavated in 1938 on a sandy elevation
of the Dutch Geest. The presence of roof-supporting timbers
was determined by discoloration in the ground from where
they had rotted away. By this evidence it was ascertained
that the roof of the main house was supported by two rows
of free-standing inner posts, ten in each row, and that they
were of quarter-split oak sunk, rounded side inward, 0.75m
into the earth. This building was buttressed by a large
number of exterior posts set at an angle to help neutralize
the outward thrust of the roof. The walls were of wattle-daubed
clay; the rounded corners and absence of any timbers
capable of supporting a gable suggest that the building's
roof was hipped over its narrow ends. Both main house and
adjacent hamlet were protected by a pallisaded fence, and
the main house additionally by a ditch.

dwellings of the coastland farmers of northern Germany and
Holland lay level with the flat land; but as the land began
to sink away, the water of the North Sea rushed in with
steadily increasing frequency and furor, and forced the
settlers to remove their dwellings to successively higher
levels. This they did by packing the floor level of their
houses with thick layers of turves and animal manure and
by re-erecting new dwellings on these mounds above the
inundation level of the heavy winter tides. As this process
continued century by century, it gave rise to a landscape of
man-made dwelling mounds attaining in their ultimate
stage a diameter of twelve or fifteen hundred feet and a
maximum inner height above the surrounding land of as
much as twelve to eighteen feet.

The effects, although not the cause, of this peculiar
geological phenomenon were known to Pliny the Elder,
who visited this territory probably in A.D. 47 and transmitted
his observations to posterity in a derisive yet
highly descriptive passage of his Historia Naturalis:

A vertical profile cut through such a tumulus or Warf
shows as a rule a sequence of several convex layers of soil
in different coloration; the cultural remains reveal layer by
layer the story of the settlement as it was abandoned and
re-erected on each successive level. The physical composition
of these mounds offers unusually favorable conditions
for the preservation of organic materials, such as wooden
uprights, wattled fences or walls, or even objects made of
leather, since each abandoned settlement was covered by a
solid layer of clay which sealed its contents against the
corrosive action of the air.

In 1930 Albert Egges van Giffen dug a trial ditch through
a mound of this type at Ezinge (Groningen), Holland, and
the ensuing excavation (1932-34)—a landmark in the history
of European house research—enabled him to trace the
development of a West-Germanic settlement from its
beginnings in the fourth century B.C. to its end in the third
century A.D.[112]

The earliest settlement of this site (layer VI) was a
single farmstead (figs. 293 and 294), erected early in the
fourth century B.C. on the natural ground of the flatland. It
consisted of a three-aisled house with walls of wattlework,
and a vast enclosure almost entirely taken up by a platform
for the storage of hay or harvest. The timbers of the roof of
the house had disappeared, but the roof-supporting posts
and the braided walls of the house were preserved to a height
of almost a feet (fig. 293). They consisted of five pairs of
freestandings inner posts and a perimeter of thinner outer
posts. The wattle walls ran independently of this system,
slightly inside the ring of outer posts.

[ILLUSTRATION]

FOCHTELOO (FRIESLAND),
THE NETHERLANDS

304.A

304.B

HOUSE OF A WEALTHY GERMANIC FARMER

1ST-4TH CENTURIES A.D.

PLAN AT LARGE SCALE (A. E. VAN GIFFEN)

Plan of the main house (A) shows that aisles of the six westermost bays are cross
partitioned into stalls for 24 cattle. Entrances in the middle of each long wall lead to a
center bay that separates stock from the dwelling (four eastern bays). B and C: Plans
of the main house, final condition.

[ILLUSTRATION]

305.A LEENS (GRONINGEN), THE NETHERLANDS

AISLED HOUSE WITH TURF WALLS, A.D. 700-1000

PLAN [after A. E. Van Giffen, 1935-40, fig. 16]

The plan above is at level B noted on the transverse section below
306.A) with horizontal fold shading. Shown at right (306.B) is another
building.

[ILLUSTRATION]

305.B SECTION, EXCAVATION,

scale horizontally & vertically, 1:150

Ground penetration at right is about 3.5m = 11.5 ft.

[ILLUSTRATION]

LEENS (GRONINGEN), THE NETHERLANDS

306.A

306.B

AISLED HOUSE WITH WALLS, A.D. 700-1000

TRANSVERSE SECTION [after Zippellius, 1953, 32, fig. 5f]

Dwellings excavated at Leens were of great historical importance since they offered the
first archaeological proof that the aisled Germanic Wohnstallhaus continued to be built
in the Middle Ages. Fig. 306.A shows a house of Warf layer B of seven strata
spanning roughly 3 centuries. The structure was 38 feet long, 16 wide (11.5 × 4.8m).
Layer B also held a house with wattlework walls, the soil structure of which indicated
it was almost 72 feet long.

When the water level of the North Sea had risen high
enough to make living on the flatland intolerable, the
single family dwelling of layer VI was buried under a
man-made mound of sods and turves (layer V) which,
after having reached a height of roughly 4 feet and a diameter
of approximately 90 feet, gave birth to a hamlet that
now comprised a total of five houses (figs. 295-297). These
houses belonged to the same construction type as did the
preceding settlement and were equally well preserved.
Three of them were provided with hearths, and hence
must have served as dwellings for people; one was inhabited
by both men and animals, evidenced by the
presence of both a hearth and two narrow strips of wattle-work
in front of the roof posts, which the excavator
interpreted as fodder mats, but which later excavations
proved to be dung mats.[113] The same condition appears to
have existed in the large house in the center, if this house,
as seems likely, had a hearth in its unexcavated eastern
section. Another smaller house, built at right angles against
this dwelling, had neither hearth nor dung mats, and hence
may have served as barn or general storage area. In the
houses that accommodated livestock the aisles were subdivided
into bays, or stalls, by means of braided cross
partitions, each of the thus-created boxes yielding sufficient
space for the stabling of two heads of cattle, facing the
outer wall perimeter of the house. Three of the houses had
their entrance broadside, two were entered axially. Pottery
shards and other cultural accessories associated with this
settlement permit a rough dating of the third century B.C.

In the second century B.C. the hamlet of layer V was
abandoned and the mound on which it stood was enlarged
to more than twice its original diameter and raised to a
level of 6 feet above the natural ground. On top of this
elevation a new village was built in a circle around an
open yard with the longitudinal axes of the house pointing
radially to the center of the Warf (layer IV).

The houses of this layer were of the same construction as
those of the preceding layers, but in general considerably
more spacious, as one may gather by glancing at the
extraordinary cattle barn reproduced in figures 298-299.
It had a length of over 75 feet (23 m.) even in its uncompleted
state of excavation. The posts and carefully
braided walls of this structure (twigs of birch daubed with
cow manure) were preserved in almost original freshness,
in spots to hip and even shoulder height. The building
contained no hearth, but dung mats ran along the inner
roof supports along the entire length of the structure, and
the aisles were systematically subdivided into stalls by
braided cross partitions.

The circular village to which this barn belonged was in
use from the second century B.C. to the first century A.D.,
but the life span of its houses was found to be considerably
shorter than that of the preceding layers. In certain sectors
van Giffen found that five to ten houses had been superimposed
upon one another in rapid succession; and intermittent
stratification of this settlement horizon with sterile

[ILLUSTRATION]

307. HODORF (HOLSTEIN), GERMANY

AISLED FARM HOUSE, 1st-2nd CENT. A.D.

PLAN [by W. Haarnagel; after Schwantes, 1939, 272, fig. 10]

At the lowest level of the Hodorf WARF lay a flatland farm consisting of a
three-aisled main house divided into living and livestock areas, and an unaisled
barn built in axial prolongation of the house. In the layout of the plan two
measures are clearly discernible, the longitudinal measure of column interval A
and its half measure A/2. This measure and submeasure make up the length of
the house and its AMBAU (= 111 feet). The width of the house (17 feet) appears
to be uniformly twice that of the center aisle. While the observation is simple and
even superficial, it hints, at this period and in this region, of an emerging
awareness of systematic measure in simple building practice and agriculture. All
trace of bulging curvature of wall line, or boat-like plan, has disappeared in
favor of a rather uniform rectangular geometry. Discipline of measure prevails
over scattered spacing and casual positioning of posts. A knot tied midway
between the ends of a braided rope could graphically solve the problem of
division by 2 for men unversed in the mystery of abstract arithmetic. It would,
too, lead to successive halving in series.

courses of sand gave evidence that this village, in its
initial stages at least, was still dangerously exposed to the
destructive action of the heavy winter tides.

In the centuries that followed, the second and third
centuries A.D., the Warf had to be raised again on two
successive occasions (layer IV-III). The house type remained
the same, except that in the later stages the wattle
walls were frequently reinforced externally by heavy layers
of turf. Toward the close of the third century, finally, the
village perished in a fire—an event that van Giffen connected
with the intrusion into the Frisian territory of the
first westward-moving Anglo-Saxons. The spacious three-aisled
houses were now superseded by small rectangular
huts which are of no interest to this study.

The excavation photos shown in figures 293 and 299 convey
in persuasive terms the unusual state of preservation in
which the Ezinge houses were found. They furnished
conclusive evidence about the construction of the walls and
the nature of the principal roof-supporting members (in
places preserved to a height of 4 feet above the ground),
but they told us nothing about the manner in which these
members were framed together at the top into a stable
roof-supporting system, nor how the roof itself was constructed.

There are, nevertheless, a few inferences that can be
made with relative safety from the conditions of the walls
and the placement of the posts. One of these is that the
roof must have been hipped over the narrow ends of the
house. This must be inferred from the fact that the two
end-walls of the house are not provided with posts that
could have carried a gable. The reconstruction of the roof
shapes shown in figures 295 and 297 render this condition
correctly.[114] Second, the principal posts must have been
framed together lengthwise by long beams which were
needed for the support of the rafters. There is no unity of
opinion, however, on whether the posts were in addition
connected transversely by crossbeams. Van Giffen felt
that, provided the posts were set sufficiently deep into the
ground, no such cross-connections were needed; and this
was also, in part at least, the opinion of Joseph Schepers.[115]
The technical soundness of this view, however, was questioned

[ILLUSTRATION]

308. HODORF (HOLSTEIN), GERMANY

EXTERIOR VIEW, AISLED HOUSE

1st-3rd CENTURIES A.D.

[by courtesy of W. Haarnagel]

Although conjectural in many details, this model in the
Niedersachsische Landesstelle für Marschen- und Wurtenforschung,
Wilhelmshaven, is nevertheless a very convincing reconstruction of
the main house of the flatland farm of Hodorf unearthed in 1936-37.
It demonstrates that the Lower Saxon Wohnstallhaus,
surviving examples of which date only to the 15th century, is in fact
a modern derivative of a prehistoric building type.

As in the similar Ezinge houses, the rafters of the roof were carried
by a row of posts placed slightly outside the independent wattle
walls. The rounded corners of these walls, and the absence of any
strong support at the building's narrow ends, suggest that its roof
was hipped. Four round posts around the hearth (fig. 307) and
unaligned with the principal posts, are correctly interpreted as
supports for a canopy raised slightly above the main roof with lateral
openings for light, and smoke escape—a device well known through
the Sagas (see p. 23ff) and crucial for interpretation of the guest
and service buildings of the Plan of St. Gall (see below, pp. 117ff).

by T. Hermanns and Adelhart Zippelius[116] who
pointed out that if the posts were only connected by long
beams, the roof-supporting frame would still be exposed
to the danger of bending and buckling under the strain of
heavy loads of snow or the thrust of the wind during
storms. Moreover, cross connection is suggested by the
extremely accurate transverse alignment of the post, as it is
found in all of the Ezinge houses, as well as in the majority
of Warf dwellings subsequently unearthed. Whether the
cross beams lay beneath the longitudinal timbers, or above
them, must remain an open question.

There appears to be general agreement that the peripheral
row of posts—standing either within the walls of the house
or at a slight distance away from them—consisted of
short uprights terminating in a fork and carrying in that
fork a course of horizontal timbers which served as footing
for the rafters. The wattle walls themselves would have
been too weak to carry the roof. In some of the Ezinge
houses the outer posts were found to lean inward in close
adjustment to the angle of the roof thrust—a feature that
was encountered again in many houses subsequently
unearthed.[117]

The construction of the roof itself has been the subject
of some penetrating, yet careful and equally cautious,
observations made by Adelhart Zippelius.[118] Zippelius feels
that the layout of the Ezinge houses suggests that they were
covered by a continuous sequence of coupled rafters
(Sparrendach). The absence of any trace of posts along the
central axis of the house precludes the assumption of a
ridge pole. In primitive ridge-pole construction the two
sides of the roof were, in general, formed by means of
poles (in German called Rofe) which were hooked into the
ridge piece with their heavy ends upward and suspended
in the pole by a hook formed by the stub of a former
branch. This type of roof construction (Rofenkonstruktion),
ideal for houses of relatively smaller dimensions, could also
be employed in connection with aisled houses, but only if
the width of the nave was not much greater than the width
of the aisles.[119] Zippelius contends that in the Ezinge houses,
where the nave is generally twice the width of the aisles,
this system would not have worked, since the overhanging
portions of the roof poles (over the nave) would have outweighed
the lower portion of the roof, which covered the
aisles. The structural stability of the Ezinge houses required
that the roof poles were laid upon the supporting frame
with their light ends upward. Conjectural as all this may
be, it is based on sound speculation, and in the absence of
more tangible archaeological evidence provides us with as
good a working hypothesis as can be found at present.

Zippelius made some further, no less persuasive, assumptions
about the manner in which these timbers might have
been jointed. The easiest, simplest, and oldest method of
carrying a horizontal log is to lay it upon a row of timbers
terminating in a natural fork (fig. 300A)—a method that
continued to be employed long after more sophisticated
forms of joining had come into use, and is practiced even

[ILLUSTRATION]

EINSWARDEN (NORDENHEIM), GERMANY

309.B

309.A

POST AND WATTLE HOUSE, around the birth of Christ

PLAN AND RECONSTRUCTION [after Zippellius, 1953, 38, fig. 8]

The site is on the estuary of the river Weser. The construction of
this house is virtually identical in all respects with the houses of the
cluster settlement of Layer V of the Ezinge Warf (figs. 295-97 and
fig. 327). Like most of those, as well as the chieftains's house at
Fochteloo (fig. 304), its living quarters (two westernmost bays) are
separated from the stables (two easternmost bays) by a center bay
entered through a door in the middle of the southern long wall,
while the cattle enter through a door in the eastern end wall. The
house is 33 feet long and 16½ feet wide.

today by primitive men throughout the entire world. When
natural forks of the desired height could not be found
among the available logs, the fork had to be shaped with
tools. The closest man-made imitation of the natural fork—
and here again I think Zippelius is correct—is a joint to
which he refers as Pfostenzange and which is obtained by
simply cogging the notched portion of a large beam into a
corresponding slit in the head of the upright beneath it
(fig. 300B). Another way of locking posts into horizontal
timbers (either at the top, bottom, or in-between) is by
means of mortice and tenon joints (Verzapfung), as shown
in figure 300C and D, or by halving them into one another.
Halving would also appear to be the most sensible joint for
the tips of the rafters, the connections being given additional
strength at this point, perhaps, by some braided strands of
willow. The reconstructions shown in figures 297 and 298
attempt to conform with this thinking.[120]

WIJCHEN, MAAS ESTUARY, THE NETHERLANDS

When the Ezinge houses were discovered in 1930-34 they
were a new and entirely isolated phenomenon on the
Continent. But in the five years that followed, before the
outbreak of World War II, every subsequent summer
brought new results. While van Giffen was still at work at
Ezinge, F. Bloemen unearthed under less favorable soil
conditions another group of aisled houses of the first
century B.C. on a mountain range near the estuary of the
river Maas, near Wijchen.[121] The ground plans showed the
transverse alignment of inner and outer posts, which was
typical of the houses of layer V and IV of the Ezinge Warf.
The outer walls consisted of an alternating sequence of one
heavy and two lighter posts; the heavy posts stood in line
with the principal posts (figs. 301, 302). In other aspects,
however, the construction differed. The houses had posts
along their central axes, an arrangement that is in general
interpreted as an indication of the presence of a ridge pole.
The excavation showed that ridge-pole construction, although
unusual, was nevertheless not absent in this territory,
an observation that was confirmed by later finds in
other places.[122]

[ILLUSTRATION]

310.X ISOMETRIC VIEW

[ILLUSTRATION]

310. CROSS SECTION

The house was 102 feet long, 29 feet wide. The nave and one aisle were 10½ feet
wide, the narrow aisle 8 feet wide. The distribution of stones—some for pavement
some for lining or packing of wall-post sockets, others for footing of principal
roof supports—reveals that the house was divided lengthwise into a nave and two
aisles, and crosswise into fourteen bays. In the first ten, only the center floor was
paved, and the aisles were strewn with sand. In the last four bays the pavement
ran across the width of the dwelling. This is the well-known T-shaped floor plan
of the Lower Saxon Wohnstallhaus.

In the house above, bay depth in the stable was 6½ to 8¼ feet. In the living area
the distance between trusses increased, and in the terminal bay containing the
hearth is almost twice as deep as the others. Since the principal inner posts of
the house were footed on stone blocks rather than in post holes, they must have
been framed at their heads by long beams and cross beams somewhat in the
manner shown above. The ANKERBALKEN (cross beams terminating in long
tenons morticed into the main posts a few feet below the tie beams) shown in
Rieck's reconstruction (Reick, 1942, fig. 2) appeared to us to be an anachronistic
feature for so early a structure and for that reason has been omitted in our cross
section.

AALBURG, near BEFORT, LUXEMBOURG

AISLED HOUSE, 5TH CENTURY B.C.

[ILLUSTRATION]

311. PLAN [after G. Rieck, 1942, 27, fig. 1] 1:125

FOCHTELOO, RHEE, SLEEN, AND LEENS,
THE NETHERLANDS

Bloemen's excavation was followed with the discovery by
van Giffen in 1935, 1936, and 1937 of a group of settlements
of the first and second centuries A.D. near the
villages of Fochteloo, Rhee, and Sleen; and in 1938, again
near Fochteloo,[123] of a settlement of the same period which
van Giffen believed to be the farm and residence of a
chieftain (figs. 303-304). This settlement comprised a long
house, protected by fence and ditch, and a nearby hamlet,
likewise fenced in, consisting of three smaller houses and
a couple of open barns. All of these houses were aisled and
were entered broadside by two entrances lying opposite
one another in the middle of the long walls and giving access
to a median crosswalk that separated the stables for the cattle
from the living quarters of the people. The long house of
the chieftain had a third additional entrance at the rear of
the stables, primarily for the use of livestock. This house
was 70 feet long and 21 feet wide (21·40 m. × 6·50 m.).

The great significance of van Giffen's excavations of
Ezinge was that they solved an enigma that had puzzled
students of European house construction for over a century.
They brought to light the prehistoric prototypes of two well-known
and closely related modern house types, namely
that of the Lower Saxon "Wohnstallhaus" and of the
Frisian "los-hus." The oldest surviving specimens of these
two widespread house types date from the early sixteenth
or, at the most, from the end of the fifteenth century.[124]

Van Giffen's excavations demonstrated that this type
was infinitely older than anybody had heretofore presumed
it to be, and their immediate prototypes could now be
traced back as far as the fourth century B.C. It was clearly
only a matter of time for the connecting medieval links to
be found. Once more it fell to van Giffen to lead the way
in this search. A trial ditch run through a Warf in the
vicinity of the village of Leens (Groningen), Holland,
revealed the profiles of a settlement whose life span started
approximately at the point where that of Ezinge ended.
And in a systematic excavation of this Warf conducted in
the subsequent year, van Giffen[125] could trace his aisled
Iron Age house through seven successive layers from the
end of the seventh century A.D. to the beginning of the
eleventh. Altogether some twenty-three houses came to
light: some of them built with wattle walls, others with
walls of turves; but all of them had their roofs supported by
two rows of freestanding inner posts. I reproduce as a
typical example the plan of a house of Layer B (fig. 305),
after van Giffen, and a cross section of this house (fig. 306),
as suggested by Zippelius.[126]

HODORF, SCHLESWIG-HOLSTEIN, GERMANY

Van Giffen's work in Holland was only a beginning. In
1936 Werner Haarnagel launched the first of an equally
exciting series of excavations in the adjacent coastlands of
northern Germany, where he discovered a Germanic flatland
farm of the first and second century, near the village
of Hodorf, Schleswig-Holstein, on the banks of the river
Stör, not far from its confluence with the river Elbe.[127] It
consisted of a three-aisled dwelling with hearth, to which a
one-aisled barn was added axially at a slightly later date
(fig. 307). The construction method employed in this dwelling
was identical, in all details, with those of van Giffen's
houses at Ezinge: six pairs of inner posts serving as principal
roof supports, an outer perimeter of wall posts serving
as footing for the rafters, plus the customary envelope of
wattle walls running in total independence of the supporting
members. The aisles were divided into cattle stalls in the
rearward part of the house, as in Ezinge, except that in
Hodorf this area was entirely matted with wattlework. A
distinctive feature of the Hodorf farm was that its hearth
was framed by four posts which were out of line with the
principal roof supports and also differed from the latter by
being round. They were obviously not part of the regular
structural system. Haarnagel thought that they might have
carried a smoke flue, or that they belonged to a separate
inner armature of poles which carried an elevated section
of the main roof over an opening in the ridge above the
hearth site, serving as light source and as smoke outlet. A
similar arrangement of poles ranged in a square around the
hearth had been observed in other Iron Age houses in
vastly distant places.[128]

Haarnagel has reconstructed the Hodorf house in a handsome
model which is displayed at the Niedersächsische
Landesstelle für Marschen- und Wurtenforschung in the
city of Wilhelmshaven (fig. 308). While many of the details
in the roof section of this model must by necessity remain
conjectural, the concept of the house as a whole is unquestionably
sound. The pottery found in the Hodorf house
indicated as time of occupancy the first and second century
A.D. Toward the close of the second century the site was
imperiled by tidal inundations. Its inhabitants made an
attempt to save the house by filling it up inside with sand,

[ILLUSTRATION]

312. WILHELMSHAVEN-HESSE, GERMANY

EXTERIOR VIEW, AISLED HOUSE, 6TH-9TH CENTURIES

MODEL IN THE NIEDERSÄCHSISCHES LANDESINSTITUT FÜR
MARSCHEN- UND WURTENFORSCHUNG

[after Haarnagel]

The excavation of Warf of Wilhelmshaven-Hesse (trial ditch in
1939 disrupted by World War II, resumed in 1949, continued in
1950) offered the first evidence that the aisled and timbered Iron
Age hall known through the excavations of Hodorf (figs. 307-308)
and Einswarden (fig. 309) continued to be used in the coastlands of
northern Germany in early medieval times. As it was excavated the
Warf revealed, in settlement horizons extending from the 7th through
10th centuries A.D., aisled and bay-divided houses ranging in
length from 39½ to 59 feet, and in width, 17½ to 21 feet.

The house model shown here is a reconstruction of one of the larger
houses of the Warf. Like those of Fochteloo and Einswarden, it had
an axial entrance for cattle in one of the narrow walls and a lateral
entrance to the living area close to the opposite end of the house.

a little more than 2 feet above its original floor level. A
number of posts were reset on this occasion, and the roof
may have been replaced entirely, but in all other respects
the house remained the same, except that now it was used
exclusively as a dwelling. It continued to be used in this
form until the end of the third century when it made room
for a new but smaller house of the same construction type.

EINSWARDEN, GERMANY

The operations at Hodorf had barely been completed, when
in the winter of 1937/38 Haarnagel was called to a site in
the vicinity of the village of Einswarden,[129] on the left bank
of the estuary of the Weser river, where the heavy machinery
of a modern land improvement project had edged into the
core of an ancient dwelling mound. Systematic excavations
were undertaken in the summer of 1938 but remained confined
to only a small sector of this large mound.

They brought to light three post-and-wattle houses of
the period around the birth of Christ and below these
dwellings, in an even earlier settlement horizon which
reached back to the second and third centuries B.C., four
additional houses of the same type. The largest of the
upper settlement measured 56 feet by 21 feet (17 m. ×
6·5 m.); the smallest, 33 feet by 16 feet (10 m. × 5 m.).
The latter, having its wood work practically intact to a
height of 16 inches (40 cm.), was especially well preserved.
Haarnagel could observe that the outer posts of house II
leaned inward. He assumed that the posts that he found
were the lower portions of rafters that rose from the ground
directly, and reconstructed the house accordingly.[130] Albert
Genrich[131] and Zippelius[132] consider it more likely that these
oblique outer posts were short, that they carried an outer
frame of horizontal poles that served as footing for the
rafters, and leaned inward in order to counteract the outward
thrust of the roof, as shown in figure 309 B.

AALBURG, NEAR BEFORT, LUXEMBOURG

With the outbreak of World War II, all of this excavation
ceased. Save for an isolated excavation conducted by Gustav
Rieck during the German occupation of Luxembourg at
Aalburg, near Befort,[133] nothing new was added to our
knowledge of the early history of the three-aisled timber
house. Rieck uncovered the foundations of an aisled timber
hall of extraordinary dimensions (102 feet long and 29 feet
broad [31 m. × 8·8 m.]) which antedated even the earliest
Ezinge houses (figs. 310, 311). Here, it seems, in a dwelling
that had been constructed as early as 500 B.C., in territory
where Celtic and Germanic influences intermingled, the
excavator had come upon a floor plan that anticipated by
one millennium the T-shaped Flet and Dele arrangement
of the Lower Saxon farmhouse. The roof-supporting posts
of this house were not sunk into holes but rose freely from
base blocks above the ground, attesting overhead a solid
frame of cross and long beams. This site raised the interesting
question, whether the aisled West-Germanic timber
house might not have been adopted at a very early date in
the territory of the neighboring Celts.

WILHELMSHAVEN-HESSE, GERMANY

When excavation work could be resumed after the war had
ended, Haarnagel added a number of excavations to his
preceding work, which enabled him to trace the history of
the aisled timber house both further back and further
forward in time. A trial ditch dug in 1939, just as the war
broke out, in Hesse, one of the suburbs of the city of
Wilhelmshaven, had suggested the presence, in a settlement
stratum of the seventh century A.D., of aisled houses of
the Hodorf-Einswarden type, such as he had previously
been able to assert only for the span of 300 B.C. to A.D. 200.
Systematic excavations undertaken in 1949 and continued
in 1950[134] surpassed all expectations by establishing the
existence of this house type in settlement layers not only of
the seventh, but also of the eighth and ninth centuries A.D.
And in 1951-53 this span was further extended into the
eleventh, the twelfth, and thirteenth centuries through the
excavation of a medieval trading settlement in the city of
Emden.[135] The result of these excavations is visually summarized
in a reconstruction model of one of the houses of
Wilhelmshaven-Hesse, here shown as figure 312.

JEMGUM, NEAR LEER, GERMANY

Conversely, in an excavation conducted in 1954 Haarnagel
had the good fortune of unearthing in a place called Jemgum
near Leer,[136] on the left bank of the river Ems, an
aisled house with pottery shards and artifacts ranging from
the beginning of the seventh to the end of the fifth century
B.C. (transition from Bronze Age to Early Iron Age.) The
walls of the Jemgum house (figs. 313-314) were a different
construction type from those of the previously discovered
houses. They were built of horizontal logs of ash, squared
off, and held in place by vertical ash saplings. In the middle
of each long wall there was an entrance protected by a
projecting porch. The roof was carried by four freestanding
inner posts of a diameter of eight inches (20 cm.), dug
sixteen inches (40 cm.) into the ground. The hearth lay in
the middle of the center aisle, in the northeastern half of the
house. On the opposite side of the house the ground was
covered by a wooden floor covering of alder planks, which
suggests that this section of the house was used as a living
and sleeping unit.

FEDDERSEN-WIERDE, NEAR BREMERHAVEN,
GERMANY

Haarnagel's most successful excavation—begun in 1955,
continued every subsequent summer, and still in progress
at the time of this writing—was undertaken on an Iron Age
Warf called Feddersen-Wierde on the right bank of the
river Weser not far from Bremerhaven. As this dwelling
mound was peeled off, layer by layer, it released the remains
of forty-eight houses; the majority were in excellent condition,
reflecting the various stages of growth of a settlement
that had started as a flatland farm at about the time
Christ was born, and was subsequently raised, in seven
stages, to successively higher levels, until around the year
400 it had reached an ultimate height of 13 feet (4 m.)
above its original starting point and a diameter of about
656 feet (200 m.). The results of this extraordinary excavation
are known so far through preliminary reports only.[137]
In figure 315 I reproduce a plan of settlement period IIB,
which shows the Warf in the stage it had reached sometime
during the first century. At this time the settlement consisted
of a principal Warf and a secondary smaller Warf,
both protected by a peripheral ditch. The principal Warf,
some 295 feet long and 98 feet wide (90 m. × 30 m.),
accommodated a cluster of four houses; the smaller, a
cluster of only two. The houses varied considerably in
size, the largest measuring 97 feet by 21 feet (29·50 m. ×
6·75 m.); the smallest, 33 feet by 16 feet (10·00 m. ×
5·00 m.) Each house formed a self-sufficient agricultural
entity, combining under one roof the living quarters of its
owner and the stables for his livestock (fig. 316). The hay
and harvest was stored in separate open sheds to the side of
the house. The layout of the main houses is identical with
that of the contemporaneous houses that van Giffen had
encountered at Fochteloo (figs. 303-304). Like them, the
houses of the Feddersen-Wierde had their principal entrance
arranged in opposite pairs in the long walls, giving
access to a crosswalk which separated the quarters of the
humans from those of the animals. In the smaller houses
where the areas of living quarters of the owner and the
stables for his livestock were more or less equal, this led to
a fairly balanced arrangement with the entrances often
exactly in the center. But in the houses of the leading
families, superior wealth in cattle led to an elongation of the
stables and to the addition in the latter of a subsidiary

[ILLUSTRATION]

C. ELEVATION

principal members only, shown; for more complete
assembly see exterior view on next page.

[ILLUSTRATION]

B. CROSS SECTION

[ILLUSTRATION]

A. PLAN

detail of plan at jamb of doorway

[ILLUSTRATION]

313.A. B, C
JEMGUM, LEER, GERMANY

AISLED HOUSE, 7TH-5TH CENTURIES B.C.

[redrawn from W. Haarnagel, 1957, 21, Plan. No. 7]

The site is on the left bank of the estuary of the river Ems. The house was small, no more than 15 feet wide and 25 feet long. It was used
exclusively as a dwelling and gave no evidence of ever having sheltered animals. In the middle of each long wall, slightly off center, were
opposing entrances protected by projecting porches.

The roof was carried by two pair of inner posts (unscantled oak trunks, dia. 20cm) dividing the house into a central area of roughly 6½ × 13
feet asymmetrically placed, and with aisles all round it. The hearth lay in the axis of this center space, in the western half of the house which
had a simple clay floor and must have served as kitchen.

The floor of the space between the eastern pair of posts and the eastern end wall was covered with wooden planks cut from alder trees; this
area, better insulated from dampness than any other in the house, must have served as living and sleeping quarters. All structural members of
the dwelling that were posted into the ground, or that lay atop the ground, were found to be in good condition, many of the boards forming the
wooden floor of the presumed sleeping area were still in place.

[ILLUSTRATION]

314. EXTERIOR VIEW

JEMGUM, LEER, GERMANY

AISLED HOUSE, 7TH-5TH CENTURIES B.C. [redrawn from W. Haarnagel]

Drawings and models are in the Niedersächsisches Landesinstitut für Marschen- und Wurtenforschung

The walls of the house were made of squared ash logs, of which the bottom course was still well preserved. They were held in place at distances
varying between 5 and 6½ feet, by paired saplings pointed and driven into the ground to a depth of about 2 feet, with five pairs in each long
wall, and three in each end wall. At their meeting points in the corners of the house, the ash logs had rotted away, and for that reason, it could
not be ascertained in what manner they were jointed. It seems reasonable to assume that they were notched into each other at right angles,
since otherwise these timbers would have been subject to displacement from the thrust of the rafters.

Since the free-standing inner posts were only set 15¾ inches into the ground, they must have been framed crosswise at their heads by tie beams,
and lengthwise by longitudinal plates serving as footing for the rafters, or supporting them in midspan. There were no roof-supporting posts in
the end walls, indicating that the roof was probably hipped over the building's narrow ends. The construction of the walls, although common in
heavily wooded areas of Scandinavia and Alpine regions, is atypical for this part of Europe.

[ILLUSTRATION]

315. FEDDERSEN-WIERDE. PLAN [after Haarnagel, 1957, fig. 2]

AISLED HOUSES OF WARF-LAYER II B, 1ST-2ND CENTURIES

Haarnagel's exploration of this Warf, conducted from 1955 onward under the auspices of the Deutsche Forschungsgemeinschaft, was the
German counterpart to van Giffen's excavation of the Warf of Ezinge (figs. 292-99). The site, on the right bank of the estuary of the Weser,
was carefully selected after many sample drillings from a chain of nine dwelling mounds running in an almost straight line south to north over a
distance of 15 kilometers. The Warf encompassed seven settlement horizons, a new one every 50-80 years, to compensate for the steadily rising
innundation level.

The earliest settlement was a flatland farm built around the birth of Christ. The Warf was abandoned around 400 A.D. when it had reached
a height of about 13 feet (4m). The dwellings buried in its various layers were as well preserved as those of Ezinge and for the same reasons
(see caption, fig. 299); and were of the same construction type. The plan above shows the Warf in the stage it had reached toward the end of
the first century A.D.

[ILLUSTRATION]

FEDDERSEN-WIERDE, NEAR BREMERHAVEN, GERMANY

316.B AUTHORS' RECONSTRUCTION (DRAWN BY WALTER SCHWARZ)

SEE FIGURE 175. PAGE 216, VOL. 1, FOR A LARGER INTERPRETATION OF THIS DRAWING

316.A PLAN [after Haarnagel, 1956, Pl. 3]

AISLED HOUSE OF A CHIEFTAIN, Warf-LAYER II B, 1ST-2ND CENTURIES

With the cattle barn of Ezinge (figs. 298-99) this is one of the finest examples of a house type widely diffused in the Germanic territories of
Holland and Northern Germany during the first millenium B.C. and throughout the entire Middle Ages. The house was 97 feet long (28.50m)
and 21 feet wide (6.75m). It combined under one roof the owner's living quarters and the stables for his livestock. In the area used by animals
(eastern 52½ feet of the house) the roof-supporting trusses were more narrowly spaced, leaving in the aisles between each pair of posts a stall
for two head of cattle (32 head altogether).

As in the chieftain's house at Fochteloo (fig. 304) stables and living area were separated by an entrance bay accessible through doors in the long
walls, while animals entered through a gate in the eastern end wall. The walls and all the internal cross partitions were done in wattlework,
daubed with manure. The stable area had the traditional mats of wattlework on which the manure was gathered, with cess trenches beneath to
allow for drainage. The walk between these mats was paved with turves laid between floor beams running parallel with the trenches.

[ILLUSTRATION]

317. FEDDERSEN-WIERDE, NEAR BREMERHAVEN, GERMANY

HOUSE OF THE EARLIEST SETTLEMENT HORIZON. DETAIL

[photograph by courtesy of W. Haarnagel]

These are remains of two of the principal roof-supporting posts of the house, for which the builder used the round trunks of relatively young
and slender oaks without debarking them. The superb state of preservation of both timbers and the wattlework of which the walls and cross
partitions in the aisles were formed owes to the fact that whenever a house was abandoned because of floods and then rebuilt on higher ground,
its remains were soon covered by layers of fine silt deposited during floods, thus sealing its contents against air and bacterial decay.

axial entrance, primarily used for livestock. The house thus
attained the distinctive T-shaped floor plan which later
became the hallmark of the Lower Saxon farmhouse. The
long house in the northwest corner of the main Warf of
settlement period II-B of the Feddersen-Wierde is one of
the finest of this type of Iron Age house known to date. In
figure 316A I reproduce its plan, after Haarnagel, and in
figure 316B a tentative reconstruction of my own. The excavation
photo shown in figure 317 of one of the cattle boxes
of house I of the oldest settlement horizon of Feddersen-Wierde,
gives an idea of the magnificent state of preservation
in which the walls and roof-supporting posts of some
of the older houses of this site were found.

The occupants of settlement-horizon II of the Warf
Feddersen-Wierde were field-ploughing and cattle-raising
farmers. In settlement-horizon III (first to second century
A.D.) the economy, and with it the entire social structure of
the village, begins to change. The dominant architectural
feature now, as well as in all the subsequent horizons (IV,
V, VI, and VII, ranging from the third into the fifth
century A.D.), is a large aisled hall (without stalls for cattle
and carefully fenced in), used as the residence of a person
of conspicuous wealth and prominence. Next to this hall is
a second hall (likewise without cattle stalls) which Haarnagel
believes was used as an assembly place for the entire
community. Animal husbandry and agriculture give way
to industry and trade, and the growth of a new class of
workmen who lived in smaller houses and worked in the
service of their trading chieftain.

[ILLUSTRATION]

ELISENHOF, NEAR TÖNNING, SCHLESWIG, GERMANY

319.

320.

AISLED HOUSE, 9TH CENTURY A.D. [excavation photos courtesy of A. Bantelmann]

The overview (fig. 319) of the Warf shows the remains of the houses; below (fig. 320), the detail shows a portion of the wattled walls of the
house with inclined posts carrying a peripheral course of poles on which the rafters were footed.

The great historical significance of the excavation of this Warf is that it closed the gap between the Iron Age and Migration Period houses
(shown in figs. 293-318) and their medieval derivatives (figs. 339-354). The settlement was started on flatland in the 7th century; its subsequent
development could be traced clear into the 11th century. In layout and construction its houses were virtually identical with those of Ezinge
(figs. 293-299), and Feddersen-Wierde (figs. 316-317). They were in some places preserved to a height of 7 feet.

[ILLUSTRATION]

321. ANTWERP, BELGIUM. UNAISLED AND AISLED HOUSES

PLANS, EARLY 11TH CENTURY A.D.

[after A. Van de Walle, 1961, 128, fig. 35]

The house plans and reconstruction shown here and in fig. 322 are in themselves of no particular architectural distinction. But they do mark the
historical point at which the aisled and bay-divided timber house, the premedieval history of which has been briefly traced in these pages,
attempted to gain a hold in the new and rapidly developing medieval cities.

Excavations conducted in 1955-1957, in what was then the old city of Antwerp (and is now the center of the modern town) brought to light
three medieval habitation levels in an average depth of 5 to 7 feet (1.50-3.50m) beneath the present street level. By pottery and other artifacts
these strata could be dated: the lowest to about 850-976, the middle to about 976-1063, and the top level to 1063-1225. On each horizon the
excavator found three houses in a row, side by side, gable walls facing the street. The houses shown here belong to the middle level. The larger
one to the right is aisled; the others, narrower and shorter, are unaisled. These two are divided internally into a main hall with hearth, and with
one or more partitions to the rear perhaps serving as private or storage rooms.

Aisled houses were well suited to the open terrain of the nonurban countryside. But in the densely built cities, with open land at a premium, the
aisled structure of one story had limited utility and future. Some wealthy individuals or institutions could, to be sure, acquire enough urban land
upon which to build expansive aisled houses on one level, and could afford the expense of their maintenance. Such was the case with ecclesiastical
overlords (see figs. 339-340) or corporate bodies such as the Church (figs. 341-343) or the guilds. But for the most part, aisled dwellings were
impractical in, and proved antithetical to the function of the city. The type came to be replaced by narrower structures of multiple stories
providing space above ground level, set with gable walls toward the street and side walls almost touching, a picturesque and early characteristic
of new urban architecture.

[ILLUSTRATION]

322. ANTWERP, BELGIUM

AISLED HOUSE, EARLY 11TH CENTURY

[the reconstruction illustrated is redrawn from A. van de Walle, 1961, 129, FIG. 36]

This isometric rendering, conjectural in detail yet fairly
certain in general lines, illustrates more persuasively than
the plans of the preceding figure why it was that aisled
houses could not survive the pressures of dense urban
development. The low, aisled house of the open country, in
the struggle to adapt it to urban row-house conditions,
soon proved to be a wasteful use of costly and limited
city space. Therefore this house type was, in the cities
quickly discarded.

In process of adaptation, the remaining nave (after
aisles were eliminated) could, to be sure, have been raised;
but the skeletal construction of the old northwest European
all-purpose house was never intended to bear the load
of superincumbent stories. A new type of timber framing
with strong load-bearing walls evolved to make timber
framing possible in construction of narrow urban houses.
But because of its total vulnerability to fire, the timber
house eventually came to give way, as the cities grew, to
masonry houses.

BÄRHORST, NEAR NAUEN, GERMANY

To the successful excavation work by van Giffen and
Haarnagel in the coastlands of Holland and northwestern
Germany, one has to add the work of others. As early as
1935-37 Otto Doppelfeld had unearthed a palisaded village
of an estimated fifty aisled houses, on the Bärhorst,[140] a
shallow, sandy plateau in a marshy swale near Nauen
(Berlin). The site had been discovered in the course of
trenching operations undertaken before the installation of
a giant sewage disposal plant. Remnants of pottery and
other cultural accessories showed that the village was
constructed around A.D. 250 and that it was held in occupancy
for about a century. Since it lay in an environment
that was utterly unsuited for successful agricultural exploitation
and perished in a fire that seems to have been
associated with a planned and systematic abandonment (no
objects of any use were left), Doppelfeld concluded that it
might have been the temporary site of a wandering Germanic
tribe who discarded the site when they found prospects
for the conquest of more suitable land. While basically
adhering to the same construction type, the Bärhorst houses
showed a great variability in the treatment of their walls
(fig. 318). Some of the houses had simple wattle walls; in
others the walls were formed by boards mounted horizontally
or vertically between the wall posts. Still others
were braided from thin split pieces of straight wood (leftovers
from the hewing of the structural timbers). The
Bärhorst village showed that the aisled pre-medieval timber
house extended into the third and fourth century A.D. eastward
as far as the longitude of the modern city of Berlin.

TOFTING AND ELISENHOF, NEAR TÖNNING,
SCHLESWIG-HOLSTEIN, GERMANY

Albert Bantelmann, on the other hand, pressed the search
northward by excavating, in the summers of 1949 and 1950,
a dwelling mound at Tofting (Schleswig-Holstein)[141] at the
mouth of the river Eider, close by the Danish border. His
excavations showed that conditions in the homeland of the
Anglo-Saxons were identical with those which van Giffen
and Haarnagel had found prevalent in the adjacent territories
of the Frisians.

Tofting was a relatively modest site; but from 1957
onward, in annual excavations as yet not terminated,
Bantelmann peeled off, layer by layer, in the Warf Elisenhof,[142]
near the town of Tönning at the mouth of the
river Eider in Schleswig, the remains of a village that was
founded in the seventh or eighth century A.D. and remained
in continuous occupation deep into the High Middle Ages.

The earliest settlement, which is so far known only
through a preliminary report, was built on natural ground
on one of the banks of the river Eider. Later the site was
raised and peripherally expanded by heavy deposits of
manure and clay, until it finally comprised an area of
roughly seven hectares. The occupants of the earliest settlement
were cattle-raising farmers, which is attested by the

[ILLUSTRATION]

CHEDDAR, SOMERSET, ENGLAND

AISLED PALACE HALLS, 12TH & 13TH CENT.

[courtesy of P. Rahtz, and Ministry of Building and Works, Crown
Copyright]

324.A

The composite plan shows the layout of East Halls I, II, and III. The original ten-bay
building dates to the early 12th century. It was replaced 100 years later by a smaller
six-bay hall 71 feet by 48 feet. At this time the roof supports were set into new square
post holes, some of which overlapped the round ones of the original hall. Toward the
end of the 13th century the second hall was replaced by a third of yet smaller size
66 by 42 feet) and without aisles.

324.B

At the time of Henry I (1100-1135) the hall was an aisled, ten-bay structure, 110 feet
long and 54 feet wide. The course of the outer walls could be identified by large post
holes, at 8-foot intervals, linked by ground timbers. The roof-supporting posts were
apparently rough-scantled logs over 15 inches in diameter, set into round post holes
packed with gravel.

presence of bone deposits of cattle, sheep, pigs, and horses
in, respectively, decreasing magnitude. The predominant
dwelling in all layers was the aisled post-and-wattle house,
giving shelter to humans and animals under the same roof.
Layout and construction were in all essential features
identical with those of the houses of Ezinge and Feddersen-Wierde,
and the state of preservation left no feature in
doubt (figs. 319, 320). Main posts and wattle walls survived
in certain cases up to a height of 7 feet (2 m.). In some
houses the outer posts consisted of inward leaning timbers
of unusually heavy scantling (fig. 320)—no doubt the supports
of a peripheral course of poles that served as footing
for the rafters. The floors of the houses were formed by
turves of clay, heavily matted with roots. The section of
the house that contained the hearth and served as living
quarters invariably lay on a higher level than the part that
contained the stalls for the cattle. From the higher level
the floor gradually slanted down to reach its lowest point
at the end of the stable section, thus affording easy drainage
for the liquid waste of the animals, which was conducted
downward in carefully constructed flues—the same type of
flues (one groundboard and two sideboards) used today
in the farmhouses of the same district, where it is called
Grüpp.

The length of the houses varied with the number of
cattle owned by each farmer. The width amounted uniformly
to about 17 feet (5 m.)—as it did in van Giffen's and
Haarnagel's early Iron Age houses—a dimension obviously
conditioned by the fact that it offers a comfortable minimum
of space for two rows of animals and a central lane of
access with drains for the waste products. As in the Iron
Age houses, the roof received its main support from two
ranges of freestanding inner posts. The cattle stood in
pairs in each stable, their heads turned toward the walls of
the house. The cross partitions by which their boxes were
formed consisted of split logs set into the ground in palisade
fashion. The outer walls were wattled, with the twigs
wound around a sequence of thin posts alternating at
regular intervals with heavy posts which must have supported
the wall plates. Only a small percentage of the
houses was oriented from east to west. The determinant
factor in the choice of the axis appears to have been the
slope of the Warf, as it offered best drainage.

Besides the standard house there was a variety of non-aisled
smaller buildings, some with wattle walls, others
with walls of turf; the latter, apparently used as weaving
houses. Two of these smaller houses were in ridge-pole
construction.

The Elisenhof is altogether a spectacular site. Its full
evaluation, which is forthcoming, will unquestionably give
us new insights into the constructional aspects of its houses.
The same may be expected from the excavation of a ninth-century
village on the Grothenkamp near Neumünster in
Holstein, which came to light in an emergency dig undertaken
in 1962.[143] Also not to be overlooked, in this connection,
are three houses (one of them aisled) which A. van de

Walle excavated in 1955 in the ancient town center of
Antwerp in a settlement-horizon that could be dated in the
eleventh century.[144] I reproduce in figures 321 and 322 a
plan and a reconstruction of this house as proposed by van
de Walle.

The great importance of these settlements, together with
those of Leens, Emden, and Wilhelmshaven-Hesse, is
that they have helped to close the gap between, on one
hand, the Iron Age houses of Ezinge, Jemgum, and Befort
and, on the other, isolated medieval house sites found in
Germany in such places as Wilhelmshaven-Krummer Weg
(eleventh-twelfth centuries), Ramm (thirteenth century),
Hungersdorf (about 1400), Hardesbüttel (thirteenth-fifthteenth
centuries);[145] and in Holland in such fourteenth-century
sites as Lievelde, Waalhaven, and in Boudewijn
Hartsland.[146] Thus, pre- and protohistory were connected
to the modern period with the story of a house type whose
historical life span, it now seemed, would have to be counted
by millenniums rather than by centuries. In this matter,
too, the last six years have brought a sensational surprise.

ELP, PARISH OF WESTERBORK, THE NETHERLANDS

After Haarnagel's excavation at Jemgum (figs. 313-314), it
was generally believed that the aisled hall had been
traced back to the period of its earliest appearance (seventhfifth
centuries B.C.). However, a Bronze Age settlement in
Elp (parish of Westerbork, province Drenthe, Holland)
excavated in 1960-62 by H. T. Waterbolk[147] disclosed that
the aisled long houses of the Ezinge-Feddersen-Wierde
type were in use as early as 1250 B.C. The excavation brought
to light the ground plan of some thirty houses of different
types belonging to a farmstead composed of a main building
and about four subsidiary buildings, all of which were rebuilt
on various occasions during a period of occupation
that lasted from roughly 1250 to roughly 850 B.C. The main
houses vary in length between 82 feet (25 m.) and 135 feet
(41 m.). They are internally divided into two equal parts.
In one half, which was used for living, the posts are
widely spaced. In the other half, which served as stables,
the interstices between the posts are smaller. Some of them
are not sunk into the ground as deeply as the principal
posts and therefore, probably, served as mainstays for stall
partitions. I show as a typical example the ground plan of
house 9 (fig. 323), which illustrates this point particularly.
A similar distinction of the post interstices between living
quarters and the stall sections of the house could be observed
in Feddersen-Wierde (figs. 315-316), in Hodorf
(fig. 307), in two Iron Age houses excavated in 1954 by
P. J. R. Modderman near Deventer (Oberijssel), Holland,[148]
and—although not quite as markedly—in Fochteloo (fig.
304). As in the later Iron Age houses, so in Elp, the narrow
ends of the house were rounded, which suggests that the
ends of the roof were hipped. The houses appear to have
been entered broadside through a passage way that separated
the living quarters from the section that was occupied
by animals—another feature that was to become a characteristic
trait of the later house tradition. Waterbolk had
reason to believe that the house type of Elp already existed
several centuries before the settlement of Elp was founded;
and before the manuscript of his preliminary report on Elp
was finished, he received the news that J. D. van der Waals
had come across a Bronze Age site with an aisled house
118 feet (36 m.) long at Angeloo (Emmen). It had many
features in common with the Elp houses, and, to judge from
its pottery, appeared to be older than the Elp settlement.[149]

The Elp settlement was occupied by an autochthonous
Bronze Age population of Holland,[150] which may or may not
be proto-Germanic. The later Iron Age sites of northwest
Germany and Holland were in the territory of the Frisii,
the Chauci, and the Saxons.[151] The homeland of the Saxons
was east of the river Elbe at the bottom of the Danish
peninsula. Their westward move into the territory of the
Frisians, around 300 A.D., seems to have led to the destruction
of the settlement of Ezinge. Toward the middle of the
fifth century, in several successive waves, they moved
across the channel into England.

YEAVERING, NORTHUMBRIA, AND CHEDDAR,
SOMERSET, ENGLAND

Until the summer of 1956, England, peculiarly enough, had
not yielded a single house site comparable to any of the
Continental finds. A small settlement, of aisled post-and-wattle
houses excavated in 1946 by Gerhard Bersu on a
promontory of the shore of Ramsay Bay on the Isle of
Man,[152] was probably not an Anglo-Saxon settlement but
an outpost of a Norse raiding party which was occupied
only intermittently. In the summer of 1956, however,
Brian Hope-Taylor came upon the site of a royal Anglo-Saxon
palace of the seventh century at Old Yeavering in
Northumbria (Bede's villa regalis ad Gefrin),[153] which contained
the remains of no fewer than twenty-five timbered
houses, most of them aisled; and in 1960-62 Philip Rahtz

[ILLUSTRATION]

WARENDORF, WESTPHALIA, GERMANY

325.A EXTERIOR VIEW REDRAWN FROM WINKELMANN

325.A.1 PLAN. HOUSE 47, LEVEL A, PERIOD 4

HOUSE TYPES OF EARLY MEDIEVAL SETTLEMENTS. 650-800 A.D.

[after Winkelmann, 1958, 500, fig. 5]

Excavations conducted in 1951 by Wilhelm Winkelmann in Warendorf, on the south side of the Ems, brought to light traces and remains in the
soil of no fewer than 186 separate buildings, which could be accurately dated by pottery and other associated artifacts. The settlement was
rebuilt four or five times, in most cases—as the remains of charcoal and fired clay found in many postholes indicate—after houses of the previous
settlement had been destroyed by fire.

Analysis of the successive stages of the site disclosed that each inhabited level consisted of a group of four to five farmyards occupying an area
of about 984 feet (100m) square. Each farmyard held a large dwelling, and fourteen to fifteen smaller auxiliary buildings (barns, stables, sheds,
weaving houses). The variety of these service structures is shown in figs. 326.A-F.

The reason for the boat shape of House 47, a feature very common in early Scandinavian architecture, is obscure. For other examples of
longhouses of this type see below, V. 17. 3.

[ILLUSTRATION]

WARENDORF, WESTPHALIA, GERMANY

325.B EXTERIOR VIEW REDRAWN FROM WINKELMANN

Arrow indicates general direction from which exterior view above is taken

325.B.1 PLAN. HOUSE 7, LEVEL A, PERIOD 1

HOUSE TYPES OF EARLY MEDIEVAL SETTLEMENTS, 650-800 A.D.

[after Winkelmann, 1958, 504, fig. 8]

Plans and reconstructions on these pages are typical of the man or principal dwellings of the Warendorf settlements. By virtue of both size and
architectural distinction, as well as the presence of a hearth, they are unquestionably the dwellings of the owners of the land. They were boat-shaped
(fig. 325.A) or of rectangular plan, as above, varying in length between 45 and 95 feet (14-29m), in width between 14½ and 32 feet
(4.5-7.0m). Their roofs were supported by a perimeter of heavy posts set vertically into the ground, each buttressed from outside by a second,
lighter post rising at an angle to meet the inner post near its head. The triangulation counteracted the thrust of the roof. Wall panels between
vertical posts were daubed with clay.

The axes of these dwellings, and most of the subsidiary structures, ran from west to east. This feature, characteristic of many pre- and
protohistoric buildings in these latitudes, was apparently determined by the desire to expose only a gable wall to the prevailing (here, western)
winds. The houses were entered on their long walls through two opposing entrances in midwall, each protected by a porch. The hearth lay in the
eastern half of the house.

[ILLUSTRATION]

WARENDORF, WESTPHALIA, GERMANY

326.A

326.B

326.C

HOUSE TYPES OF EARLY MEDIEVAL SETTLEMENTS, 650-800 A.D.

[redrawn after Winkelmann, 1958, 504, fig. 8]

These pages give a visual account of the variety of service structures that one might have expected to find in the farmyards of Warendorf.
They have been redrawn from models made under the direction of Wilhelm Winkelmann, and are on display at the Museum für Vor- und
Frühgeschichte, Münster, under the aegis of which the excavations at Warendorf were conducted.

There were forty service buildings of rectangular plan with vertically boarded walls at Warendorf (326.A), used as barns and stables. Small
houses with hearths built in the manner of the larger dwellings presumably provided housing for serfs (326.C). Numerous small sheds with open
walls (326.B) were probably utility buildings for various kinds of storage.

[ILLUSTRATION]

WARENDORF, WESTPHALIA, GERMANY

326.D

326.E

326.F

HOUSE TYPES OF EARLY MEDIEVAL SETTLEMENTS, 650-800 A.D.

[redrawn after Winkelmann, 1958, 504, fig. 8]

The small, walled rectangular shed (326.D) could have served many needs from implement storage to animal shelter. A great number of small
buildings of A-frame construction (326.E) were found in the Warendorf farmyards. They were partly dug into the ground to gain standing
height, and some buildings of this type were identifiable as weaving houses by the artifacts they contained. But a building with below-ground
storage could have served equally well for winter storage of root crops.

The polygonal framework was used to store loose hay or sheaves of grain (326.F); its conical thatched roof could slide down its poles to
accommodate and adequately shelter the diminishing harvest as it was used up through the winter. The floor grid of lashed poles (326.X)
afforded aeration and drainage for the hay or grain; the plans show typical posting patterns for the structures.

[ILLUSTRATION]

326.X HAYSTACK SHELTER, left, and
DRYING PLATFORM, right

Haystacks and aeration platforms of this
type are common in Germany and the
Netherlands even today. For a fine
15th-cent. portrayal, see fig. 368.

[ILLUSTRATION]

326.G WARENDORF, WESTPHALIA, GERMANY

HOUSE TYPES OF EARLY MEDIEVAL SETTLEMENT, 650-800 A.D.

[redrawn from Winkelmann, 1958, 500, 504]

Each of the buildings shown on this and the preceding pages is referred to in the Alamannic, Bajuvarian, and Salic laws discussed above
(p. 26ff), where they are designated in the following manner:

The house of a free man = DOMUS or SALA LIBERI (figs. 325A-B); the house of a serf = DOMUS SERVI (fig. 326.C); a wall-enclosed
barn = SCORIA (fig. 326.A); an unenclosed barn = SCOF (fig. 326.B); a granary = PARC (fig. 326.D); a haystack, or stacked sheaves of
wheat = MITA (fig. 326.F). For a more complete account and the etymology of these terms see above, p. 29, notes 14-18, and the chart in
Winkelmann, 1954, 210, fig. 12.

unearthed at Cheddar, Somerset,[154] another Saxon palace
(sedes regalis aet Ceodre) with the remains of a timbered
long hall from the time of King Alfred (871-900)[155] and an
aisled hall, likewise in timber, from the reign of Henry I
(1100-35), a plan of which is reproduced in figure 324. The
results of these excavations once fully published will put
an end to another enigmatic chapter of early medieval
house research.

THE SINGLE-NAVED HOUSE OF
WARENDORF, NEAR MÜNSTER, WESTPHALIA,
AND IN OTHER GERMAN SITES

If I have given primary consideration to aisled structures
in the preceding account of prehistoric and early medieval
house construction, I have done so because the excavations
conducted during the last three decades seem to indicate,
with mounting conclusiveness, that this was the principal
dwelling type used during the Iron Age and in the Early
Middle Ages in the barbaric territories north of the Alps.
In emphasizing this fact, I do not wish to convey the
impression that it was the only one. Other excavations,
recently conducted, have brought clear evidence of the
existence of a simpler type of house that was not provided
with any aisles.

A good sampling of this latter type was brought to light
between 1951 and 1954 when Wilhelm Winkelmann excavated
a medieval settlement in the vicinity of Warendorf,
near Münster, Westphalia.[156] The pottery found in this
settlement suggested that it was occupied from about 650
to 800 A.D. The leading house type was a structure of
rectangular plan, accessible by two porch-surmounted entrances
facing each other in the middle of the long walls
(figs. 325 A-B and below, pp. 271ff). Similar houses were
encountered in Bucholtwelmen (district of Dinslaken) in
1939, in Haldern (near Wesel, Lower Rhine) in 1938, and
in Westick (district of Unna, Westphalia) in 1935—the
latter a rare example of Continental cruck construction.[157]
Warendorf was of particular interest because on this site
the principal dwellings were surrounded by an entire host
of subsidiary structures of smaller dimensions and lesser
significance, the like of which one would expect to encounter
in all those settlements where the various domestic functions,
rather than being assembled under one roof, are
scattered throughout a variety of separate structures.
Winkelmann has gathered all these types together in a

visual chart which reads like a model book of early medieval
house construction (figs. 325 and 326). He had found an
equivalent for virtually each and every variant mentioned
in the Alamannic, Bajuvarian, and Salic laws.[158]

LACK OF COMPARABLE FINDS IN FRANCE
AND IN THE SOUTH OF GERMANY

The great shortcoming of our present knowledge of
early medieval house construction in transalpine Europe
is that it is based on the results of excavations confined to
the Scandinavian countries and to the northern parts of
Holland and Germany, with the recent addition of Anglo-Saxon
England. France, to this day, has remained a terra
incognita. Bursting with treasures of unsurpassable beauty
created in more advanced and more sophisticated periods
as well as in more permanent materials, she is unlikely
to engage in the near future in any concerted search
for the shadows that the rotting timbers of her humbler
early medieval houses left in the subsoil underneath the
stately structures that replaced them. But even in southern
Germany the soil has as yet been rather unyielding. An
excavation of a Frankish settlement of the sixth to ninth
centuries A.D., made as early as 1937 in Gladbach (district
of Neuwied),[159] brought to light a great variety of smaller
subsidiary structures, all in post-and-wattle work—but no
houses of any primary significance. The same holds true
for an Alamannic settlement of the Early Middle Ages,
excavated in 1947 in the vicinity of Merdingen (district of
Freiburg),[160] and for a Bajuvarian early medieval village
near Burgheim (district Neuburg on the Donau.)[161]

The reason for this scarcity of finds on the mainland is
probably very simple. The dwelling mounds of the coastal
lowlands which yielded such rich information are not only
conspicuous scenic landmarks, but also of a physical
composition (marsh, clay, manure—sealed off by intermittent
layers of silt) that offers unusually favorable conditions
for the preservation of wood, an advantage that is
otherwise only encountered in peat bogs or in sites that lie
below the normal water level. On the mainland these
conditions, in general, are wanting.[162]