CARPETS.
It is not my intention in this chapter to consider in detail the
various kinds of carpet which are common in our market, nor even to
review the history of their manufacture, interesting as it would be to
do so; for we must confine ourselves more particularly to an
examination of the art-qualities which they present, and to the
particular form of pattern which may be applied to them with
advantage.
Although we cannot here enter into a consideration of the manufacture
of carpets, I cannot too strongly recommend all who intend preparing
designs for them to consider minutely the powers of the carpet loom;
for the nature of the effect produced will depend to a large extent
upon the knowledge which the designer possesses of the capabilities of
the manufacture for which he designs patterns. In the case of any
manufacture it is highly desirable, if not absolutely essential, that
the designer of the patterns to be wrought should be acquainted with
the process by which his design is to be converted into the particular
material for which the pattern has been prepared; for this knowledge,
even when not absolutely essential, gives an amount of freedom and
power which nothing else can supply.
The carpets most extensively in use are "Brussels;" but there are many
other kinds both of better and inferior qualities. "Kidderminster
carpet" (a carpet not now made by even one Kidderminster manufacturer)
is a common fabric suited to the bedrooms of middle-class houses; but
the art-capabilities of this material are very small, as it can only
have two colours in any line running throughout its length. This
carpet consists of two thicknesses, which are imperfectly united, and
is not durable. "Brussels carpeting," now made chiefly in Great
Britain, is a good carpet for general purposes. Its surface consists
of loops, and it may have five, or, if made of extra quality, six
colours in any line running throughout its length. If with five
colours in the same line the carpet will, in a sense, consist of five
thicknesses of worsted; yet these are united into one fabric. In some
cases a "Brussels carpet" is woven of very close texture, with the
loops cut through; thus we have a "velvet pile" or "Wilton carpet"--a
fabric which is very rich-looking, and durable.
Those called real "Axminster" carpets are, perhaps, the best made.
They are formed by the knotting together of threads by hand,
consequently any number of colours may be used in their formation; but
such are necessarily most costly. A "patent Axminster" carpet is made
by a double process of hand-weaving, by which fine results are
achieved, and any number of colours used. In the first weaving a rough
"cloth" is formed, which is cut into strips called "chenille threads,"
and these are again woven into the carpet. This process is most
ingenious, and the carpets produced by it are very good; but they are
costly.
Some few years since a most ingenious process of manufacturing what
are known as "tapestry" carpets was patented--a process resembling in
its nature that of the patent Axminster manufacture, but differing in
this particular, that the "warp" threads are coloured by printing, and
thus the first process of weaving is dispensed with. These carpets
are, like Brussels, made with a looped surface, and also with a pile.
They cannot be said to compare in any way with the patent Axminster
carpets, which are of a pretentious and costly character, nor even
with a good "Brussels;" but they are low in price, and meet a want, as
is proved by their enormous sale.
Besides these varieties of carpet there are a number of kinds of
foreign production, most of which are hand-made, and are very
beautiful. By far the greater number of these have a "pile," although
this is sometimes rough and uneven, yet rarely, if ever, inartistic;
but a few are without pile; still these are not without that
indescribable something which renders them estimable in the eye of an
artist.
Having hastily noticed the chief kinds of carpet in use in this
country, and we might say in almost all countries, we come to the
question--what form of pattern, or what character of ornament, should
form the "enrichment" of such a fabric?
When speaking in a previous chapter (see page 92) of wall decorations,
we noticed that a wall-paper pattern, or, indeed, a wall pattern of
any kind, might desirably have an upward direction and a bilateral
symmetry. This can never be the case, however, with a carpet pattern,
which must be equally extended all over the surface, or have a simple
radiating symmetry, as Fig. 56; and this rule will apply whether the
pattern be simple or complicated. It is not wrong, as we have said
before, to have a radiating pattern on a wall, but it is wrong to have
a bilateral pattern on a floor.
The reason of this is obvious. If such an object as we have indicated
is placed on a wall, from whatever point the occupants of the room may
view it, it is yet right way upwards to them; but if such an object
were placed on a floor it would be wrong way upwards, or sideways, or
oblique to most of those who viewed it; and to employ a pattern of
this character in such a position is highly absurd, when a pattern can
as readily be formed which will avoid this unpleasantness. What would
we think were we asked to view a picture, or even to visit an
apartment containing such, were this work of art presented to our view
in an inverted manner? We should feel astonished at the absurdity; yet
this would be no worse than expecting us to view a carpet while the
pattern is to us in an inverted position.
And the principle which we have just set forth is one taught by a
consideration of plants. If we wander over the moor, where we tread
on Nature's carpet, we find that all the little plants which nestle in
the short mossy grass are "radiating ornaments"--that is, they are
pretty objects which consist of parts spreading regularly from a
centre.
[Illustration: Fig. 70.]
[Illustration: Fig. 71.]
[Illustration: Fig. 72.]
[Illustration: Fig. 73.]
I cannot too strongly advise the young ornamentist to study the
principles on which Nature works. Knowledge of the laws which govern
the development of plant-growth is very desirable; but it is not our
place to _imitate_ even the most beautiful of plant-forms--this being
the work of the pictorial artists. Yet it is ours to study Nature's
laws, and to observe all her beauty, even to her most subtle effects,
and then we may safely pillage from her all that we can _consistently_
adapt to our own purposes. But in order that we produce ornament, we
must infuse mind or soul into whatever we borrow from her. (See page
2.)
With the view of more fully impressing the manner in which Nature
teaches us principles which we may apply in art, and of aiding the
student in his inquiries, we will give one or two illustrations. Thus
Fig. 64 is a drawing of a spray of the guelder rose (_Viburnum
opulus_) when seen from the side, or, as I might express it, when
viewed as a wall decoration; and Fig. 70 is the same spray as seen
from above, or, to use the same manner of expression, when seen as a
floor pattern. Further, Fig. 71 represents a young plant of a species
of speedwell (_Veronica_) as a wall ornament, and Fig. 72, the same
plant when seen as a floor ornament; and Figs. 65 and 73 represent a
portion of the goosegrass (_Galium Aparine_) as seen in the same two
views.
[Illustration: Fig. 74.]
[Illustration: Fig. 75.]
[Illustration: Fig. 76.]
From these illustrations we see that plants furnish us with types of
two essentially different ornaments, which are adapted to the
decoration of the two positions of wall and floor, and may be
introduced with truthful expression and effect into wall-paper or
carpet.
[Illustration: Fig. 77.]
[Illustration: Fig. 78.]
[Illustration: Fig. 79.]
Even when the leaves appear somewhat dispersed upon the stem, a
principle of order can yet be distinctly traced in the manner of their
arrangement, as is diagrammatically expressed in Figs. 74, 75, 76; and
here, also, the top view gives us a regular radiating ornament.[25]
[25] The spray here represented is that of the oak, and the diagram
(Fig. 74) shows the orderly spiral manner in which the leaves spring
from the stem.
The same law prevails in the flower that we have traced as existing in
the arrangement of leaves upon the stem: thus Fig. 77, which
represents the London pride (_Saxifraga umbrosa_), affords an example
of a regular radiating flower, which we find so placed, in different
examples, as to appear as a floor or wall ornament; and Figs. 78 and
79, the former being the flower of the speedwell (Veronica), and the
latter that of the common pansy (_Viola tricolor_), furnish us with
illustrations of bilateral flowers intended only as wall ornaments. In
order to secure our seeing the pansy only laterally, it is furnished
with a bent stalk; hence it never rests horizontally upon the summit
of its stem, but always hangs so that it is perfectly seen only from
the side.
There are cases, however, in which bilateral flowers are placed
horizontally; but it is very interesting to notice that when this
occurs the disposition or arrangement of the flowers is such as to
restore the radiating symmetry. Thus, if we take the candytuft
(_Iberis_) or the common hemlock (_Conium_), we find that while each
flower is bilateral in character, the flowers are yet arranged around
a centre in such a manner that the smaller portion of each flower
points to the centre of the flower-head, while the larger parts point
outwards from the centre of the group. These, then, are the teachings
of plants, to which we are called upon to hearken.
The above illustrations are not only useful examples of the
suggestions of plant-forms to the ornamentist, but form excellent
material to the art-student for the conventional treatment of leaves
and sprays, buds and blossoms. They will also serve to indicate the
kind of plant-forms that should be chosen for decorative purposes.
Students of this branch of art would find it a useful practice to make
a collection of flowers and plants or parts of plants that appear to
offer features similar to those of which we have been writing, and
test their capabilities for decorative purposes, by endeavouring to
arrange them for the ornamentation of wall and floor, as we have
treated the plant-forms indicated in this chapter.
We have now seen the principle on which all carpet patterns should be
constructed as distinctive from wall patterns, and in order to impress
the necessity of giving a radiating basis to the ornaments placed upon
carpets, and not a bilateral structure, we have referred to the
principle of plant growth, where we noticed that all plants, when
viewed as floor ornaments (when viewed from above), are of a radiating
character; whereas if they are seen as wall or vertical ornaments,
they are either radiating or bilateral. This is a necessity of a
carpet pattern, that it have a radiating structure, or, in other
words, that it point in more than two directions.
Man naturally accustomed to tread on grass, when brought into a state
of civilisation, seeks some covering for his floor which shall be
softer to the tread and richer in colour than stone or brick. And in
our northern climate he seeks also warmth; hence he chooses not a mere
matting, or lattice of reeds, but a covering such as shall satisfy his
requirements.
In early times our floors appear to have been strewn with sand--a
custom still lingering in some country districts; then came the habit
of strewing reeds over the floor, and on the part of the opulent,
sweet-scented reeds (_Acorus calamus_). And it is curious to notice,
in connection with this subject, that one of the charges brought by
Henry VIII. against Cardinal Wolsey was that of extravagance in the
use of sweet reeds. This use of reeds was succeeded by the employment
of mats of simple appearance, formed of a kind of grass, and these by
the introduction of wool mats, which, at first, were chiefly imported,
but afterwards manufactured in our own country. The wool mats were in
their turn replaced by carpets, which gradually increased in size till
their proportions became such as to cover the entire floor on which
they were placed.
This brief history brings us to notice what is required of a
carpet:--it should be soft in texture, rich in appearance, and of
"bloomy" effect.
We may add to these requirements by saying that a carpet should also
be a suitable background to all works of furniture or other objects
placed upon it, and that in character it should accord with the
objects with which it is associated in any particular apartment.
Considering more fully these requirements, we notice that a carpet
should be soft. This is very desirable, for softness gives a sense of
comfort, and with softness is generally combined durability of the
fabric; but softness can scarcely be regarded as an art-quality. Yet
as the art which an object bears is more leniently viewed when the
fitness of the object to the purpose for which it is intended is
apparent, we may safely regard softness as a very desirable quality of
a carpet.
The Eastern carpets are pre-eminent in this quality of softness, and
of English-made carpets "Brussels" and tapestry are the least
satisfactory in this way; as usually made, they have a hard "backing."
A kind of Brussels carpeting with a soft back has recently been
brought out, but at present it is not general in the trade. If the
carpet employed in any apartment as a floor covering is harsh in
character, it is desirable to place soft felt under it (felt for this
purpose can be got at carpet warehouses), or evenly spread soft hay,
for by so doing the wear of the fabric will be greatly increased, and
the pleasure of walking on it will also be correspondingly greater.
The next quality of a carpet is richness. No carpet is satisfactory
which is "washy" or faded in appearance. There must be "depth" of
effect, a "fulness" of art-quality. Hangings may be delicate,
wall-decorations soft in tint, but a carpet must be rich and "full" in
effect, yet a general softness of tone is desirable.
But this richness must be of singular character, for the most
desirable effect which a carpet can present is that of a glowing
neutral bloom.
I hope that my language does not appear mystical to the general reader
or young student. To the ornamentist I think it will be intelligible.
What I wish to say is that the effect should be glowing, or radiant,
or bright, as opposed to dull, quiet, or heavy; that it should be such
as results from the use of a predominance of bright and warm colours,
rather than of cold and neutral hues; that it should be neutral,
inasmuch as it should not present large masses of positive colour, hut
should have an equality of rich harmonious colours throughout; that it
should be "bloomy," or have the effect of a garden full of flowers, or
better, of the slope of a Swiss alp, where the flowers combine to form
one vast harmonious "glow" of colour. This is the effect which a
carpet should present, yet it should never present flowers,
imitatively rendered, as its ornamentation. Such imitative renderings
are not to be produced by the ornamentist; they must come from the
pictorial artist, for they are pictures. They cannot form suitable
backgrounds to furniture and living objects, for they are positive,
and not neutral, in their general effect. A picture, also, will not
bear repetition: whoever heard of one person having two copies of the
same picture in one room? Yet a pictorial group of flowers may be
seen repeated many times over a floor, which is very objectionable.
The effect to be produced is that of a rich "colour-bloom;" but the
skilled ornamentist will achieve this without violating any laws of
fitness, and will gently and delicately hint at the beauty of a
profusion of blossom through his tenderly formed pattern.
[Illustration: Fig. 80.]
Yet a carpet must be neutral in its general effect, as it is the
background on which objects rest. Neutrality of effect is of two
kinds. Large masses of tertiary or neutral colours will achieve its
production, so also will the juxtaposition of the primary colours in
small quantities, either alone or with the secondary colours, and
black or white; but there will be this difference between the two
effects--that produced by low-toned colours will be simply neutral,
while that produced by the primary colours will be "bloomy" as well as
neutral, and if yellows and reds slightly predominate in the
intermingling of colours, the effect will be glowing or radiant.
The radiant, or glowing, bloomy neutrality of effect is that which it
is most desirable that a carpet should present.
[Illustration: Fig. 81.]
This effect is rarely produced in English carpets, owing either to the
want of skill on the part of the ornamentist, who is unable to produce
such works; the want of judgment on the part of the manufacturer,
whereby he fails to produce such patterns; or the want of taste on the
part of the consumer, owing to which he buys works of a more vulgar
character. I have designed carpets in which I have sought to realise
as much of this effect as I could with six colours--the number to
which I have been limited by the conditions of manufacture, and
fortunately these appear to be commanding a large sale, and to be
setting a fashion in carpets; but those who wish to study these bloomy
effects in their more perfect forms, must do so in the carpets of
India, Persia, Smyrna, and Morocco, but especially in the Indian rugs.
Some of the carpets from India are perfect marvels of colour-harmony,
and of radiant bloom. They appear to glow as a bed of flowers in the
sunshine, and yet they are neutral in their general effect, and when
placed in an apartment do not usurp a primary place, as does any
pictorially treated pattern.
This "bloom" was seen to perfection in one or two silk rugs which were
shown at the International Exhibition of 1862 in London, and it was
not much less apparent in some of the carpets from India shown in the
Paris Exhibition of 1867. Most Indian carpets have this colour-bloom
to some extent, and few are unworthy of careful study.
Persian carpets (Fig. 80) are also models of what carpets should be;
they are less radiant than many of the Indian works, but are almost
more mingled in colour-effect. In pattern many of the Indian and
Persian carpets are identical, being traditional, yet in colour they
differ, and both are worthy of much consideration.
The Morocco carpets (Fig. 81) differ again from both those of India
and Persia, and even to a greater degree than the Persian carpet
differs from the Indian. In these there is often a prevalence of soft
yellows and juicy yellow-greens, intermingled with reds, blues, and
grey-whites, in such a manner as to produce a most harmonious and
artistic effect. To the young student, and to any who may desire to
cultivate his taste in respect to such matters, I say, Study the
carpets of the East most carefully, especially those of India, Persia,
and Morocco.
[Illustration: Fig. 82.]
Indian carpets, such as we have just referred to, may be seen at the
museum in the building of the new India Office at Whitehall, which
museum is open free to the public (for examples, see Figs. 82, 83,
84).
As to the nature of the pattern which may be applied to a carpet, we
have "all-over" patterns, or patterns spreading regularly all over the
surface; "geometrical" patterns, or those which have an apparent
regularity of structure; and panel patterns, or those in which
particular parts are, as it were, framed off from other parts.
First, as to "all-over" patterns. These are what we almost always find
in both Indian and Persian carpets, and are, undoubtedly, the true
form of decoration for a woven floor covering. What is desirable is an
evenly spread pattern, such as will give richness without destroying
the unity of the entire effect. The pattern may have parts slightly
accentuated or emphasised beyond other parts, but not strongly so, and
this emphasising of parts must be arranged with the view of securing
to the pattern special interest. Thus, if a carpet is viewed at a
distance it should not appear as devoid of all pattern, but through
the slight predominance of certain leading features (in Indian
carpets, generally of ornamental flowers) the plan of the design
should be indicated. More detail should be apparent when the work is
seen from a nearer point of view, and still more upon close
inspection; but in no case should any parts appear strongly
pronounced, or otherwise than refined and beautiful, and in no case
should there be a want of interest manifested by the pattern.
[Illustration: Fig. 83.]
Carpet patterns are generally better if founded on a geometrical plan.
In this way most of the Indian and Persian patterns are constructed. A
geometrical plan secures to the design a manifestation of order and
thought in its formation. Panel patterns, unless very carefully
managed, become coarse. In some Indian carpets we find a sort of panel
in which the colour of the ground is changed from that of the general
ground of the carpet, but here the panel has usually a truly
ornamental form, and is, indeed, rather a large ornament than a sort
of frame enclosing a distinct space. Whenever a panel occurs in an
Indian, Persian, or Moorish carpet, it is so managed, and its
surroundings are such, as to cause it to appear as a part natural to
the general design; but it is far otherwise with the panel patterns
which we occasionally see in our shop-windows as the produce of native
industry, and it is far otherwise with those which are used in vast
quantities by the Americans. Judging from the carpets which they
order, I imagine that nowhere on earth is taste in matters of
decorative art so depraved as it is in America. It is true that the
great floral patterns have ceased to be demanded by them, but they are
only replaced by coarse, raw-looking panel patterns, coloured in the
most vulgar manner, and without even a hint at refinement or harmony
of colour. Let the pattern be "loud" and inharmoniously coloured, and
the chances of its sale in the American market are great.
[Illustration: Fig. 84.]
But we must not forget that even in our own country bad patterns sell
equally as well as good, inartistic patterns as well as those which
are of a more refined character, and that even here in Great Britain
more of the indifferent, if not of the very bad, sells than of the
good. Let us cast the beam, then, from our own eye, before we try to
extract it from that of another.
The ground colour of a carpet may vary much, as we all know; it may be
black, blue, red, green, or white, or any other colour. If the ground
of a carpet is pure white, it is almost impossible that it look well.
When I make this assertion I am often told that some of the Indian
carpets which I so much admire have white grounds. This is a mistake.
Some of them have light grounds, but not pure white. They have light
cream-grey, or green-white grounds, but not pure white, and this
variety of tone altogether alters the case. Yet even with a
light-toned ground it is not an easy matter to make a carpet which
shall appear as a suitable background to the furniture of a room; it
can be done, but it is a thing difficult to achieve. The safest and
best ground for a carpet is black or indigo blue. If on this a closely
fitting, well-studied pattern be arranged, drawn in small masses of
bright colour, a beautiful bloomy effect may be achieved, and a glance
at our best shop-windows will show that the most satisfactory carpets
are coloured in this way.
As to the size of the pattern we can say but little, as this will be
determined by the coarseness or fineness of the fabric. In a Brussels
carpet each stitch is about the one-tenth of an inch square. In some
Turkey carpets each stitch is a quarter of an inch square. It is
obvious that a much smaller and finer pattern can be produced in
Brussels than in Turkey carpet.
A carpet pattern is best small, or at least small in detail if not in
the extent of the design. A pattern may repeat three or four times in
the width of the fabric (twenty-seven inches if Brussels), or but one
figure may be shown, yet in this latter case the detail of the pattern
may be as great as in the former. That degree of smallness which is
compatible with tolerable distinctness of detail is desirable. For
this reason Turkey carpets are not altogether satisfactory; no fine
pattern can be worked in them, and besides this they have no
colour-bloom and little colour-harmony. In some respects they are
good, but altogether they are not satisfying.
Before I close these remarks upon carpets, let me say that, as
designers, manufacturers, and consumers, we are one and all timid of
new things. We want daring--the energy to produce new things, to
manufacture them, to use them. What if the pattern is "extreme," if it
is better than others? what if Mrs. Grundy should think us
eccentric?--better be eccentric than ever harping on one monotony. If
we could but bear calmly the derisive smiles of the ignorant,
art-progress would be easy.
With us carpets cover the entire floor. In London these carpets are
nailed to the boards, and but seldom taken up. In some parts of
England we find rings sewn around the under edge of the carpet, which
rings are looped to the heads of nails. Carpets so furnished can be
more readily removed for cleaning than those which are nailed to the
floor. Square carpets, such as the Turkey, Indian, and Persian, are
spread loosely on the boards, and can be taken up and shaken without
difficulty. This is unquestionably the most healthy plan of using a
carpet, and it is also an artistic plan. If the outer portion of the
room floor is formed of inlaid wood of simple and suitable pattern,
and a loose square carpet is spread in the centre, we have an artistic
effect, and the desirable knowledge that cleanliness is also
attainable with a reasonable expenditure of labour.
Before we leave the consideration of carpets we will state in
axiomatic form the conditions which govern the application of ornament
to them, as reference can more easily be made to short concise
sentences than to more extended remarks.
1st. Carpet patterns may with advantage have a geometrical formation,
for this gives to the mind an idea of order or arrangement.
2nd. When the pattern has not a geometrical basis, a general evenness
of surface should be preserved.
3rd. Carpets are better not formed into "panels," as though they were
works of wood or stone; on the contrary, they should have a general
"all-over" effect without any great accentuation of particular parts.
The Indian and Persian carpets meet this requirement.
4th. While a carpet should present a general appearance of evenness,
parts may yet be slightly "pronounced" or emphasised, so as to give to
the mind the idea of centres from which the pattern radiates.
5th. A carpet should, in some respects, resemble a bank richly covered
with flowers; thus, when seen from a distance the effect should be
that of a general "bloom" of colour; when viewed from a nearer point
it should present certain features of somewhat special interest; and
when looked at closely new beauties should make their appearance.
6th. As a floor is a flat surface, no ornamental covering placed on it
should make it appear otherwise.
7th. A carpet, having to serve as a background to furniture, should be
of a somewhat neutral character.
8th. Every carpet, however small, should have a border, which is as
necessary to it as a frame is to a picture.
Having thus summarised the principles that govern the application of
ornament to carpets, we may proceed to notice the conditions governing
the decoration of other woven fabrics.