_GEOMETRICAL DECORATION_
_The Line._
One of the simplest and the earliest kinds of ornament that we find is
the zigzag line, which occurs on the oldest tombs, 4000 B.C. So simple
is this, that it might be supposed that every possible variety of it
would be soon played out. Yet, strange to say, two of the simplest
modifications are not found till a couple of thousand years after the
plain zigzag had been used. The wavy line in curves instead of angular
waves is not found till the XVIIIth dynasty, or about 1500 B.C.; while
the zigzag with spots in the spaces is equally late, and is generally
foreign to Egypt.
[Illustration: 1.—VI. dyn., L.D., II. 98.]
[Illustration: 2.—IV. dyn., Mery, Louvre.]
[Illustration: 3.—V. dyn., Ptah-hotep, Perrot XIII.]
The plain repeated zigzag line is used down to late times, but generally
with variety in colour to give it interest. From the earliest times this
was symmetrically doubled, so as to give a row of squares with parallel
borders; or with repeated zigzag borders in alternate light and dark
colours. This same type lasted onward to the XIXth dynasty (belt Ramessu
II. C.M.X.), and is found, with the addition of spots in the outer
angles, in the foreign dress of Shekh Absha, at Benihasan, in the XIIth
dynasty.
[Illustration: 4.—Prisse, Art. 84.]
[Illustration: 5.—L.D., II. 130.]
A later stage was to repeat the squares with varieties of colour; and
also to introduce details into the squares, and so make them compound
patterns, as in the XVIIth dynasty at El Kab, where the sequence of the
blue, green, and red lines makes a brilliant effect from these simple
elements. Not only a square, but also a hexagon, was worked into the
same design. This, from the nature of it, suggests a rush-work screen,
and probably it was plaited with rushes in three directions, and hence
the production of this particular angle. The previous zigzag patterns
all suggest weaving; and in some in Ptah-hotep’s tomb (Vth dyn.) closely
woven and complex zigzag patterns are shown which are evidently copied
from textiles, as we shall see further on in the chequer patterns.
[Illustration: 6.—XII. dyn. Amu dress.]
[Illustration: 7.—XVIII., Keft dress. C.M. cxcl.]
[Illustration: 8.—XX. Vase, C.M. cclix.]
The use of spots for filling in corners was foreign to the Egyptian. We
first find it in the garments of the Amu, or people of northern Arabia,
in the XIIth dynasty. Till then a spot is never seen, except for the
centre of a square; but the Amu dresses are covered with spots in every
space, and even along the bars and stripes of colour. The same is seen
on the later dresses of the Amu in the XIXth dynasty, and also in the
dress of the Phœnicians, or Keft people. It recurs on the foreign vases
probably brought in from the Aegean; and it is only found in Egyptian
products during the XVIIIth dynasty, when foreign fashions prevailed,
though it is but rare then. Hence we may fairly set aside this use of
spots as a foreign or Asiatic element, akin to the filling in of spaces
on early Greek vases with rosettes and other small ornaments.
[Illustration: 9.—XVIII., P. I. xvii. 7.]
[Illustration: 10.—XVIII. Vase, R.C. lvii.]
The zigzag line only became changed into a rounded wavy line in the
later time of the XVIIIth dynasty. This probably results from the
earlier patterns being all direct copies of textiles which maintained
rectilinear patterns; but when the same came to be used on pottery (as
above), or on metal work (shield border, L.D. iii. 64), then curves were
readily introduced. On a golden bowl repeated waves are shown, deepened
so as to receive further figures.
_The Spiral._
The spiral, or scroll, is one of the greatest elements of Egyptian
decoration; it is only second to the lotus in importance, and shares
with that the origination of a great part of the ornament of the world.
The source of the spiral and its meaning are alike uncertain. It has
been attributed to a development of the lotus pattern; but it is known
in every variety of treatment without any trace of connection with the
lotus. It has been said to represent the wanderings of the soul; why, or
how, is not specified; nor why some souls should wander in circular
spirals, others in oval spirals, some in spirals with ends, others in
spirals that are endless. And what a soul was supposed to do when on the
track of a triple diverging spiral, how it could go two ways at once, or
which line it was to take—all these difficulties suggest that the
theorist’s soul was on a remarkable spiral.
[Illustration: 11.—F.P. coll.]
[Illustration: 12.—F.P.]
[Illustration: 13.—F.P.]
The subject of spirals fall into two groups. The older group by far are
the scarabs, which contain spirals on a limited and small field; the
other group are those continuous patterns on ceilings, furniture, &c.,
which are capable of indefinite extension by repetition. As the scarabs
are far the older examples, there is a presumption that spirals may have
even originated on scarab designs; and the hesitating and simple manner
of the oldest instances on scarabs indeed seems as if the engravers were
merely filling a space, and not copying any well-known pattern. The
earliest that can be certainly dated is one of Assa, of the Vth dynasty,
on which a bordering line is interrupted at the ends and turned in to
fill the space on either side of the name. From the cramped way in which
this is done, and the want of uniformity in the spirals, it seems as if
no regular pattern were in view, but only the need of avoiding an
unsightly gap in the design. We next see spirals used in the same way to
fill up at the sides of the inscription on the scarabs of Pepy, without
any attempt to connect them into a continuous pattern; and on the
scarabs of Ma·abra, probably soon after, the same loose spirals are seen
thrown in to fill up. In none of these cases is the ornament anything
but the means of supplementing the required inscription; nothing is
arranged for the sake of it, and it is treated as a mere afterthought.
Nor is it until the XIIth dynasty that any continuous spiral design can
be dated. For over a thousand years, then, the spiral is only to be
found as an accessory on scarabs, a fact which strongly suggests that it
originated in this manner.
Before describing spirals further, it is needful to settle some definite
names for their varieties. Where the lines are coiled closely in a
circular curve, as in Assa’s scarab, they may be termed _coils_; where
lengthened out, as in Pepy’s, we may term them _hooks_; where lengthy in
the body between the turns, as in Ma·abra’s, they are rather _links_.
Where the line is broken at each spiral, as in all the above, it is a
_chain_ of spirals; but where the same line is maintained unbroken
throughout it is a _continuous_ spiral, and these are found in all
varieties of coils, hooks, or links. Sometimes the continuous line has
separate ends, but more usually it is _endless_, returning into itself.
These terms will suffice to distinguish the varieties, and enable us to
speak of a spiral with definiteness.
[Illustration: 14.—Louvre.]
[Illustration: 15.—Ghizeh.]
These detached spirals continued in use in the XIIth dynasty, generally
as loose links, often not hooking together, as in this of Usertesen II.
In the XVIIIth dynasty this is still found as a general surface ornament
on the boat covers of Hatshepsut at Deir al Bahri and on the base of a
Kohl vase in the Ghizeh Museum.
[Illustration: Fig. 16. F.P. coll. Fig. 17.]
But the spiral was developed, apparently under Usertesen I., into a
chain of coils, which are drawn with great beauty and regularity. Such
care indicates that the design was a novelty, which was not yet
stereotyped and reproduced as a matter of course. In no later reign were
spirals ever so beautifully and perfectly executed. This type was
revived under Amenhotep II. (H. S. 1097). In about the XIIth dynasty it
was combined with the lotus in perhaps the most perfect design that
remains on any scarab—a continuous coil with flowers and buds in the
spaces.
[Illustration: 18.—Turin.]
[Illustration: 19.—F.P.]
But it was felt that the spirals all round occupied too much of the
field, so the top and bottom were left free for inscribing, and the
ornament was limited to the sides, as in this chain of hook pattern of
Usertesen I. This design, with the line continued around the top as well
as the base, was the staple decoration of the private scarabs of the
XIIth-XIIIth dynasties, many of which are of great beauty. Both types
are found, but the hook pattern is more usual than the coils.
[Illustration: Fig. 20. F.P. coll. Fig. 21.]
[Illustration: Fig. 22. F.P. coll. Fig. 23.]
In the finest work, however, the line is made endless, a single
continuous line forming the whole pattern, as in the endless hook
pattern of Setmes, and the endless coil pattern of Ptaherduen.
[Illustration: 24.—Paris.]
[Illustration: 25.—F.P.]
In the few spiral scarabs of later times the pattern is not only placed
at the sides, but is carried all round, as we see in that of Amenhotep
I. and one of Ramessu II., which latter is the latest spiral pattern
known on scarabs.
[Illustration: 26.—F.P. coll.]
The long links were seldom used in continuous patterns around scarabs,
as in this, but were more usually employed for independent spiral
patterns without any inscriptions.
[Illustration: 27.—F.P. coll.]
[Illustration: 28.—K. x. 50.]
[Illustration: 29.—I. viii. 69.]
[Illustration: 30.—K. x. 28.]
[Illustration: 31.—K. x. 40.]
After serving as adjuncts to inscriptions, the spirals became elaborated
as sole patterns. These are at first a few simple coils, as on one
which, from the side pattern, can be dated to about the VIIIth dynasty.
These, when elaborated with more coils or links, sometimes developed to
great length.
[Illustration: 32.—K. x. 17.]
[Illustration: 33.—F.P.]
[Illustration: 34.—I. x. 176.]
Such patterns required but little ingenuity, and it is rather in the
design of continuous spirals that the Egyptian showed his skill. The
problem was how to arrange a number of coils in a symmetrical system
uniformly covering the surface of the scarab, and yet to connect them in
a true series. This was done in various ways, usually by introducing
long loop lines around the edge. One of the simplest type is— In another
a cross pattern is formed which is entirely of C coils, like frequent
patterns at Mykenae.
[Illustration: 35.—F.P.]
[Illustration: 36.—F.P.]
[Illustration: 37.—F.P.]
Others fill up by establishing a repeating pattern, which might be
indefinitely multiplied, as — and the difficulty is avoided on a large
silver scarab of early date by shortening the links to allow of the
connecting line passing the ends.
This difficulty of designing good covering patterns out of true
continuous lines probably led to the evasion of introducing false links.
Thus what would otherwise have been an opening in the middle was barred
across.
[Illustration: 38.—I. x. 158.]
[Illustration: 39.—K. x. 27.]
[Illustration: 40.—K. x. 48.]
[Illustration: 41.—F.P.]
[Illustration: 42.—F.P.]
Some beautiful effects were obtained by this false barring, which does
not, at first sight, catch the eye, as in these two examples.
[Illustration: 43.—F.P.]
In the latter, two complete lop-sided spiral groups are joined by long
false links around the outside. Another favourite device which often
occurs is also compounded of lop-sided groups, or rather of a cross
group, like Fig. 43, with four false links joining in the middle.
[Illustration: 44.—I. x. 144.]
[Illustration: 45.—I. x. 155.]
[Illustration: 46.—F.P.]
[Illustration: 47.—Turin.]
Some other devices did not profess to cover the whole field, as in Figs.
44 and 45; and sometimes two separate lines of design were superposed, a
single element of the same design being found as late as Tahutmes III.
The spiral had thus been greatly developed as a detached ornament for a
small surface; but in architecture and furniture it was required as a
continuous decoration on borders and on large surfaces. Hence its
development was in many ways different, and—so far as we know—later by a
whole cycle of history than the development on the scarabs. On those
small objects it started in the Vth dynasty, became fully elaborated in
the XIIth, is common in the XIIIth, and only very occasionally found in
the XVIIIth, disappearing altogether in the XIXth. On walls and
furniture it is rare in the XIIth dynasty, becomes usual in the XVIIIth,
flourishes in the XIXth and XXth, and is decadent in the XXVIth.
[Illustration: Fig. 48.]
The simplest form in which it is found is as a chequered pattern series
of S spirals, apparently on cloths thrown over boat cabins. On
Hatshepsut’s boat the spirals are close together (Duem. XXI.); but
rather later, on the boat of Neferhotep, they are spread with chequers
of red and blue between them (W.M.C. lxvii.).
[Illustration: Fig. 49.]
About the same period they appear as a continuous coil pattern in relief
on the columns of the _harim_ well at Tell el Amarna. The spiral in
relief being in yellow, it probably was copied from a jewellery pattern
in which a strip of gold was twisted into spirals, and the spaces filled
with squares of coloured stones or pastes, judging from the analogy of
the inlaid capitals. This example being earlier than most of the spiral
decorations of surfaces may thus open our eyes to the meaning of some
such designs; and, in general, a close continuous coil returning on
itself may well be a copy of a strip of sheet metal, doubled, and rolled
up.
[Illustration: 50.—P. 85. I.]
The next stage is where continuous lines of spiral patterns are placed
side by side, and other patterns developed in the spaces between them.
Sometimes the intervening patterns become so complex as to overshadow
the mere spirals, as in the splendid ceiling of Neferhotep, in the
XVIIIth dynasty. And in this the far more complex quadruple spiral
begins to appear, as we shall see presently.
[Illustration: 51.—P. 85.]
[Illustration: 52.—C.M. cclv.]
The lines of spirals were not only placed parallel, but were also
crossed. For some reason this type was never well developed, but
remained one of the coldest and most mechanical of all, looking in the
later stage of the XXVIth dynasty like a most debased wall paper.
But the glory of Egyptian line decoration was in the quadruple spiral,
of which the most elementary example is on a boat cover as late as the
XXth dynasty (Ramessu IV.); though it has passed through this stage long
before that time—if indeed this may not be regarded as a degraded
simplification of it. It is also sometimes rhombic in plan.
[Illustration: 53.—P. 86.]
[Illustration: 54.—XIIth dyn. R.C. lxxii.]
From this was developed a peculiar pattern by the omission of the lines
which define the spirals, thus reducing it to a system of rows of
hollow-sided quadrangles without any apparent connection.
The main development of the quadruple spiral was with rosettes or lotus
filling the hollow squares.
This became a stock subject with the Egyptian, and from thence a main
pattern in other lands. The filling in was either a flower pattern or a
rosette, which might be either a flower or a leather pattern, as we
shall notice further on.
[Illustration: Fig. 55.]
[Illustration: 56.—P. 86.]
The insertion also became more complex, four lotus flowers being placed
in each angle of the hollow square; and the spirals being more heavily
developed, in order to gain enough space for complexity in the squares
between them. Such a system could hardly be carried further, but reached
its limits; like the limit of size in the Great Hall of Karnak, where
the columns occupy too large an area in proportion to the clear space.
[Illustration: 57.—P. 83.]
In another direction, however, the spiral blossomed further, in the
parallel lines of spiral pattern. These became developed by introducing
link lines so as to form a quintuple spiral, which was further
complicated by lotus flowers and buds in the hollows and recesses.
In this direction, again, the Egyptians had reached the limit beyond
which more detail would be merely confusing. By careful use of colour to
separate the various parts, these complex patterns remain clear and
pleasing in spite of their richness of detail.
The quadruple spiral had, however, another development, of C links,
which is rather too formal to be beautiful, and lacks the flamboyant
grace of the chains of spirals. Still it has a simple dignity, related
to the scarab spirals rather than the flowing surface patterns. This
became formalised into a torturing kind of design, which can only be
described as “cursedly ingenious.” By simplifying the previous pattern,
a wave was invented which was equal in each direction, and four of these
were crossed in a manner which nothing but bold colouring could make
intelligible.
[Illustration: 58.—P. 85.]
[Illustration: 59.—P. 83.]
[Illustration: 60.—L.D. II. 57.]
The fret patterns are all modifications of corresponding spirals. The
cause of such change is obviously the influence of weaving. As early as
the Vth dynasty we find a fret of rhombic form in basket-work in the
screen behind the figure of Ptah·bau·nefer, at Gizeh. The angles show
that the plaiting was in three directions, as we saw in the basket-work
pattern at Benihasan (Fig. 3). But frets in general are very rare until
a late period, and they doubtless depend on the adaptation of spirals to
textiles. We see no trace of the fret in the Mykenaean art, the spiral
there being figured on stone or metal, while the women wore flounced
dresses with scale pattern. But in the pre-Persian age fret pattern
weaving in borders was the standard design, as we see on the coloured
robes of the Parthenon statues; and immediately after that the stiffest
of square frets swarms over Greek art, to the exclusion of the graceful
spirals and scroll borders.
[Illustration: 61.—P. 82.]
[Illustration: 62.—P. 83.]
[Illustration: 63.—P. 83.]
The chains of links were copied in the fret pattern with no difference
except in squaring up the curves. The same is true of the quadruple
spirals, which appear likewise modified; and this change seems to have
led to another simplified form, which is on the same idea as the
torturing design (Fig. 59), but which is less ingenious, and is still
possible as an ornament.
[Illustration: 64.—Schuck. 256.]
So far we have viewed only the course of Egyptian design, nor can we
travel far outside of it within these pages. Moreover, as it is dated
before any other such decoration in other countries, it is well to view
its course as a whole without confusing it with the various fragments
borrowed from it by other lands. Yet we may well turn now to see the
beginning of the course of European decoration at Mykenae, and observe
its close contact with that of Egypt. The spiral is the main element of
prehistoric decoration in Greece; the parallel chains of links occur
almost exactly as we have already seen them in the pattern of
Neferhotep, but omitting the inner details added in the spaces.
[Illustration: 65.—Schuck. 290.]
The quadruple spiral is splendidly shown in the ceiling of Orchomenos,
with a lotus flower in each space; also as a simpler form without any
filling in of the squares on the grave stele (Schuck. 146). While even
the ox head with a rosette between the horns, in the grand quintuple
spiral pattern (Fig. 57), is strangely paralleled by an ox head of
silver with a large rosette on the forehead found at Mykenae (Schuck.
248).
In observing these equivalents it must be noted that whole patterns with
their detail are taken over complete from Egypt. There are none of the
series of intermediate steps which we have traced in the mother country;
and where a simpler form occurs it is known to be later, the grave
steles being after the age of the great ceiling. Thus there is the
surest sign of a borrowed art, apart from the facts of the exact
resemblances we have noted. Of course the Mykenaean designs are mostly
influenced by the taste of the race. Many of them are strongly European,
and might be of Celtic or Norse work, as has been shown by Mr. Arthur
Evans; but the source of the designs lies in the two thousand years’
start which Egypt had before Europe awoke.
[Illustration: Fig. 66.]
[Illustration: 67.—R.C. lvii.]
[Illustration: 68.—P. 97. 105.]
[Illustration: 69.—R.C. lxii.]
A separate form of the spiral pattern is that used for borders,
otherwise called the wave or maeander, which merged into the guilloche.
Although the chain of coils on the scarab borders in the XIIth dynasty
may be regarded as a wave border, yet no example is known of this border
on other objects until the XVIIIth dynasty. At that time it appears as
often on foreign objects as on Egyptian, and the only instance of the
guilloche is on foreign dress. Hence this development of the spiral idea
may well be due more to the Aegean civilisation than to that of Egypt.
This will agree with the occurrence of the guilloche on black pottery
from Kahun, which class, wherever it can be dated, is found to belong to
the XIIth-XIIIth dynasty. The metal vases shown on the monuments of the
XVIIIth-XXth dynasties are mostly foreign tributes, and on them the wave
border is common, merging into a twisted rope border which is also
found—though rarely—on scarabs of the Middle Kingdom.
[Illustration: 70.—R.C. lvii.]
[Illustration: Fig. 71.]
[Illustration: Fig. 72.]
In Egyptian use this border is seldom found. A box in the Louvre had a
line of long links; and a scroll edge appears to the standard of Ramessu
II. But more usually the scroll is associated with the lotus, as in
these—
[Illustration: 73.—P. 89.]
[Illustration: 74.—P. 89.]
The innumerable adaptations of this in Greek and later designs are
familiar enough to us.
The influence of weaving has been very great upon these wave borders. As
I have before noticed, the woven borders, reducing the pattern to a
fret, are shown on the pre-Persian statuary at Athens, and precede the
most common and oft-repeated use of the fret or key pattern borders in
Greece, and thence in all classical, mediæval, and modern times.
[Illustration: 75.—R.C. cxxi.]
[Illustration: 76.—R.C. lxi.]
[Illustration: 77.—P. 103.]
Another type of border, which may be connected with this, is found in
the Ramesside age. As it occurs as stitching on leather, and is well
adapted to quilting or sewing bands together, it may well have been
derived from that; but it is also found on metal work, with which it
does not seem to be connected by origin.
[Illustration: 78.—P. and C. xiii.]
The source of chequer patterns is unmistakably in plaiting and weaving.
On the oldest monuments the basket sign, _neb_, is chequered in
different colours; so are also the baskets of farm produce carried by
the servants, as shown in the tombs. The modern Nubian basket-work is
well known for the many patterns which it bears like the ancient
Egyptian. The chequer pattern is found in every period in Egypt, and is
perhaps most common in the latest forms on the sides of thrones in the
Ptolemaic age. In the Old Kingdom many varieties were in use. The plain
chequers of red or black with white, the squares filled with black and
red crosses on a green and yellow chequer; or diagonal square patterns
developed by lines of chequers, which are often not square but
elongated, thus forming general and wide-spread patterns which attract
the eye on large surfaces. These are best seen in the tomb of Ptah-hotep
(P. and C. xiii.) and in that of Peheniuka (L.D. I. 41), both of the Vth
dynasty, Sakkara.
[Illustration: 79.—P. and C. xiii.]
[Illustration: 80.—L.D. I. 41.]
[Illustration: 81.—L.D. II. 130.]
In the Middle Kingdom we find chequers covered with bars of colour, red
and green, at Benihasan.
[Illustration: 82.—P. 81.]
Under the empire chequers are less common owing to the greater
development of more elaborate decoration. A pleasing variety was formed
by lengthening the squares, a change doubtless copied from weaving,
where oblong squares serve to break the monotony of the pattern.
[Illustration: 83.—L.D. iv. 77.]
In later ages of the Saitic and Greek times the chequer is a common
resource, but is seldom treated with originality or grace, and we do not
find any new departure or advance in the mechanical execution of the
later examples. One slight novelty was the alternation of whole and
divided squares of colour, under Claudius.
Somewhat analogous are the net-work patterns. They seem to be probably
derived from stitch-pattern over dresses. Though found in the XIIth
dynasty they are not usual until the XVIIIth dynasty, and they are
generally on the dresses of goddesses. A simple example is on a
horse-cloth of Ramesside age, which shows that these can hardly
represent long beads, but rather stitching or quilting. A more elaborate
form is on the dress of Bast in the tomb of Seti I., in hexagons.
[Illustration: 84.—R.S. lxxxii.]
[Illustration: 85.—C.M. ccxlii. cccx.]
But this design rose to importance when it was introduced as an
architectural element in the decoration of columns at Tell el Amarna.
There it is coloured yellow, and the spaces are alternate red and blue.
[Illustration: Fig. 86.]
The Egyptians never used circles freely in decoration; no examples are
known before the XVIIIth dynasty, and but few then.
[Illustration: 87.—P. 79.]
[Illustration: 88.—P. 84.]
[Illustration: 89.—P. 86.]
The intersecting circles, forming a kind of net-work, are found in the
XVIIIth dynasty in blue on a yellow ground; and the same occurs in black
on blue and red ground, in later times (L.D. I. 41). Besides the
rosettes other patterns were introduced into the spaces, which were
coloured red and green alternately. But the most beautiful type was with
contiguous circles not intersecting, and each containing four lotus
flowers.
The circle, however, never became of importance, probably because it was
too stiff and mechanical for the Egyptian, who delighted in the waving
spiral patterns and the unlimited variety of lotus developments. It is
remarkable that there is not a single example of the circle divided into
six, or with six segmental arms, which is so common a motive in Assyria
and Syria, and which results so readily from stepping the radius around
the circle. This seems to show that the Egyptian did not use compasses
at any time, but always worked with a string and points. The absence of
a simple and self-evident motive like the sixth of the circle is almost
more striking than a peculiar motive being present.