Beautiful rug with Rub el Hizb medallion

[Pierre Galafassi]

Hi all,

The simplicity of its motifs has made this small square rug (*) one of my favorites among Renaissance studio props. A single 8-branches « Rub el Hizb» star hovering over a beautiful red background and a forceful & original border. No «horror vacui» here! Pity that della Francesca did not show a little bit more of it.

FIG 1 & 2
http://www.turkotek.com/old_masters/tiled_arch.jpg

http://www.turkotek.com/old_masters/tiled_arch_2.jpg
P. della Francesca. Virgin and Child with Federigo de Montefeltro. 1472. Brera. Milano.

The knot density seems quite high despite the simple pattern. The «Rub el Hizb» motif makes an Islamic origin of the rug highly likely. The analogy with the square nineteenth-century Moroccan flag below is quite interesting.
FIG 3.
http://www.turkotek.com/old_masters/moroccan_flag.jpg

(*) The picture definition does not allow to be sure that this is indeed a rug, rather than another type of textile, a velvet for example. However, it is generally assumed to be a rug in the literature.

Regards
Pierre

[Filiberto Boncompagni]

Well, Pierre, there is a good commentary about this painting and its rug in the book of Luca Emilio Brancati “I tappeti dei pittori”. Here it is, with related scans:

“The balanced composition and iconography present
in this painting, which has led to numerous
questions on its interpretation, leads one to believe
that the presence of the rug is not casual but is the
direct wish of either the artist or the client. Leaving
aside possible comparisons for which there is no
longer any trace, the main purpose of the rug here
is to frame and raise the sacred space occupied by
the Madonna and Child. This space is different and
more privileged to that occupied by the angels, the
saints standing barefoot on the stone floor, and by
the kneeling offerer. It seems to capture an echo
between the geometric rigour of the carpet and the
figure of the monumental, aloof and indecipherable
Madonna, similar in many respects to Oriental statuary.

As with the smallest details of the scene, the finest
details of the rug are also clearly depicted. The folds
and undulations of the border provide depth to the
material. The pile is represented with a 'salt and
pepper' effect to portray typical colour variations in
the wool. Other small technical details convince the
viewer of the nature of this textile: the presence of a
fringe made of thin white yarn that comes from the
red ends to the left, the thin line that defines the selvage,
and the criss-cross painted on the back of the
rug to convey the idea of a warp and weft weave.
A digital reconstruction of the rug (fig.32) has

http://www.turkotek.com/old_masters/32_DigRicostr.jpg
(Fig.32)

shown that it is almost square and that, despite being
covered by the Virginis mantle, the star medallion
at the centre consists of the typical star and bar
design format which is common on the so-called
'large pattern Holbeins' or 'wheel' carpets (fig. 35).
In addition, Piero della Francesca's scrupulous perspective
of the rug can now be seen clearly: after
having used computer graphics to eliminate any effects
of perspective on the visible part of the rug, no
deformations as such were registered other than a
logical loss in the definition of the drawing. This
highlights a quality that is in fact Piero's stylistic signature,
and which leaves no doubt about the careful
rendition of a real carpet for which there are none
in existence today that might conceivably be used
for comparison.
Critics have not always agreed on the attribution of
the rug. First, Julius Lessing, talking about the border
motif noted that the representation "does not
show the material clearly and its oriental provenance
is uncertain" (Venetian school: 1879, p. 21 ,Pl.
28 E). Wilhelm von Bode and Ernst Kühnel believed
this rug to be of Anatolian provenance, but
together with Foppa's fresco (Pl. no. 5) he quoted
them as being examples that were altered by artists
for figurative requirements (1914, p.138); a consideration
which remained unchallenged in Charles
Grant Ellis's translation of the same work (Bode-
Kühnel, 1984, p. l6). More generically, Gustave
Soulier included it among carpets of "geometric
combination" (1924, p.208), whereas Michele
Campana, in the list published for his 1945 work,
involuntarily provides us with two conflicting attibutions:
on the one hand, quoting Bode, he listed
the Brera altarpiece, correctly attibuted to Piero, as
a rug from Asia Minor (1945, p. 172) whereas a few
pages previously, with the old attribution to Fra'
Carnevale da Urbino, he talks of a "superb
Damascene or Syrian rug" with an emphasis sug-
gesting that this was to be considered his opinion
(ibidem, p.167). Finally, more recently, Ian Bennett
placed this carpet among the so-called 'large pattern
Holbeins', thus suggesting Anatolian provenance
(1977 , p.99).
Although there are clearly features that are common
to rugs from the eastern Mediterranean basin, mainly
Mamluks and Anatolians, we also find various aspects
which distance us from that area: a single star
and cross border, an uncommon format, a yellow lily
with little stylisation (unusual for middle-eastern
rugs) that can be found in the triangular points of
the star medallion. Surviving examples of 'large pattern
Holbein' rugs do not provide us with satisfactory
comparisons. But some originals of the thirteenth
century and various examples featured in oriental
miniatures of the Timurid epoch suggest something
different. Comparing the border motifs with rugs
from the Seljukid period (thirteenth century) found
in mosques at Konya and Beyshir (Istanbul, Türk ve
Islam Eserleri Müzesi, inv. nos. 655, fig.37, 688,
1034; Vakiflar, inv. n. A-344, fig. 36)

http://www.turkotek.com/old_masters/Scan35_36_37.jpg
(Fig. 35, 36 and 37)

we find there are several common
denominators: the succession of eight-pointed
stars, the rhythmic feel, the simple
composition, and the deep saturated colours.
Furthermore, a comparison with decorative motifs
from carpets in Timurid miniatures shows a perfect
adherence to the star and cross sequence in the border
of our rug, with the grid defined by Amy Briggs
as being of type II (Briggs, 1940,p.23;here fig.22).
One might even suggest that Piero's rug represents
an evolutionary mid-way point between the Seljukid
and Ottoman 'Holbein', or possibly Timurid, carpets.
It was certainly a precious piece in
Montefeltro's time which is why it was so highly regarded.
With documents to confirm that Piero was in
Urbino during these years, and assuming that
Federico de Montefeltro was the client as shown by
the painting, one might suppose that the artist had a
chance to see the carpet at his court. Perhaps the
duke had a penchant for rugs. In the almost coeval
miniature on parchment by Francesco di Giorgio
Martini for the Disputationem Camaldulensium by
Cristoforo Landino (1475; Biblioteca Apostolica
Vaticana, doc. Urb. Zat. 508) the duke is shown in
profile as usual, talking with a person (Landino) at a
window for which the only decoration is a 'small
pattern Holbein' or 'wheel' rug (fig. 34).”

http://www.turkotek.com/old_masters/Scan34.jpg
(Fig. 34)

Regards,
Filiberto

[Pierre Galafassi]

Hi Filiberto,

Thanks a lot for sharing with us Brancati’s informations and the highly interesting computer reconstruction of the rug. I like the little stylized lilies included in the eight Rub el Hizb points. There is a, probably fortuitous, analogy with the «kochak» motif of Salor rugs.

The following sixteenth century extant anatolian rug is the closest thing to della Francesca’s carpet which I could dig out so far. Sure, the red field is more crowded, but it features the bold eight-pointed star with the eight lilies (?) and also the large octagon inside the star.

http://www.turkotek.com/old_masters/star_medallion.jpg
FIG. Western Anatolia, sixteenth century, 212x135, Heinrich Kirchheim, Orient Stars, page 237.

Oh, and several extant fifteenth- and sixteenth century Mamluk rugs also feature a large Rub el Hizb medallion. However, the esthetics of della Francesca’s rug and of Mamluk carpets (very low contrast, overcrowded field etc..) are really as far apart as can be. IMHO a common origin is not credible.
Best regards
Pierre