What Gets Made When Figures Are Avoided

The standard account is that Islam forbids images, which is why Islamic art is geometric. Almost every clause of that sentence needs work.

Start with the source. The Quran does not prohibit depicting living beings. It condemns idolatry, repeatedly and strongly, and that is a different prohibition: it is about worship, not about representation. The explicit prohibitions on making images are in the hadith, the body of reported sayings, where painters are challenged to breathe life into what they have made and warned about the Day of Judgement.

Hadith are interpreted, and these have been interpreted differently across thirteen centuries, four legal schools, Sunni and Shia practice, and a dozen regions. Sunni thought has generally been more explicit against depicting human beings; Shia practice has often been more relaxed. There was outright iconoclasm at moments, including an order from the Umayyad caliph Yazid II in 721 to destroy representational images, which did not last.

Now the practice, which is the part the standard account simply omits.

Figurative painting flourished. The Persian miniature tradition has been dominant since roughly the thirteenth century and strongly shaped both the Ottoman and the Mughal traditions. These are not marginal: they are large bodies of work full of faces, battles, hunts, portraits, court scenes and legendary narrative. Mughal portraits are usually in profile and very finely drawn in a realist manner; the best Ottoman work is vigorously stylised. Representations of Muhammad exist, sometimes with the face veiled and sometimes not, in manuscripts from Persia and India, one of the earliest in a manuscript painted in Konya around 1250.

The reason this coexists with the prohibition is context. Miniatures are private objects. They live in albums and illustrated manuscripts, handled by a few people, not displayed on a wall in a public religious space. The constraints that governed the decoration of a mosque did not govern a book in a library.

So the restriction operates on religious space, and it is there that the alternative was developed.

Calligraphy was the highest art in the Islamic world, and calligraphers held higher status than other artists. They signed their work. Connoisseurs could identify a particular master's hand on sight. That is a description of a prestige art with named individual practitioners, which is not what an art form under suppression looks like.

And geometry, which is where the subject gets genuinely surprising.

In 2007 Peter Lu and Paul Steinhardt published a paper in Science arguing that the most elaborate Islamic tilings were not laid out with compass and straightedge line by line, as had been assumed. They were built from girih tiles: a small set of shapes, from pentagons to bow ties, each carrying a fixed pattern of lines across it. Tessellate the shapes and the lines join into the pattern. The tiles are simple enough to construct with tools documented in medieval Islamic sources.

Their striking example is the Darb-i Imam shrine in Isfahan, dated 1453, whose tiling comes very close to the quasi-periodic patterns Roger Penrose described in 1973 and that physicists found in metal alloys in 1984.

The authors were careful and the reporting was not. The claim is that the pattern has properties consistent with quasicrystalline tilings, produced by a constructive method. The claim is not that fifteenth-century artisans possessed the twentieth-century mathematics. A method that generates a result is a different thing from a theory that explains it, and the distinction is worth keeping, because the careful version is still one of the more remarkable facts in the history of ornament.