They say that, in the ruins of the kingdoms of the Safavids, there lie settlements and even cities that are designed, not by the hand of man, nor by the geometric arts of the artillery captain, nor by the fall and flow of the land, but by machines. In villages around near, neatly laid waste by marauders from the Afghan kingdoms to the east, the craftspeople were held to be artificers most adroit in all manner of clockworkings and horological objects. Accounts record them as organised into guilds and chapterhouses, working much like the Helvetian Termineur, assembling well-wrought work with elements taken from any number of diverse artisans. So that a Parsi artificer might assemble a clock from fragments from as far afield as St Petersburg, or Stamboul, or Tashkent.

Chief amongst the marvels attested to these now vanished personages, is a thing described to me as 'Al-Madan Meherkat,' literally 'An Engine of Cities.' While the phrase may put one to mind of one of Newcommen's sooty leviathans, the Al-Madan was something more delicate – a kind of near-horolage, perhaps as large as an outstretched hand. In all accounts, one is never described as larger than a foot across – and we are led to believe that the very largest were inferior imitations of more finely crafted works. It is otherwise difficult to divine a sense of consistency in the descriptions offered – it seems that each chapterhouse or manufactory possessed their own design, with their own peculiarities and arrangements.

What is shared, in all accounts, is the general disposition of the device. The device is a shallow, scalloped disc – fashioned like the enclosure of a pocketwatch. The face is customarily of bone or of ivory or of nacre, and must be, at the barest, inscribed with a series of uniform divisions around its edge. In the main, specimens exhibit divisions of thirty-six, or of seventy-two. In the simplest of devices, the face remains otherwise unmarked, although the Kunstkammeret in Copenhagen is in possession of a particulaly fine specimen with an engraved face that reproduces, variously, the construction and destruction by fire of a walled township.

A number of study hands are pinned to the centre wheel. Common to all Al-Madan is a central arm which one may reposition so as to set the location of true north. Below that are a series of thicker hands that may be positioned so that they come to rest over any of the division markings, or over another hand, or hands. These never numbered less than five, and the finer specimens were suggested to have between eight and ten. Each these hands are terminated by a small roundel that can be adjusted by infinitesimal degrees to declare the importance of the hand in the subsequent tabulation. A series of small thumb-wheels are positioned around the top-most arc of the case. These are used to adjust the inscrutable inner workings of the mechanism, and are recorded as 'Al-Anraf', Deviation, or 'Marab', Squareness, or any one of the many other terms that might be used to describe, by degrees, the arrangement or qualities of the lanes or streets or market-places of any great city. A great hob is situated somewhere on the rim, for winding the mechanism. Finally, the obverse of the shell houses a single counter that the operator must increment for each chebel – a distance of perhaps twenty yards - he walks. This, one is struck, appears as the single most tedious act involved in engaging the Al-Madan – as all else is mercifully automated.

Few documents survive that describe the use of the Al-Madan – the most trustworthy, an account from the celebrated envoy Jean Chantille, described in his usual, ponderous technicality;

First, the opérateur sets a northing, by compass, or by the position of the sun. One suspects this originally took the form of a deep and solemn ritual, but as one must recalibrate the infernal engine at each point of confluence, the opérateur employs a remarkable economy of movement and of time – barely glancing at the device as he swings the hand around to ready it for another calculation. Having set north, the opérateur uses the subsidiary hands to set the lay of any existing roads or paths or rude tracks – picturing himself as standing at the centre of the face, and the features radiating out from his feet. Hands that do not correspond with existing features are left where they lie, with only their terminal roundel set to the null mark to indicate their paucity of importance.

For the other hands, the roundels are similarly adjusted – a principal road, a paved highway, perhaps, might be set to the twelfth mark on the roundel. A cobbled street might merit the fifth mark, and a shepherds' track might be indicated by a single, solitary incision. Thus recorded, the opérateur may alter the prédicats – the dials that set the pattern and arrangement of the city. I say may as the process is a long and involved one, and small errors in judgement can be compounded and amplified through repeated operations of the device, much as a single misplaced mark on a Pascaline may spiral out of control and render a tabulation useless. It is the practice of apprentices and journeymen to take to the field with prédicats already decreed by the superiors, and to alter them only slightly, or not at all. But there are tales, too, of opérateurs with such dexterity and insight into clockwork soul of their machines, that they would alter their prédicats as they stalked the ghostly half-sketched lines of their planned streets and boulevards. They say the city of Aleph was conceived this way, by one madman, in one mad day.

(But that I believe to be a confabulation – a local folktale. Opérateurs always work and walk their sacred rounds in teams of ten or twenty – checking and re-checking each others pathways and patternmaking.)

Having set the lay of the land, and satisfied by their prédicats, the opérateur then winds the mechanism, providing a motive force for its tabulation and its calculation. With alacrity, the device re-arranges the order of the unused hands – pointing out new directions for new roads. These may be understood to be something like the branches and tributaries of a river and they mesh seamlessly with the existing pathways and thoroughfares. At the tip of each of these hands, the terminal roundels have been adjusted by the mechanism to indicate their relative importance.

The opérateur then moves the northing arm so that it overlaps with one of the hands. This is to be the path he walks.

He readies himself, and sets out in the direction indicated by the hand. Every twenty yards he winds a little ratchet in the back of the device, and every twenty yards the hands whir and shift, and renegotiate themselves, and he follows this new bearing. Thus, the device can equally describe a straight boulevard, or a gently curving path, by the simple decomposition of these forms into a series of discrete points. At times, the hands may shift into new arrangements that suggest forks or intersections in this pathway. The opérateur is at liberty to move down these new lines as he sees fit, repeating the process as he goes.

It is in this manner that a company of capable opérateurs may set out across a pristine landscape, following the dictate of their devices, laying out the lines and sketches of settlement – streets and plazas and wells and walls – all marked in the soil by stout staves with coloured pennants and bands of coloured gypsum and talc.

Chantille goes on to mention the use of a subsidiary device that suggested partitions between properties and the correct alignments for sewers and culverts. He also hints at mythic, half-hidden engine that was capable of 'the most workmanlike division of geometries and pattern in religious and secular facades,' which, alas is now lost to Occidental knowledge. He comments on much, but what he does not comment on is the sheer profligacy of the act of those empowered by the Al-Madan.

The lines that he records were, it is implied, the sketches and boundaries of cities to come. But they outstripped their need and became lines that would have to wait for their inhabitants, a people that might arrive in five years, or ten, or fifty, or never at all.

And one must be aware that a well trained company was reportedly able to cover many hundreds acres in a single day – marking and recording and inscribing as they went. There were two-hundred and fifteen registered companies, working for the better part of a hundred years! And thus, there are manifold pathways in the plains beyond Isfahan, in the lands of Uqbar; routes and passages and defiles in the sand and the scree – a city beyond the city, uninhabited and uninhabitable, stretching out as far as the eye can see. I have recorded traces of it – palisades of bone white staves, skeins of thin, coloured cotton stretched taut between them. Inspection reveals that these arcs and lines connect back to the rudimentary tracks and broken roads of the province, and appear to stitch together the villages, hamlets, holy places and shrines that have survived nearly two decades of war and depredation. Here, in valley and in dry copses, are the not-yet-ruins of markets and souks. Here, on high hills and exposed bluffs are the shadows of tight lanes and tall towers, unbuilt and unbuildable. Here is a city greater than London, or Paris, or the cities of the Far East – a city that stretches clear from the shores of the Caspian to the marches of the Punjab. That it is inhabited solely by scattered bands of shepherd-folk, and their utterly contentious sheep, almost seems inconsequential – it is the design that has been achieved, and all else that follows is contingent, almost unnecessary.

This is the paradox raised by the Al-Madan. Is the city the artifact, or the intent? Are these unbuilt cities – fantastic dreams of the urbis- to be regarded as having failed by their phsyical absence. Or are we to see construction, embedding, emplacement as the failure – as the compromise, as the contamination of a purer, idealised frame for the city?

They say that, in the ruins of the kingdoms of the Safavids, there is a city designed by machine. It is an endless city, unceasing.

It is the only city we will ever need.