I am looking for a way to describe the last two months of activity which have sometimes felt like being lost in a maze, or like falling down a hole into another world to which there is no end, and no obvious structure, that has you wandering in a wilderness of moss, a wide expanse of the tiny, where an obsession with detail makes you lose sight of the larger themes from which you originally set off. The series of objects, in this case Indian temples, keeps unfurling and leading you on, unsure whether it’s a boon or a curse that the series has no end or obvious shape.
The model lurking here seems to be that of finding forms concealed in the ground itself, discovering buildings in the living rock like the figures Michelangelo senses waiting to get out of the stone block, buildings which combine the qualities of sculpture and architecture, which you release from captivity rather than invent or devise according to the rules of a human craft.
Bruegel’s Tower of Babel is not generally considered a mythological painting, but it taps into primitive ideas about the connectedness of different life forms, in particular of human societies and mountains, combining god-like scale and a multitude of petty human devices like cranes and hoists. It depicts a faltering technology and a huge and concentrated effort that will set human civilisation back a stage or two via burgeoning misunderstandings. But the fact remains that someone has imagined a symbiosis, though in ruined form, between geology and building, the one growing into and out of the other, like a weird actualisation in the 16th century of the creation myth in which the largest distinct natural form, a mountain, gives birth to the full later complexity of species and cultures, like a comprehensive explanation of what we are all doing here.
Tremors in consciousness provoked by that much later composition together with the Cambodian creation stories can help us understand what Indian architects might have been driving at in searching out solid masses of rock near the surface, signalled sometimes by the caves already tunnelled through them by slippage or erosion, in which with minimal removal they could discover buildings.
It was never a high proportion of Indian religious buildings which were made or half-found In this way, but they had an imaginative force out of all proportion to their numbers. Whenever you come across them, they take you back to the mythical origins of architecture, spaces found not made, and then brought up to the surface and into the light. That is the direction we imagine such spaces heading in, but for us the excavated temples usually speak strongly of a darkness we have mostly left behind, which it seems part of the task of the temple, whether rock-cut or not, to drag us part-way back into.
The first time in India our only rock-cut temple was only partly discovered in the hill. Most of it was added onto the cave-bit, so the whole effect was like the tower of Babel, built bits merged with more primitive elements to make a patchwork whole, all of which resembled bricolage, a hybrid tumbled together like a rock fall, not entirely stable.
The temple lay at the foot of the hill as if partly hidden by scree which had slid downward as the hill eroded. The entry porch and the mound rising behind it didn’t look as if they were all in the right order, but scrambled, as in a half-collapsed structure.
Crawling round the interior was a powerful experience. I’m not sure you could follow the ambulatory passage the whole way round. At a certain point your way was entirely cut off after you had crouched or crept through the lowest bits. Certainly you were bothered by the bees. They had set up their hive in the furthest reaches and came and went continually, their buzzing amplified by the vault.
The plan in the Encyclopaedia of Indian Temple Architecture gives such a bland idea of this dangerously impeded interior and doesn’t attempt to show architecture turning abruptly into crags along the temple’s right flank or at its west end.
But the whole force and value of the Maladevi temple at Gyaraspur, which makes it a great beacon among all the buildings I have seen, is this uneasy truce between the violence of geology and the ingenuities of architecture.
If I had it to do all over again, I would go on to Gwalior (as we did) and make a stop at the little Caturbhuja shrine in the Fort (as we didn’t) to gauge how the raw power of rock makes itself felt in a rock-cut building the size of a plaything.
Next I would stop at Dhamnar (a substitute for the grandest of all rock-cut temples at Ellora), an instance of the fascinating type that finds an entire world below ground level, ground level which still exists on every side at Dhamnar, where eight temple buildings form a tight cluster, a main shrine and seven complete children of the parent, which each possesses all the parts of a temple on a reduced scale. Or I would have done this in 2001, but I am not sure I would now, because this complex has apparently been renovated by drastic cleaning and the addition of a protective coating that contains a lot of brick dust, which gives it an orange colour, most un-stone-like, like the healthy glow favoured by failing Presidents.
Even in the old days the buildings at Dhamnar were rough and raw in a wholly different way from Gyraspur. Sculptural detail had the smudged look of attempts in very hard stones like granite, but here it was the stone’s softness that had made it easy for time to erase all sharpness, until you felt the day looming near when it would all disappear. Hence the well-meaning renovator, who didn’t want to hear that he had replaced a beautiful ruin with a lifeless model born yesterday. Did he know how the building was made? Turning it to brick was such a cruel travesty.
My next stop (on my would-be journey) is safe from such destructive interference because it was left so incomplete that the effect is like camouflage. In this group of magical buildings it is perhaps the most magical of all.
It is like a sketch for a large temple complex more begun than completed, blocked-in lightly across the whole site, so it is all there and full size, but barely detectable. Perhaps uniquely in the whole history of architecture, this temple group at Masrur in the Himalayan foothills preserves the natural inspiration of the building and even the full value of its magical materials before they are spoiled by being squared up and smoothed, yet conveys the entire architectural concept in a shapely and complete form as well.
It is both a building and a vivid landscape, a mountain range bristling with crags and a symmetrical city of towers, an ideal vision like a Chinese landscape representing heaven, and a whole world of natural rock always entirely itself and (almost) nothing else, the most natural as well as the most perfect temple.
Some readers may think they’ve already seen something like this in their local zoo, artificial crags constructed for mountain goats to climb on in captivity. But the distance between constructed and actual crags is unbridgeable. Not that hybrids can’t contribute something to the discussion, like the miniature rock-cut pavilions at Mamallipuram lined up in a row as the outcroppings seem to have allowed, with the quaintest indication that these were carved from the top down and (in at least two of them) left deliberately incomplete so you couldn’t miss the point. Bhima ratha and Valayankuttai ratha turn back into wild rock for a last few moments before they reach the ground, which makes them at one and the same time, levitating architecture and a natural growth rooted in the earth, a botanical/geological marvel giving birth to a strange child, the phantasm of civilisation.
Instead of purely human constructions these are Eternal Forms like those which emerge on the walls of caves as a teeming population, buildings something like creatures with their own internal principles of life.
In a sense it’s wonderful there is no end to the territory and no single logic according to which it is laid out. At the other end of the field of possibilities from temples camouflaged as mountains are seemingly overplanned complexes leaving nothing to chance, which look in plan more like wiring diagrams than rich plastic compositions binding together their widely dispersed elements. Among examples of this type, both monotonous and scattered when seen first in plan, the so-called Harihara temple 1 at Osian in Rajastan stands out, ‘so-called’ because the interesting dedication to Harihara the bifurcated deity, who suits the site which can’t consolidate or make up its mind, this name has been retracted for something blander.
The bases of Indian temples are one of their most distinctive features, elements more central to the building’s way of being than any equivalent in Western buildings, elements which often attract careful diagrams in the Encyclopaedia. Bases come with many stages and bristle with Sanskrit terms in the Encyclopaedia entries.
The platform-bases of both Harihara 1 & 2 pile up seven distinct stages, like multi-storey structures in miniature. all of which is distinguishing the building from its setting in the world and asserting its essential complexity. The platform-base at Harihara 1, which is like an elephantine enlargement of the bases of its component shrines, has the unity and coherence of a whole symmetrical cosmos made of clearly marked layers and dotted with architectural miniatures, niches which contain their own versions of walls, roofs, thresholds, openings and inhabitants. The resident spirits of the Harihara temples are the figure sculptures which appear three to a side on the walls of the platform, and then at least five to a side on walls of each of the five shrines planted in the peculiar symmetrical system on the roof of the platform.
The Encyclopaedia includes neat little charts of the sculptures placed on the four cardinal aspects of the building, charts which take for granted that these layouts carry crucial meanings. So I found myself becoming obsessed with pinning down who was looking out from where, as I reconstructed a visitor’s journey around this multitudinous complex.
To begin with, this exercise required being sure of the compass points. Hindu temples normally face east, contrary to the usual orientation of Christian churches. Anyone who deals with the plans of western religious buildings gets used to finding the east end at the top and the west entrance at the bottom, north to the left, south to the right. With Hindu buildings these norms are reversed. Except that a few important Eastern buildings, Angkor Wat, for instance, Kailasa at Ellora, and Harihara 1 and 2 at Osian, face west.
This anomaly has caused confusion in the Indian Encyclopaedia, where the charts of sculptures on the Harihara shrines show the sequences of deities on all four sides of the platform and the five shrines reversed from their actual order.
I started out not knowing what some of these deities should look like and only began to notice that they weren’t in the right places when the elephant’s trunk appeared on Kubera not Ganesa, and the boar’s snout on Buddha not Varaha. At this stage it felt exactly like solving a puzzle, and no one would believe the satisfaction I got from putting Buddha in the right place. But Buddha on a Hindu temple? Was it tolerance or co-opting?
How pleasing to learn that the very same order is repeated on Harihara 2, not the attitudes and emotional tonalities or relations between other creatures in the scenes, but the basic sequence of deities was the same, so the content must be at some level deeply valid, and therefore it was probably a structure worth pondering. It took a long time to dawn on me that Harihara 2 also has the same floor plan as Harihara 1, but with a large intruder plunked down in the middle of it, an overscaled colonnade made of diverse column-forms which allows the roofing-in of the open space between the shrines. I felt let down by Michael Meister, my favourite among the different Encyclopaedia authors, above all for his responsiveness to natural settings and his appreciation of all kinds of architectural novelty. Why didn’t he announce the startling alteration in Harihara 2 more emphatically?
Why didn’t he make plain how radical it was to stick this heavy awning on an essentially outdoor space? And why weren’t there any photos showing how the new elements collided with or related to the existing shrines, which they treated as buildings within a much larger building, where they were now lost or marooned or holding court in a sort of surreal parody? Which was it? Were contradictions exaggerated or suppressed, enjoyed or disguised out of existence? One of the really explosive moments in Hindu architecture had been slipped past us unawares, a missed opportunity which made me wonder if my hero hadn’t been paying attention. But there was proof that he admired these buildings tremendously, so he had certainly noticed.
One of the pleasures of Harihara 1 is the assembly of five (four surviving) exquisite separate works into a new whole. Harihara 2 gives up those pleasures to make a more imposing singularity, or is it a more ungainly diversity? You would need to go there to decide which. In any case architect no. 2 wasn’t content to repeat. I still miss the photos capturing the bold new spatial effects where the canopy meets the shrines.
Your analogy or model for the process, that it is like solving or putting together a jigsaw puzzle of separate pieces, is faulty and much too confined for what is taking place, because a jigsaw has one answer and follows a narrowing process to a goal that is almost meaninglessly clear and definite. This other process is a loosening and tightening as you go, sometimes a limited task like identifying all the figures which swarm on the outsides of buildings, then finding that the sculptures on two related buildings follow the same sequence and can help solve each other, because different ones are recognisable in each, and others are obscure, and some are missing entirely, or not in their proper places but lying some distance away, like the semi-human creature planted temporarily in a blank space on the back wall of Harihari 1 at Osian (Harihara, who isn’t the single deity linking all three of these buildings–Harihara temples 1, 2 and 3–after all).
Chasing the Hindu stories round the outsides of these buildings can seem a childish activity. In truth we are now reduced to chasing them mainly through photographic archives of disconnected views, shattered but in some odd way a more continuously sensuous activity than many visits to actual buildings can consistently be. The photographer is making choices, continually selecting. And leaving out the wider context can result in more intense experience.
I find myself thinking of a remarkable French series that concentrates on obscure Romanesque buildings in obscure or at least very particular corners of (mostly) rural France. In some sense it is literally true that I have never been closer to the textures of carved limestone than when transfixed by the black and white photos in these modest-sized books.
‘Black and white’ isn’t good enough. These were images printed in heliogravure and bled off the edges of the page, leaving no room for captions or other distracting words diluting the confrontation with all the tones between light and darkness, glare and shadow in all their heights and depths, in a total concentration on the grain of the stone, the scuffs and breakages that describe its life over time, the contest between tools and the rock’s varying resistance, between the slow taming of mineral surface by wind and water, and the bursting forth of rude ideas about animal energy, and emotion crossing or breaking out on human features, all this filtered through a photographer’s eye, who’d been brought up on early modernism which had played havoc with religious belief.
For the Zodiaque series of Puritanical (in the best sense) treatments of Romanesque buildings were the brain child in the first place of a single Jesuit monk, trained as an artist and a priest, who combined these two strands in strong forms to produce (with committed collaborators) one of the most compelling visions of a phase in the story of art, especially vivid and alert to forces beyond a narrowly rational view of human culture and especially of animal life, a spirit it would be apt and inspiring to bring to bear on Indian architecture and sculpture of an equivalent period to the French Romanesque.
Perhaps our best hope of such an encounter lies in the photographic archive of around 120,000 images of Hindu temples assembled by the American Institute of Indian Studies, mostly in the late 1960s and early 70s. I haven’t tried to pin down images to particular photographers but have picked out a few that come closest to those in the Zodiaque series, especially the volumes in which Dom Angelico Surcamp took a sizable part.
To try and distinguish different photographers’ contributions in the Indian pictures is a project beyond me at the moment. So far I’ve barely thought of these Indian photos as works in themselves, but used them to understand the buildings.
But the Hindu stories—there I am still at an early stage. I come across ‘Natesa’, and after 4 or 5 occurrences I realise it’s a name for Siva, meaning ‘Dancer’, which is attached to him when he’s quelling demons by dancing on their heads.
I keep seeing Nrsimha, a god in the form of a man-lion, with a much smaller creature– human with an animal head?– the images too ruined and me too inexpert to make out these figures clearly. Lion-man seems to be tearing the little person open and letting his innards spill out. The lion-man must be an aspect of one of the main gods, given the prime positions he is awarded. I am putting off looking him up.
I get a kick out of Siva dancing on the heads of demons. I am fascinated but appalled at Nrsimha sitting there calmly eviscerating a child-victim. A few days ago I watched a film which E soon realised she didn’t want to see called Map to the Stars that was loaded with the exhilarating crudity of Greek myth. Children were doing awful things, setting fires in which they accidentally burned up themselves as well, strangling smaller children across their knees (cf Nrsimha) in Portacabins. Therapists acted out their clients’ fantasies, crouching over them like predators. These events took place mainly in Hollywood, and star maps showed you where the huge egos of film had their castles. Stardom was obscene and mysterious, yet had some connection to the heavens. All the grisly violence wasn’t just senseless. Hindu gods can also seem quite un-benign, but you need to know about them, and feel as you learn that you’re in touch with something that matters.
I am late realising that E and I are embarked on similar quests. She is working her way through a limited number (a quarantine, as it happens), 40 holy men (including only a token scattering of holy women) from all over the world, a number to which there is an end, which she can break down into a compassable number of distinct tasks and can even take a week off to do more pressing work, an inventory of an existing population, the furniture and ornaments of the house, or something which has a fixed terminus, like a gigantic shopping list, a survey in its way of all creation, but one which has a submission date by which it will be done, whereas mine keeps expanding from the dimensions of a single blog post to that of a book, or a couple of them, as if in cataloguing certain contents you went on discovering further series of rooms in a ramifying structure which kept on growing like Topsy, like a god who developed new limbs to accommodate new functions or new tasks, which were or became new identities, so the total number of gods might range anywhere from three to three hundred million.
[Excursus on Kanchipuran and Pattadakal] Here was meant to come a brief treatment of the Kailasa temple at Kanchipuram in Tamil Nadu, a fascinating instance of a building as a mountain that I was put onto by a witty drawing of my friend Adam Hardy’s, which clarifies the organisation of this super-intricate, angular ‘mountain’, built of an impossible number of sub-units each complete in itself with a final result like an enormous, many-faceted lump of quartz. The drawing brings out the cartoon-like quality in the battery of horrific and comical lions who follow the twists of walls intensely indented, like an abstract rendering of rocky crags.
I intended to follow Kanchipuram with a treatment of the Virupaksha temple at Pattadakal, so the imaginary journey would begin and end with buildings I had actually visited, and Pattadakal would allow a final summation of the plenitude of Indian architecture, the whole human and natural worlds collected and summarised on the outer surfaces of a single building.
This plan came a-cropper through a discovery that seemed at first a miraculous validation. Somehow I came across a description of Ellora, the biggest and best of all rock cut temples, which derived it from the very buildings at Kanchipuram and Pattadakal I had chosen for purposes of my own. The account was even embellished with a sick king and his fasting queen, like inhabitants of a fairy tale, and I was off on a lengthy burrowing in the complexities of Ellora, plentiful sculptures, decorative innovations (the Rococo many centuries before its time), undreamt-of forms. Would I never be done? Every ending sprouted a further beginning. Except that this time the link between Ellora and the other temples was a fantasy, and the story of the building finished before its initial courses were laid, saving the queen from wasting away, was an opportunistic appropriation of an architectural paradox.
The number of interesting old temples in India kept growing and was the most numerous population anywhere in the world, which would only be manageable if I were 20 or 30 or 40 years younger and could fit in 5 or 10 or 20 annual trips to keep up with the expanding and deepening field. The Shell Guide to English Parish Churches might be the template – surprising it took me so long to notice the parallel, or Pevsner’s twenty-four years covering England which began at least ten years before it surfaced in a form visible to anyone else. In some sense this was the ideal ancestor, which kept popping up or beckoning, seeming to stand for any sustained human effort, a plan so ambitious it encompassed an entire place, a large island that resembled a continent, a task so huge it was probably not do-able. I set forth on my truncated version of such a task, consuming two years, not twenty-five, which still became a trap I was dying to get free of, as now I regretted being still a prisoner of Indian temples after almost two months.
The beginning of this obsession was lost in the mists of one of the intensest and at the same time blankest periods, when I could hardly leave the house or escape an isolation that would perhaps never end, except that here ‘never’ meant only a short span, a year or two until you inadvertently caught the disease you wouldn’t survive. Life had become both a nothing and a gigantic cosmic allegory, like the ones medieval folk went around thinking they had always been engaged in.
So that was a kind of template and its content was a series of temples, dictated by an unfathomably complex series of examples in a couple of books organised according to a series of local rulers who were locked into an extremely foreign geography or a history of exotic styles and the shifting stimulus of a big collection of images which all sat in four over-lapping volumes you kept picking up in no fixed sequence. Four was just enough to feel unencompassable like India, though it covered only two arbitrary blobs of territory over a not easily identifiable set of years Far Away and Long Ago. (The title of a book which bewitched me when I was just beginning to read on my own.)
![34 ptg twombly untitled [say goodbye catullus to the shores of asia minor]1994 room w huge mural menil houston.jpg](https://robertharbisonsblognet.files.wordpress.com/2020/01/34-ptg-twombly-untitled-say-goodbye-catullus-to-the-shores-of-asia-minor1994-room-w-huge-mural-menil-houston.jpg?w=739)
The work consists of another steep ascent and precipitous fall. A slender frame contains an exuberantly molten platform-mound of plaster heaving with life, but the overall impression is something like a guillotine waiting to descend. The childish quality of Twombly’s inscribings makes me think of a-semic writing, writing that looks like words but isn’t, a mode with which Twombly filled whole canvases in certain phases of his career.
From certain angles the ‘feet’ of the giant look covered in feathers, and the work’s whole effect seems one of the most disunified ever, diverse as only forms produced by a centrifuge could ever be in the real world.
Paolozzi himself drew attention to the diversity of his sources, even reading a list during a lecture at the ICA to show the whimsical range of all the various objects which had caught his eye. It’s amusing and deliberately alarming, but it sets us barking up a lot of wrong and non-existent trees.
The trouble is that none of these things are there any more in the sculpture, only impressions such as you might capture in hot wax or castings in a single material that levels out the variety, as if you had buried them all in the same earth (or metal, for they are now all uniformly a messy, unbeautiful bronze). So they are like the things in Wallace Stevens’ poems, tantalising ghosts of their sisters in ordinary reality or even worse, barely recognisable, partly overlaid by something else, no longer nameless because turned to liquid and run out across the flat background sheet. And many of Paolozzi’s ‘things’ are only parts of things—handles, tubes, eyes (as in hooks and eyes), washers, circuit boards, many of them only vaguely familiar to un-mechanical man.
In some sense it is a true entry into this hidden realm of Paolozzi’s activity to plunge right into the phantasmagoric textures without allowing an overall orientation to start with, but it is also a misrepresentation because you do recognise the figure before you get swamped by the detail, which may be the essential experience of these works, but isn’t the starting place.
So I set about naming the strange beings: limping man, hideous puckered man, triangulated man (or lopsided man, semaphore man, glued-together man, splat-man—all names for one of my favourites, so a good place to start). He is off-centre, deliberately so, and seems to be sliding sideways. I can’t explain why this unworkable geometry is so compelling, or why I love the idea of an uncountable number of pieces so unreliably bound together.
There’s Robot, whom I know as toga-man or Roman senator, who unfurls a scroll in front view but shocks us from the side and behind where he looks uncannily like Snoopy the cartoon-dog, but then you notice punctures in the dog’s head which allow you to thread the object like a Chinese landscape on a scroll, where you get lost in a series of miniature interiors.
There are sculptures to which accidents seem to have happened, like Shattered Head, for whom I’ve invented a narrative, in which he was intact and harmonious to start with, but was dropped on a hard surface and smashed, after which he was carelessly reassembled, so that the openings in his face are no longer in the right places, but we read them as eyes and nostrils anyway, now grotesquely misplaced as we have sometimes seen with badly wounded veterans.
There’s even one called Damaged Warrior, ambiguous name—is it the sculpture or the man who has suffered? He is Truncated Man, sliced in two by a bomb or by the artist’s decision, but how could you choose to cut this torso in just this way?
I can’t remember any work by other sculptors which goes further or gets separated more radically from likelihood. Yet Paolozzi soon grew tired of the endless transformations. Perhaps the various discontinuities are too great to go on thinking up new ones forever, and there’s an almost inevitable urge to return to the world of everyday possibility, but while it lasts, Paolozzi’s 5-year excursion into fully three-dimensional surrealism is without equal.
And then, something unexpected, which I mistook for a nineteenth century replica, a set of spindly wooden stalls covered in carving, a thicket of close-packed detail I mistook for recent because the wood looked new and yellowish, like fresh oak that hadn’t had time to mellow. Later I found an old description that complained about ‘the rich treacly brown the nineteenth century liked’, now (in 2019) banished by aggressive cleaning, which had removed every visible sign of the centuries’ passage.
There are those thickets of carving, but the Winchester stalls consist primarily of a lofty fictitious architecture towering overhead, decorated with perforated elements looking through to further spaces closed off by miniature vaults. First architect-masons devised novel methods of spanning odd-shaped spaces and then their successors made vaults into intricate terminations for spaces primarily conceptual.
In the 1970s Gordon Matta-Clark went round carving out new spaces in existing buildings, like chantries in reverse, or voids as independent structures, which it was only possible to discover lurking in nondescript industrial or commercial buildings because they were already corpses, that is to say, scheduled for demolition. A negative can’t ever be simply a positive, but Matta-Clark had remarkable successes in these figure-ground experiments that made absences feel stronger than presences. The effect could seem almost metaphysical, and the non-existent acquired inescapable reality. Perhaps the chantries too are always flying in the face of fundamental facts, and like fables or myths able to make you believe in something which isn’t entirely there.
Something similar operates with the so called ‘fragmentary mask’ below, like an animal’s skull abandoned in a field, that’s now brought indoors where every tremor of its surface becomes full of meaning.
Some are cartoon-illustrations of the history of architecture, like the one called ‘lintel shadow’ which projects a monumental gateway like those three-part compositions at Stonehenge but taller and spindlier, in one sense more imposing, in another more precarious. It led to a ‘stone’ enclosure like an introduction to an underground tomb. It led ‘underground’ or nowhere, and fit the idea of a shadow of architecture by being out of true in every axis and every dimension. It was stony in its form–big lumps, scored with accidental grooves and gouges, which lent a kind of ‘authenticity’ to the ruined masses, yet also made you suspicious. As you got nearer, a sliver of the air beyond appeared between adjacent ‘blocks’, which were coloured a convincing mottled gray but gave out a hollow sound if you tapped them.
They were a figment or a fiction, an insubstantial shadow. Doubtless the lintel too, far out of reach, was a partly convincing fake, hoisted up on rickety poles which had had to be extended by bolting smaller pieces to them with crude splints, our first encounter with the sculptor’s habit of flaunting a few ‘mistakes’–revisions or changes of direction she preferred not to smooth over or clean up. ‘Admitting your mistakes’ has here a wonderful feeling of being at ease with your materials (no quotes) and your project. It evolves, and your audience can watch that happening.
Some of the solidest elements are the shadows, especially the sloping platform ‘cast’ by the massive ruined column called ‘barrel’ in Barlow’s title for it (all her titles follow directly after untitled:– ‘untitled: barrel’). This looks like a waffle-structure in metal, or a model of a curving three-storey block of modernist flats, except that unlike other shadows it is supported by unstable poles driven into swampy ground and poking through the surface of the swamp so crookedly you lose faith in them completely. I had enjoyed the messy punctures in the fibrous board that constitutes the horizontal (but sloping) surface, calling up a forlorn watery landscape.
The most entertaining conglomerate is saved for the last of the three rooms. It is called ‘blocks on stilts’, which doesn’t begin to do it justice. It consists of four towers (you will have to count them more than once before you believe there are only four, and you will think you have disentangled them only to find they have mixed themselves together again). In some sense it is a simple idea, a set of four-legged frames, each of them existing to raise one impossibly bulky rectangular-solid impossibly high.
Some time after finishing this, I remembered my calculation that there are 12 legs in ‘blocks on stilts’, so there must be three towers, not four. But in another sense there are four—when you are there, the parts are magically multiplied and counting them doesn’t settle the question, strictly speaking.
robert harbison's blog 