The Atlas Collection
Travel posters
For places worth remembering. Modern prints inspired by favourite coastlines, towns and landmarks.
New destinations added regularly
Posters from the places you love
Over 320 curated artists
Browse all artists
Curated by Artists
Museum Quality Prints
Crafted to Last
Designed for Interiors
Style It Your Way
Popular prints available framed or on canvas
Timeless Art For Beautifully Designed Homes
Why Choose Us ?
What customers say
Over 5000
Happy Art Lovers
From Our Journal
Along the River During Qingming, the Song Scroll of 800 People
In the early twelfth century, in the Northern Song capital of Bianjing (modern Kaifeng), a court painter finished a handscroll 25 centimetres high and about 5.25 metres long. Unrolled section by section, right to left, it shows the city and its river in one continuous view: open fields give way to suburbs, suburbs to a crowded bridge, the bridge to a gated wall and a market street beyond. A modern count puts the human population of the scene at 814, alongside dozens of boats, pack animals and buildings. The scroll carries no signature and no date. Both the painting's title and its attribution to the painter Zhang Zeduan come from a single note, written by the scholar Zhang Zhu in 1186 and mounted onto the scroll as a colophon, decades after the paint had dried. A painting this size, this old and this precisely populated would be remarkable if it merely survived. By modern count it shows 28 boats, 60 animals, about 170 trees and some 30 buildings, and, at its centre, one bridge that very nearly fails. That it survives at all, carrying this much information, makes it less a single image than a continuous record, made by someone who had clearly watched before he painted. An Unnamed Painter, Named Later Zhang Zhu's colophon is short. It states that the painter, Zhang Zeduan, was a native of Dongwu (present day Zhucheng, in Shandong). He studied as a young man, then travelled to the capital and took up painting, excelling at jiehua, ruled-line painting done with a straight edge rather than free brushwork: boats, carts, bridges, market streets, moats and city walls. Zhang Zhu cites an earlier catalogue, Mr Xiang's Evaluation of Pictures, which places two of the artist's works in the "divine class", the highest rank available. One was a Regatta on West Lake, not otherwise known today. The other, named only by title, fits this scroll better than anything else Zhang Zhu mentions. That is the entire biographical record. Later reference works give Zhang Zeduan dates, 1085 to 1145, and a post at the Hanlin Painting Academy under Emperor Huizong, but these rest on inference rather than on any surviving document. Much of what might have confirmed or corrected them was lost a few decades after the scroll was painted, when Jin dynasty armies overran Bianjing in 1127 and the Song court fled south. The painting may be a rare survivor of exactly the world it shows. Even the title is less settled than a museum label suggests. Zhang Zhu wrote his colophon on the day after the Qingming festival of 1186, which may be no more than an accident of timing, though it has encouraged readers who want Qingming to mean the spring festival of tomb sweeping rather than the alternative sense, peace and brightness, a stock phrase for good government. Scholars have argued for both readings, and for a deliberate double meaning. Zhang Zhu did not resolve the question; he simply supplied the title that later cataloguers kept. A Painting Built to Be Unrolled Along the River During the Qingming Festival is painted in ink and colour on silk, and it was never meant to hang on a wall. A handscroll is stored rolled up in a box and opened a stretch at a time on a table, read right to left the way Chinese text runs, and rolled closed again behind the viewer's hands as a new stretch opens ahead. There is no single vantage point from which the whole composition can be seen at once, not even now, when a photograph can flatten the whole 5.25 metres onto one screen. The route follows the river outward from open country. Willows and a train of pack animals appear first, then the outskirts of the city: inns and a boat under repair, farmers bringing produce in by cart. The pace quickens as the buildings close in, building toward the painting's acknowledged centre: a wooden bridge arcing over the water, thick with pedestrians, above a river crammed with boats. Past the bridge, the scroll runs through a city gate and along a street lined with shops and stalls, before the paint stops and the silk ends. Shops, Doctors and a Line of Camels What the scroll records is commerce, not ceremony. Restaurants advertise themselves with cloth banners and timber signboards. Wine shops, noodle stalls, a doctor's clinic and a fortune teller's table appear along the main street, each with customers and staff occupied in specific, legible tasks rather than posed for the painter. Porters carry loads on shoulder poles. Ox carts and handcarts compete with sedan chairs for road space, and a line of camels, heavy with cargo, files through one of the city gates, a reminder that Bianjing's markets were fed by overland trade as well as by the river. A camel train and carts pass through the timber city gate of Bianjing into the market street, a detail from the Song original scroll. The river itself was a working waterway before it became a subject for a court painter. The Bian canal linked Bianjing to the network that carried grain from the south to feed the capital, and the scroll shows the traffic that depended on it: barges being loaded and unloaded alongside a boatyard where rope and rigging are under repair. Zhang Zhu's colophon singles out boats and bridges as the artist's particular strength, and the density of nautical detail, different hull types, the way a mast is raised or struck, supports the idea that whoever painted this had watched real boats closely before drawing one. The Bridge That Nearly Failed The scroll's best-known passage is a single moment of near disaster. At the painting's midpoint, a wooden bridge without a single pier in the water arcs across the Bian river in a single span of about 20 metres, built from interlocking timbers pinned and wedged together rather than nailed, a technique sometimes called beam weaving. Contemporaries called it the Rainbow Bridge, Hongqiao, for its shape. In the painting, a large boat approaches with its mast still raised, seconds from striking the underside of the arch. Boatmen shout and haul at ropes. A man leans from the rail to pass down a line, while a crowd presses in along the bridge to watch. The bridge was real enough to rebuild. In 2000, the bridge scholar Tang Huancheng, working with a team from MIT for a television documentary, reconstructed a span using only the joinery visible in the painting: no nails, no piers, interlocking timbers carrying the load through friction and geometry alone. The finished bridge, built by hand in a Chinese village, held the weight of a water buffalo without strain. A nine-hundred-year-old painting had supplied a working engineering drawing. Why It Is Called China's Mona Lisa The comparison is a modern one, and it does not rest on shared subject matter or style. The Mona Lisa is a single face; the Qingming scroll is a crowd of 814 individually posed strangers. What the two paintings share is a fame that has outgrown ordinary art history: both are the one work most people can name from their tradition, and both draw crowds who queue for hours to see an object that photographs already show them in full. The Chinese comparison also carries a note the Louvre's painting does not. The scroll has been read for centuries as a document as much as a picture, cited by economic historians for shop layouts and by naval historians for hull design. Leonardo's picture keeps its secret (why does she smile) with total economy. Zhang Zeduan's scroll works by accumulation, 814 separate lives rather than one ambiguity, and its fame rests on being exhaustive rather than enigmatic. That is arguably the harder trick. The Qing Court's Version, Enlarged The original scroll was copied and reinvented throughout the rest of imperial history, each new version updating the fashions and architecture to its own period while keeping the title and the same journey from country to city. The most elaborate remake was finished in 1736, the first year of the Qianlong Emperor's reign, the work of five court painters: Chen Mei, Sun Hu, Jin Kun, Dai Hong and Cheng Zhidao. Their version runs to about 11.5 metres, more than double the length of the Song original, painted in stronger colour and finer, more even brushwork suited to Qing taste rather than Northern Song habit. This Qing court version now belongs to the National Palace Museum in Taipei, part of the imperial collection moved from Beijing during the Chinese civil war. It has been exhibited and reproduced more often than the fragile Song original, and for many visitors it is the version they meet first: more colourful and easier to photograph from end to end. A Painting Too Fragile to Show The Song original stays in Beijing, in the Palace Museum inside the Forbidden City, and it is shown rarely and briefly. Nearly a millennium of rolling and unrolling has left the silk fragile, and conservators limit its exposure to light and to the physical stress of display. When it has travelled (to Hong Kong in 2007, to Tokyo in 2012) queues formed for hours, and viewing time per visitor was cut to seconds on a moving walkway. For most people, in China and outside it, the scroll is known through reproduction: prints, screens, textbook plates, and now high-resolution scans that let a viewer do on screen what the silk itself can no longer safely do very often, unroll from end to end. The Same Dynasty, a Different Register It is worth setting this scroll against the other object the Northern Song is best known for among collectors: its ceramics. The celadon bowls and Ru ware of the same court, the same decades, pursue almost the opposite effect: plain surfaces and a single glaze, nothing added for the eye to read as a story. Zhang Zeduan's scroll is the crowded, noisy counterpart to that restraint, proof that the dynasty which prized a bowl for what it left out also produced a painter who left almost nothing out. Read side by side, the two traditions describe one court from opposite ends: the quiet of the kiln and the noise of the river. Art of China Along the River During Qingming Chinese Landscape Painting Art of China I: Neolithic Pottery & Jade Art of China II: Shang Bronzes Art of China III: Han Silk & Lacquer Art of China IV: Tang Gold & Silver Art of China V: Song Porcelain Art of China VI: Ming & Qing
Sharaku Made 150 Actor Prints in Ten Months, Then Vanished
In the fifth month of 1794, kabuki fans in Edo started seeing a name they had never encountered before: Tōshūsai Sharaku. Over the following ten months, ending in the first month of 1795, the publisher Tsutaya Jūzaburō issued somewhere around 140 to 150 print designs under that name, almost all of them portraits of kabuki actors, with a few sumo wrestlers and warriors included. Then the prints stopped, and Sharaku vanished. No later work bears his name, no contemporary diary mentions him, and his identity is still argued over today. The strongest candidate is a minor Noh actor called Saitō Jūrōbei. Ten Months, One Publisher, About 150 Prints Japanese lunar-calendar months do not map cleanly onto Western ones, so scholars date Sharaku's debut to the fifth month of Kansei 6, which ran from 29 May to 26 June 1794. His opening batch was twenty-eight ōkubi-e, or large-head bust portraits, in the ōban sheet size, tied to plays running that month at Edo's three licensed kabuki theatres: the Miyako-za, the Kiri-za and the Kawarazaki-za. All twenty-eight were printed against mica grounds, most of them dark grey or near-black, that made the actors' faces glint under lamplight. Every print in the Sharaku catalogue, without exception, was issued by one man: Tsutaya Jūzaburō, the leading print and book publisher in Edo. No rival publisher's name appears anywhere in his output. The catalogue itself is unusually short. Reference works list him simply as "Tōshūsai Sharaku, active 1794 to 1795" because that is all that can be proven: a working name attached to a run of prints between two fixed dates, nothing before and nothing after. Even the total is an estimate. Specialists have counted as few as 140 and as many as roughly 150 designs, depending on whether preparatory drawings and disputed sheets are included in the tally. Faces Built to Unsettle The twenty-eight ōkubi-e are what made Sharaku's name, and what still draw crowds to exhibitions of his work today. Take the actor Ōtani Oniji III, shown as the servant Edobei in the play Koi Nyōbō Somewake Tazuna, staged at the Kawarazaki-za. Sharaku crops in tight on the face: a jutting nose, a curled lip, fingers spread and hooked as the character gropes toward his victim. It is not a flattering picture. It is a picture of a man performing villainy, drawn by someone more interested in the psychology of the role than in the vanity of the actor playing it. Sharaku applied the same scrutiny to heroic roles, not just villainous ones. In a companion print from the same production, Ichikawa Ebizō appears as the samurai Takemura Sadanoshin in the moment before he takes his own life to preserve his honour, hands clenched, eyes strained under the weight of the decision. Earlier designers working in bust portraiture, Katsukawa Shunkō and Shun'ei among them, had stayed closer to flattering convention. Sharaku's contemporary Utamaro was doing something related in his portraits of women, reading mood and character rather than settling for prettiness, and Sharaku pushed that same instinct further, to the edge of caricature. Tōshūsai Sharaku, Ichikawa Ebizō as Takemura Sadanoshin, 1794. Printed from the same May 1794 production as the Edobei bust. A Falling Off, Batch by Batch After the ōkubi-e, Sharaku's output declined in stages that scholars still use to divide his career into four periods. A second batch of thirty-eight designs followed in the seventh and eighth months of 1794: thirty half-length portraits on plain yellow grounds in the narrow hosoban format, and eight pairs of actors kept on the white or dark mica grounds of the debut series. A third batch, in the eleventh month of 1794, moved into hosoban and the mid-sized aiban format and is generally judged weaker again. A fourth and final batch, fifteen hosoban prints from the first month of 1795, carries a different signature form from the rest and is the least admired of the four periods. Somewhere in this run Sharaku also produced two warrior prints and at least one sumo design, minor by count but proof that Tsutaya was still willing to try him in other genres before the commissions ended for good. Why the Prints May Not Have Sold The oldest explanation for Sharaku's disappearance is commercial: his prints stopped selling. Edo's actor-print buyers generally wanted a favourite performer flattered. Sharaku gave them faces distorted by effort, vanity or malice, and later writers assumed that actors, their fans, or the theatres themselves objected. Tsutaya had reason to be careful about objections. Three years earlier, in 1791, censors enforcing the Kansei Reforms had fined him half his personal wealth over books he had published by the writer Santō Kyōden, ruled to have broken the new decency laws. A publisher already punished once for pushing too far had a clear motive to drop a designer causing him trouble a second time. Recent researchers are less certain that unpopularity explains everything. No sales record survives, and no contemporary complaint documents actors or audiences reacting against him; the idea rests mostly on inference from the prints themselves. Other explanations compete with it: that Sharaku lost the co-operation of the actors he depicted and so could no longer sit close enough to draw them, or that Tsutaya simply redirected his money toward other designers, Utamaro among them, once the novelty wore off. None of these theories has displaced the others. What is certain is that after the first month of 1795, the commissions ended. Fifty Years of Silence, Then a Name For half a century after 1795, nobody wrote down who Sharaku was. The earliest surviving attempt appears only in 1844, in an expanded edition of the Ukiyo-e Ruikō (Ukiyo-e Miscellany) revised by Saitō Gesshin. It gives Sharaku's real name as Saitō Jūrōbei and an address: Kyōbashi Hatchōbori, in what is now central Tokyo. That is the entire biographical record. No portrait of him survives, no family history, no explanation of why a name attached to roughly 150 prints should surface only after fifty years of silence. Ignored at Home, Rediscovered Abroad Sharaku's reputation did not travel well within Japan. Through the nineteenth century, Japanese writing on ukiyo-e treated him as a minor figure beside more prolific and more popular designers. What changed his standing was a book published in Munich in 1910: Sharaku, by the German art historian Julius Kurth. Kurth set the actor busts beside Rembrandt and Velázquez, arguing that Sharaku belonged among the major portraitists of any tradition, not just among ukiyo-e specialists. The book arrived at the height of Japonisme, when European collectors were already assembling print collections and looking for the next name to champion. A second, expanded edition followed in 1922. Japanese scholarship caught up afterwards, reclaiming an artist whose reputation had, by then, already been made somewhere else. The Noh Actor Theory Saitō Gesshin's 1844 identification remains the theory most specialists take seriously, and later research has added circumstantial detail to it. Saitō Jūrōbei was a Noh performer employed by the Hachisuka family, daimyō of the Awa domain (present-day Tokushima), who kept a residence in Edo at Kyōbashi Hatchōbori, matching the 1844 address exactly. Domain records show the Awa lord in Edo from the sixth day of the fourth month of 1793, then absent from the twenty-first day of the fourth month of 1794 until the second day of the fourth month of 1796, travelling to his domain under the shogunate's sankin-kōtai (alternate-attendance) system. If Jūrōbei did not have to travel with him, he would have been free in Edo for almost exactly the window in which Sharaku was active. A Noh programme from 1816 lists a performer of the same name, confirming that a Noh actor called Saitō Jūrōbei genuinely existed, though that alone does not prove he was Sharaku. Other candidates have been proposed since, none with comparable supporting detail. Most standard references now treat the identification as the leading theory rather than as settled fact. If Saitō Jūrōbei really was Sharaku, he went back to performing Noh and never designed another print. Kabuki's most unsparing portraitist worked for ten months, then chose, or was made, to disappear. Masters of Ukiyo-e What Is Ukiyo-e? Hokusai Hiroshige Utamaro Sharaku Kuniyoshi Yoshitoshi Ohara Koson
Agnes Martin Stopped Painting for Seven Years and Drove to Taos
In summer 1967, Agnes Martin walked out of her New York studio at the bottom of Manhattan, gave away most of her materials, sold or destroyed the canvases she had not been able to sell, bought a pickup truck and a small camping trailer, and drove west. She was fifty-five years old. She did not paint again for seven years. She had been working in New York since 1957 and had been part of the loose group of artists living at Coenties Slip, a small block of derelict buildings near the South Street Seaport. Her neighbours had included Robert Indiana, Ellsworth Kelly, Jack Youngerman, and Lenore Tawney. By 1965 her paintings were being shown at the Robert Elkon Gallery and acquired by major collectors. She had a New York reputation. She was leaving it deliberately. Saskatchewan Martin was born in 1912 in Maklin, a small farming town in the Canadian province of Saskatchewan. Her father died when she was two. Her mother, a strict Presbyterian, raised her and her three brothers on a small farm during the prairie depression of the 1920s and 1930s. The childhood was poor, isolated, and unhappy. Martin would later describe her mother's emotional coldness as one of the formative facts of her life. She left home as soon as she could. She trained as a teacher in Vancouver in the early 1930s, taught in rural Canadian schools, and moved to the United States in 1932 to take care of a pregnant sister. She became an American citizen in 1950. She studied painting at Columbia Teachers College in New York from 1941 to 1942 and again from 1951 to 1952, after which she moved to New Mexico for the first time and taught painting at the University of New Mexico. New York The first significant period in her painting life began in 1957, when Betty Parsons, the New York dealer, visited her in New Mexico and invited her to come to New York. Parsons, who had been Pollock's and Rothko's first dealer in the 1940s, recognised something in Martin's early gridded work and offered her a New York show. Martin moved to New York that year. She took the small loft at Coenties Slip and began producing the work that would make her reputation: large square canvases (most often 1.83 metres on a side) covered with delicately ruled grids of horizontal and vertical lines, drawn in pencil over a pale ground of oil or acrylic wash. The grids were calculated but slightly imperfect. The hand of the maker was always visible in the slight wavering of the lines and the small variations in pressure. The paintings were difficult to photograph and almost impossible to reproduce. They had to be seen in person. At close range they revealed the labour of the pencil work; at a distance of a few metres they began to fill the visual field with a kind of even, vibrating light. The paintings were neither minimalist (the hand was too visible) nor abstract expressionist (the gesture was too restrained). They occupied a position of their own. The Departure The departure from New York in 1967 was triggered by several events in close succession. Ad Reinhardt, the painter Martin most admired among her contemporaries, died in August 1967. The Coenties Slip buildings, where she had lived for ten years, were being demolished for new construction. Martin had also suffered (and would continue to suffer) from auditory hallucinations and what she described as catatonic episodes, which had brought her to McLean Hospital outside Boston for several extended stays in the early 1960s. Her doctors had diagnosed her with paranoid schizophrenia. The combination of personal grief, the loss of the studio neighbourhood, and the strain of New York life made her decide to leave. She had been considering it for some time. The summer of 1967 was when she acted. The Seven Years For the next seven years, Martin lived in various places in New Mexico. She camped in the truck for the first year, then bought a small piece of land near the Cuba area in 1968 and built an adobe house and studio with her own hands. She did not paint. She wrote: small notebooks of meditations and prose poems, several of which would later be published. She read Daoist and Zen Buddhist texts. She listened to classical music. She lived almost entirely alone. The seven-year break has been the subject of much speculation. Martin herself, asked about it later in life, was usually opaque. She said the break had been about "wanting to think." She gave no other detailed explanation. The break should probably be understood as a deliberate strategic withdrawal: she had been working at a certain level in New York and wanted to come back to painting only when she had something different to say. A second installation view of works by Agnes Martin, San Francisco Museum of Modern Art. The Return Martin began painting again in 1974, in her adobe studio outside Cuba. She was sixty-two. The new work was related to the pre-1967 grids but was different in important ways: the colour was no longer just a pale ground but had become the primary visual event of the paintings; the grids were less prominent, sometimes replaced by horizontal bands of pale wash; the formats had widened slightly. The first new paintings were shown at the Pace Gallery in New York in 1975. The reception was strong but cautious; many critics needed to be reminded who Martin was, after the seven-year absence. By the late 1970s, her reputation had been re-established. Taos In 1992, after eighteen years in the Cuba house, Martin moved to Taos, the small town in the Sangre de Cristo Mountains north of Santa Fe. She was eighty. She bought a small modern studio building near the centre of town and an adjacent house. She painted there nearly every day until her death. The Taos years produced her most concentrated late work. The paintings were now slightly smaller (often 1.5 metres square), with horizontal bands of pale colour washed across the canvas: pinks, blues, yellows, and greys. The grids were sometimes invisible, the colour bands carrying the painting on their own. Each painting was titled with a single word: Friendship, Innocent Love, Untroubled Mind, Lovely Life. The titles read as instructions to the viewer about what the painting was offering. Martin had become explicit, in her late writing and her late conversations, about her view that the paintings were vehicles for specific emotional states. "The paintings are not landscapes," she said in a 1989 interview. "They are about what life feels like when it is at its best." The Influence By the 1990s, Martin had become a foundational figure for a younger generation of painters who were working in pared-down, attentive, contemplative modes. Painters as different as Sean Scully, Susan Rothenberg, Brice Marden, and Vija Celmins all named her as an important precedent. The Tate Modern's 2015 retrospective in London, which travelled to the Solomon R. Guggenheim in New York and Los Angeles County Museum of Art, made her standing as one of the central American painters of the second half of the twentieth century explicit. The reception was not universal. Some critics found the late work too repetitive. Others found the late writing's invocation of "joy" and "innocent love" too earnest. The work was also accused of being apolitical in a period when much of the most-discussed American art was explicitly political. The defence of the work has usually rested on the structure rather than the content. Martin had developed, over forty years of painting after 1957, a way of using a small set of formal devices (grid, band, pale colour, soft graphite) to produce paintings that operated on the viewer at a specific physiological frequency. The paintings work, when they work, because of how the eye reads them at close range. The End Martin died on 16 December 2004, aged ninety-two, in Taos. She had continued to paint until shortly before her death. The last canvases (now mostly in private collections) are smaller than the standard 1.5-metre square, and the bands of colour are even more reduced. The Harwood Museum in Taos has a permanent room of seven late Agnes Martin paintings. The Dia Beacon, in upstate New York, has a permanent installation of the major mid-period paintings. The Tate Modern and the Museum of Modern Art each hold significant individual works. The Agnes Martin Foundation, established in her will, manages her archive and the disposition of her remaining works. What the Departure Was For The seven-year break from painting between 1967 and 1974 has come to look, in retrospect, like the most important decision of Martin's career. She had been a successful New York painter; she could have continued and done well. By leaving, she let her work reset. By returning at sixty-two, she gave herself another thirty years of working at the highest level she ever reached. Martin painted for forty-seven years total. Almost all the work that matters most was made after she was sixty. The pre-1967 grids are foundational, but the post-1974 paintings (the Cuba paintings, the Taos paintings, the late horizontal bands) are what most viewers now think of when they think of an Agnes Martin painting. She had decided, in 1967, that what she had to say required her to stop talking first. The decision was unusual. It produced, in the end, the strongest second act in twentieth-century American painting.