A figure circulates through museum outreach texts with the authority of the self-evident: the average visitor looks at the Mona Lisa for fifteen seconds. It is repeated in articles, talks, and guides to suggest that we no longer know how to look, that contemplation is dead, that we pass before works of art as if before shop windows. The figure is convenient because it confirms what we already believed. It does not appear in any study. I have searched for it to the primary source and it does not exist: neither the Louvre publishes it, nor is there any research that supports it. It is an orphaned datum that everyone cites and no one signs.
In 2001, Jeffrey and Lisa Smith stood in the Metropolitan Museum of Art in New York and timed one hundred and fifty visitors before six masterpieces. Result: twenty-seven seconds on average, seventeen as the median. Fifteen years later, the couple repeated the experiment at the Art Institute of Chicago, already in the full era of the camera phone. The average barely moved: twenty-eight and a half seconds. Subsequent studies with eye-tracking, measuring the gaze instant by instant, yield figures of the same order, close to thirty seconds per work.
Half a minute. That is, with data, the real time of the encounter with a painting. Not fifteen seconds of cultural agony, but neither the reverential and prolonged contemplation that we imagine when we idealise a past without screens. First uncomfortable finding: brevity was not brought by the smartphone. In 2001 we were already looking for half a minute, when the phone had no camera and no one took photos with the works. The brief gaze was there before. What the phone changed was the composition of that time: now it includes posing, capturing, checking, leaving evidence of having been there. We look for the same number of seconds, but a portion is taken by the gesture of registering that we are looking.
What has truly changed is not how much we look. It is that the museum has learned to measure it. This deserves attention, more so than the umpteenth lamentation regarding lost attention.
Today, time spent before a work of art is a data point. Studies call it dwell time, and they record it with eye-tracking. Bluetooth sensors have tracked where visitors circulate in the Louvre, how long they pause at each point, and how a crowded room expels people towards others. There are heat maps of the gaze, paths, densities, and fatigue curves. Recent work demonstrates that we look longer at the works we like, and from there it deduces that dwell time could serve to automatically profile the visitor, to know what they like without asking them. The gaze, which was the most private gesture possible before a work of art, is becoming a measurable signal, an exploitable datum, an input for recommendation.
Ese es el desplazamiento que importa. La pregunta ha pasado de «¿cuánto tiempo hay que mirar una obra?» a «¿cuánto tiempo mide el sistema que la has mirado?». No son la misma pregunta. La primera es interior, y no tiene respuesta general: depende de la obra, del día, de quién mira. La segunda es una métrica, y precisamente por serlo deja fuera lo único que hacía valioso el tiempo ante la obra.
The aesthetic problem is not speed. One can stand for thirty seconds before a painting and receive a real jolt, something that reorganises how it is seen. One can stand for ten minutes and not move beyond what one already knew how to say about it. Chronological time guarantees nothing. A dwell time datum records that someone paused, but it cannot record whether something happened during that pause or nothing happened at all. It measures the shadow of the encounter, not the encounter.
Defending the interval is not defending that the visitor should stand mute before a work of art until a revelation arrives. Studies on wall labels show something: when a text accompanying the work is added, the total time increases from thirty to forty-five seconds, and the gaze returns to the image after reading. Good mediation does not steal time from the work: it purchases it. The danger is not explaining, it is over-explaining, closing the work with a key so strong that it is no longer necessary to look at it again. A poor label replaces the painting with its answer. A good label returns the visitor to the difficulty of looking.
The slow gaze that some claim is not looking for more minutes to understand sooner. It is almost the opposite: looking for long enough that the work ceases to coincide with what we already knew how to say about it. So that a distance opens between what we recognise at first glance and what is actually there. That interval —the gap between the first glance and what appears if one does not close it off— is exactly what no metric can capture, because it is not a quantity of seconds. It is a suspension of closure.
This is why the accounting of the gaze, however refined it may be, always arrives late to what it sought to measure. It can tell us that someone was there for twenty-seven seconds. It cannot tell us whether, in those twenty-seven seconds, the world reorganised itself slightly or remained exactly the same. And that difference, which cannot fit into any sensor, is the entire difference.
About the open conversation
This text intersects with a line of inquiry I have been pursuing: the defence of the interval, of slow time, and of that which resists measurement in aesthetic experience. Attention before a work, when it becomes data, enters the same logic of capture that I have addressed when discussing infrastructure and recognition. If anyone wishes to intervene from the fields of visitor studies, the psychology of perception, museology, or the practice of cultural mediation, this notebook remains open.
Sources
Smith, Jeffrey K. and Lisa F. Smith. “Spending Time on Art”. Empirical Studies of the Arts 19(2), 2001, pp. 229-236. DOI: 10.2190/5MQM-59JH-X21R-JN5J.
Smith, Lisa F., Jeffrey K. Smith and Pablo P. L. Tinio. “Time Spent Viewing Art and Reading Labels”. Psychology of Aesthetics, Creativity, and the Arts 11(1), 2017, pp. 77-85. DOI: 10.1037/aca0000049.
Krukar, Jakub and Ruth Conroy Dalton. “How the Visitors’ Cognitive Engagement Is Driven (but Not Dictated) by the Visibility and Co-visibility of Art Exhibits”. Frontiers in Psychology 11:350, 2020. DOI: 10.3389/fpsyg.2020.00350.
Reitstätter, Luise, Karolin Galter and Flora Bakondi. “Looking to Read: How Visitors Use Exhibit Labels in the Art Museum”. Visitor Studies 25(2), 2022, pp. 127-150. DOI: 10.1080/10645578.2021.2018251.
Tishman, Shari. Slow Looking: The Art and Practice of Learning Through Observation. Routledge, 2018 (print ed.) / 2017 (ebook). DOI: 10.4324/9781315283814.
Yoshimura, Yuji, Anne Krebs and Carlo Ratti. “Noninvasive Bluetooth Monitoring of Visitors’ Length of Stay at the Louvre”. IEEE Pervasive Computing 16(2), 2017, pp. 26-34. DOI: 10.1109/MPRV.2017.33.
Musée du Louvre. “FAQ — While visiting”. louvre.fr.