When Abstract Art Remembers

People cry at Rothko. Not metaphorically — actually cry, on the bench, longer than they planned to.

And there is nothing in the painting. No library, no street, no face, no thing at all. Rothko refused to paint anything in particular, on principle, for his whole career, and what he got in exchange was the sublime. Fair trade. Nobody should want it undone.

So let’s kill the obvious version of this essay right now, because it’s wrong. Abstraction did not need to keep one foot in the visible world to move anybody. Some of the most devastating abstract paintings ever made are the ones with the least in them. Newman’s zips aren’t pictures. Franz Marc’s Fighting Forms is a red-and-black storm that depicts no war whatsoever — and he painted it in 1914 and died at Verdun two years later, and good luck looking at it now without seeing what was coming for him.

Except — look at what just happened there. The dread came from outside the picture. You had to know about Verdun. Hand that canvas to somebody who’s never heard of Franz Marc and it’s still a magnificent storm. It’s just not about a war anymore.

That’s not a one-off. It’s a genre. Franz Kline named a painting Chief after a train from his childhood, and there’s no train in it. Motherwell spent forty years on his Spanish Elegies, and the Spanish Civil War sits in those paintings the way a caption sits under a photograph — authoritative, and completely outside the frame.

All of them reach for something we hold in common. Not one of them can be seen to. The painting isn’t doing the remembering. The label is.

Which is the whole argument, and it’s smaller than you’d think. Forget abstraction versus representation — that fight is over and everyone lost. Forget cold versus emotional; Rothko settles that in about four seconds. The line that actually matters is between paintings you get told about and paintings you can see. Any title can smuggle in the shared world. What a bit of legibility buys — a bit, not a lot — is that the smuggling happens through your eye, at the speed of looking, before you’ve read a single word on the wall.

Worth saying that two of the five paintings below have no title at all. Nobody knows what their painters called them. They still hand you something you recognise, which is about as clean a demonstration as this argument is ever going to get.

There’s a narrow band where this works. Enough of the world left to recognise. Not so much that you’re just looking at a picture of a thing. Five paintings, fifty-five years apart at the ends, all sitting in that band.

I can’t show you any of them — every one is still in copyright, which is its own small comedy. The links go to the works. Open them. I know you probably won’t. Do it anyway; this is the part of the essay that actually works.

Maria Helena Vieira da Silva, Bibliothèque, 1949 — Centre Pompidou

Try to figure out how far back the wall is. You can’t. The shelves recede, and then a plane you’d read as distance turns out to be sitting in front of something you’d read as closer, and the room folds up like a badly refolded map. Which is the point, and which is exactly right. Anyone who has stood in a big old library knows this specific vertigo — aisles running back further than the building can possibly allow, the collection obviously exceeding the room. She painted the feeling and threw the architecture away. A photograph can’t do this. A photograph has to make the geometry work.

She painted it the year she got back to Paris from Brazil, where she’d sat out the war. And she knew exactly what she was doing: a friend asked her why on earth she was still doing perspective, since that wasn’t done anymore, and she said she knew it wasn’t done and she had to do it anyway.

Norman Lewis, Processional, 1965

No link for this one. It’s in a private collection and there’s no museum page to send you to, which is its own quiet comment on which paintings the world decides to keep track of.

Count them. Go on. You’ll start — those pale marks against the black are unmistakably people — and you’ll give up, and that’s the painting. A march is a thing nobody can count. Not from inside it, not from the sidewalk. That’s what makes it a march instead of a crowd. Lewis painted it the year of Selma, and he found the one true fact about a procession that a news photograph structurally cannot give you: the number is the entire point, and the number is unavailable.

This is the best painting of the five and the one you’re least likely to have heard of. Those two facts are related and neither is an accident.

There’s another Lewis from the same year, and nobody knows what he called it. The Amon Carter catalogues it as Title Unknown (March on Washington) — the parenthesis is a curator’s, arrived at by standing in front of it and working out what it was. Not a label importing a meaning. A label reporting one.

Cy Twombly, Untitled, 1968 — MoMA

House paint and crayon. Not oil, not anything you’d buy in an art shop — the stuff you’d paint a hallway with, and a crayon, which is what a child gets handed. The schoolroom is in the materials before it’s anywhere in the marks.

Then the marks. Loops, slanted, endless, the kind you were made to fill a page with while staring out the window. Your hand gets this before your eye does. What comes back isn’t a classroom. It’s the boredom. The chalk dust. Being seven and having somebody else decide what your hand would do until lunch. Nothing in it depicts a school and it doesn’t need to.

Julie Mehretu, Empirical Construction, Istanbul, 2003 — MoMA

Ten feet by fifteen. You don’t look at this one so much as stand in front of it and lose an argument with it.

She calls it a portrait of a city. MoMA is careful to add that the city is never actually pictured as a place — and that’s right, and it is not the objection to my argument that it sounds like. There’s no view of Istanbul here. What there is: a star and crescent, flags, Islamic decorative motifs, bright rectangles going off like signage. Underneath, real architectural drawings, traced and projected and layered over and half-consumed. The Hagia Sophia sits dead in the middle, quiet, while the rest of the canvas loses its mind around it.

You’re not looking at Istanbul. You’re looking at what a city does to the drawings people made of it. Mehretu has said that history gets made one layer on top of another, erasing itself, consuming itself, making something else out of the same material — which is also a fair description of how she builds the picture.

Julie Mehretu, Stadia II, 2004 — Carnegie Museum of Art · Smarthistory

You hear this one before you see it. The tiered arcs are a stadium bowl, no mistaking it, and spilling off them is a confetti of pennants and flags and heraldic junk, and the roar arrives immediately — that specific noise nobody makes individually and everybody inside is producing.

She isn’t sentimental about it. She was thinking about stadiums and amphitheatres as perfectly constructed spaces, built to organise a crowd democratically and carrying, underneath that, every possibility of chaos and violence. She was thinking about it in 2004, with the Olympics going back to Athens and the War on Terror running. The same painting holds the pageantry of nations and the banners of teams and the machinery of mass feeling, and it flatly refuses to tell you where sport stops and something worse starts. It only gets to ask that because you recognised the stadium first.

Five paintings, five decades, one thing in common. Not a movement — Vieira da Silva and Lewis worked from what they saw, Mehretu builds from geometry, Twombly was following a rule. Not a period; the first one is four years after the war and the last is four years after the millennium. What they share is a single setting on a single dial. Legible enough to walk into. Abstract enough that no camera could have taken you there.

That’s the case for Abstract Impressionism, and it is not the case that abstraction should be prettier, or gentler, or done in pastels. It’s that there’s a whole register of feeling — the ache of a place you’ve been, or somewhere near enough to one you’ve been, the recognition that lands in your body before it reaches your mouth — that abstraction can only have if it leaves a door open. Rothko shut the door and got the sublime, and that is a magnificent trade. The figure went out through the same door, and that trade has been examined a good deal less.

But the sublime is what you get when you meet something too big to be a memory.

The other thing — the smaller one, the human one — needs a library in it. Or a march. Or a blackboard. Just enough of one to see.