The edit at the end of my entry on Walter Benjamin’s ‘The Work of Art…’ reminded me of how the work of Matisse changed the way I thought about the role of abstraction and figuration in art and architecture. I spent a great deal of energy becoming acquainted with and understanding the importance of painterly abstraction on architecture. Colin Rowe and Robert Slutzky’s essay ‘Transparency’ makes clear the fruitful relationship. We also know about the relationship and important of Le Corbusier’s painting on his architecture. We could add the work of the Russian constructivist painters, de Stijl and so on. This history had led me to consider that the most abstract means produced the most worthwhile results. Following this line I ended up studying and appreciating the early work of Peter Eisenman – the so-called ‘cardboard’ projects, in particular houses I-IV. But something was troubling about this work and perhaps it came from having read Benjamin and his distinction between how art and architecture differ, primarily in the way that they are ‘absorbed’. Eisenman’s work, as an example, started to seem more like a work of art or pedagogical exercise rather than a proper work of architecture . I went back to the history of modern painting to examine the problem. Cubism, though now accepted, was originally seen as degenerate, not painting, and such. The reaction was the same for abstract expressionism, minimalism, colour-field painting, to name a few key abstract movements. In each case what the artists were doing was questioning the status, definition, and meaning of painting. This was done to such an extent and at such a level that it was natural that the general public would not, at first, understand the premise. In the process we have then some great abstract painters – Picasso, Gris, Mondrian, Malevich, Pollock, Johns, Albers, and etc.
Matisse, though considered one of the great painters of the 20th century, was never in the same group of the great innovators. The reason is that Matisse’s work was always more figural than that other modern painters. His was a more popular art, pretty; still proper painting with recognisable scenes – though, here and there Matisse would paint a canvas that shocked the establishment. These shocks were always of a different order than say that of Picasso. ‘The Red Studio’ of 1911 is a flood of red space, which almost does away with gravity, floors, walls and ceilings. Another painting was shocking for painting figures with open mouths (‘La Musique’, 1910). ‘The Dance’ also from 1910 was not initially well received. Still, everything is recognisable – there are figures, foregrounds, backgrounds and middle-grounds. Traditional space still abounds.
One painting, ‘The Blue Window’ 1912, on permanent display at the Museum of Modern Art stopped me in my tracks. We can recognize a lamp, a vase, a flower, a sculpture sitting on a desk…but wait, why do I think this is a desk. Could it not be the lower part of the window? Well, that would mean that the object are floating in space, and Matisse wouldn’t do that. Except he does – Where is the vase on the far left sitting? It’s roughly in the same line as the other objects. It could be sitting on the floor, unless we are actually looking at an elevation in which case we can see the floor and the objects are, once again, floating. And what is that golden oval object floating in the ‘sky, beyond the window’? It’s a cloud, obviously, but then why is it in such a saturated colour that makes it sit in the same space as the odd yellow sculpture and yellow ‘ashtray’? And just above the ‘ashtray’ is a lamp that just happens to have a tree growing from it, so it’s a vase? And is the black line emanating from the centre of the yellow-gold sculpture a window mullion or anther tree? Is the tree really a tree – if Matisse is deliberately playing with spatial readings, than could that not be a large houseplant, which is in fact indoors? I could go on, and I did, visit after visit, for a couple of years. I found the spatial play more and more shocking, exhilarating, inventive, and ultimately extremely abstract precisely because he was doing it through figuration.
I began to believe that Matisse was doing something incredibly difficult and perhaps more significant, at least for architects, than Picasso and the others. And I think he knew it. Did Matisse reject or not believe in pure abstraction? Was he incapable of doing it? His ‘French Window at Collioure’ of 1914 (the same year Malevich painted his black square) is instructive. It comes as close to a purely abstract, i.e., non-figural, painting as he ever did. He got there gradually, so he knew what he was doing, but then he backed off. I think it wasn’t because it became too difficult, but because it may have, for him, become too easy. The manipulation of space in what he continued to paint is extraordinary.
Now, what about the link to Benjamin? Although I follow Benjamin’s formulation quite literally – that you absorb artworks in a state of concentration, not as background music – I think that there are different levels of concentration. Abstraction was important because it makes you stop and work in front of a canvas, at least, as compared to what came before it, in which you simply ‘took in the scene’ . Matisse’s work oscillates between the abstract and the figural, between perspectival and abstract space, between the real and the imaginary. His is an art that is pushing boundaries as much as anyone else’s but doing it while maintaining a conversation with the public. This is not at all necessary for a painter or any artist; it was simply Matisse’s way. But it seems, to me, an appropriate approach for architecture. The discourses around drawings, representations, processes and the buildings themselves are too dependent on what they mean for architects. The interest is only in how they push boundaries and critique norms. These are useful endeavours, but have to be understood in relation within a broader understanding and definition of architecture. Could there be an architecture (or for starters, a discourse around architecture) that pushed the boundaries as hard as Matisse did while still managing to engage with issues that the general public still recognises? Would this not perhaps make for an interesting definition of how architecture works? Think, for example, of how much classicism was smuggled into Le Corbusier’s and Mies van der Rohe’s work or of how Aalto held on to certain phenomenological expectations within is modernist idiom. Who could we nominate today as an ‘architectural Matisse’? 
 The word ‘proper’ here is intentional and meant to piss off the hardcore post-structuralist reader.
 This is a very crude formulation. Any painter worth their salt, even the classical painters of the 15th, 16th and 17th centuries, played with the way pictorial space was presented and rarely was what you saw so straightforward. However, museum and gallery visitors reacted negatively to the early abstract paintings because they could not recognise a ‘scene’.
 The end of this entry leads quite naturally into two other entries, not yet written. One is on Neave Brown, whose Alexandra Road may be the most ‘Matessian’ project of the second half of the 20th century. The second architect brought to mind is Willem Dudok who skirted with de Stijl but rendered it with reference to traditional Dutch brick construction.