Paul Klee drew roughly fifty angels in the final year of his life. Not during his decade teaching at the Bauhaus in Weimar and Dessau — after. The angel series belongs to 1939 and the first months of 1940, when Klee was already ill with the scleroderma that would kill him, working on paper in Bern because the Nazis had closed every door in Germany that mattered to him. This piece reads the angels as a Bauhaus master's last argument about form: what remains when colour, workshop, and ideology are stripped out, and only the line is left.
Methodology
We treated the angel series the way a desk treats a file: sort it by date, separate the canonical outlier from the body of work, and only then describe what the shapes are doing. The sources are the broadly established record of Klee's biography and the public catalogue of his late work held by the Zentrum Paul Klee in Bern. We counted only drawings and paintings Klee himself titled with the word "Engel" or an obvious synonym; we did not sweep in every winged figure the literature has ever called an angel, because that invites the thing we are trying to avoid — projecting Walter Benjamin's famous reading of a single 1920 monoprint backwards and forwards across a career.
Three limitations are worth naming before we start. First, exact counts of the 1939–1940 series vary slightly depending on which cataloguer you follow; we use the broadly cited figure of about fifty. Second, Klee's dating is his own — he inscribed year and sequence number on his sheets, and we trust those. Third, we are a design-history desk, not an iconographer: we read the angels as form before we read them as theology.
Finding #1: The Angels Are a 1939–1940 Project, Not a Bauhaus-Era One
The most common mistake in writing about Klee's angels is treating them as a career-long theme. They are not. The great majority of the series belongs to the last fifteen months of his life, between early 1939 and his death in June 1940. Klee had left the Bauhaus in 1931, taught briefly at the Düsseldorf Academy, been dismissed by the Nazis in 1933, and returned to Bern. By 1935 he had been diagnosed with scleroderma, a slow systemic disease that gradually stiffens the connective tissue and, in his case, the hands.
So the chronology matters. His Bauhaus years at Weimar and Dessau, roughly 1921 to 1931, were years of colour theory, the preliminary course rotation, the pedagogical sketchbooks, and the workshop. The angels do not belong to that decade. They belong to a sick man in a small Bern apartment who was producing drawings at an astonishing late-career rate — over 1,200 works in 1939 alone, by the standard count — because his body was telling him there was no time.
This reframes everything about how the angels look. They are not a mature artist's considered summation of a theme he has worked at for twenty years. They are a final sprint. The thickened, almost calligraphic line that readers find touching is partly a consequence of hands that could no longer draw a fine line reliably, and partly a deliberate reduction toward the essential. Both readings are true; neither alone is enough.
Finding #2: Angelus Novus (1920) Is the Outlier Everyone Mistakes for the Template
The one angel everybody knows — *Angelus Novus*, the small oil-transfer-and-watercolour on paper from 1920 — is a stylistic outlier that happens to have become the most-cited piece in the whole corpus, because Walter Benjamin owned it and wrote about it in the ninth thesis of *On the Concept of History* in 1940. Benjamin's reading of the figure as the "angel of history" being blown backwards into the future by the storm of progress has done more to shape how the piece is seen than anything else, including the piece itself.
The problem is that *Angelus Novus* looks almost nothing like the 1939–1940 angels. It has colour — ochres, brown, a hint of gold. It has a frontal, heraldic composition. The wings are rendered as feather-like horizontal bands. The face is schematic but modeled, not reduced to line. It belongs to Klee's early experiments with the oil-transfer technique he was developing just after joining the Bauhaus in Weimar.
Treating *Angelus Novus* as the template for the series is roughly like treating a single 1920 watercolour by Kandinsky as the template for his 1930s Bauhaus work. The gap is the whole decade of teaching in between. The late angels are not junior *Angelus Novus* figures. They are a different project, done in different conditions, and they answer different questions. Benjamin's reading is essential cultural history and should be read on its own terms, but it is a reading of one sheet, not of the series.
Finding #3: The Late Angels Are Line, Not Colour — Geometry Stripped Bare
Open any folder of the 1939–1940 angels and the first thing you notice is what has gone. The colour theory Klee spent a decade teaching at the Bauhaus — the hue wheels in the pedagogical sketchbook, the exercises in how a warm patch pushes forward against a cool ground — has been subtracted out. The late angels are, overwhelmingly, drawings. Pencil, coloured pencil, chalk, oil transfer, sometimes a thin wash. The sheet is often a found scrap: newsprint, wrapping paper, torn edges.
What remains is line. And the line is doing a specific job. The bodies are built from a small vocabulary of modernist primitives — a circle for a head, a triangle or wedge for a torso, two opposed arcs for wings, a single stroke for an arm. This is the same alphabet Kandinsky set out in his Bauhaus questionnaire of 1923 and the same one Klee drilled his own students on in the preliminary course rotation: triangle, square, circle; point, line, plane.
The angels, then, are the preliminary course exercise carried into a final year. They are not expressionist cries. They are formal compositions in which a figure is assembled from the fewest possible shapes that will still read as an angel. Some of the titles are tellingly pedagogical — *Angel, Still Ugly*; *Angel, Still Groping*; *Forgetful Angel*; *Angel in the Kindergarten* — as if Klee were marking student exercises, except the student is the angel and the exercise is becoming.
Finding #4: Klee's Bauhaus Teaching Survives in the Angels as Reduction
The Bauhaus taught reduction. The preliminary course under Johannes Itten, and then under Josef Albers from 1923, asked students to begin by stripping materials and forms down to their simplest readable state before adding anything back. Albers would later carry this to its most famous conclusion with the squares, but the principle was in the school from its 1919 founding in Weimar and stayed through Dessau and the short Berlin year before closure in 1933.
The late angels are the strongest surviving evidence of what that teaching meant to Klee in his own practice. They are not Bauhaus posters. They are not grids or primary-colour compositions in the Foundation sense. They are figures reduced until one more subtraction would make them illegible, and held one step short of that line. This is the discipline the preliminary course was trying to transmit, and in the angels Klee demonstrates it on himself, not on students.
There is also what is absent. The angels contain no workshop, no industrial material, no architectural ambition, no social programme. The Bauhaus's utopian frame — design in service of a new mass life — is gone. What is left is the formal core: a geometric figure, held to the minimum, inked onto a scrap of paper by someone whose hand was failing. If you want to know which part of the Bauhaus Klee actually thought was load-bearing, the angels are the answer. Not the ideology. The reduction.
| Attribute | Angelus Novus (1920) | Late Angels (1939–1940) | Bauhaus Foundation Exercises |
|---|---|---|---|
| Period | Pre-Bauhaus Weimar (Klee joins 1921) | Post-Bauhaus Bern years | Weimar and Dessau, 1919–1931 |
| Medium | Oil transfer and watercolour on paper | Pencil, coloured pencil, chalk on scrap | Gouache, cut paper, workshop materials |
| Colour role | Load-bearing; ochre and gold tonality | Minimal or absent; line carries the work | Primary hues used as formal demonstration |
| Figure logic | Heraldic, frontal, modeled face | Assembled from circle, triangle, arc, line | Pure geometric primitives, no figure |
| Count in corpus | 1 (unique monoprint) | ~50 titled Engel works | Hundreds of student and teacher sheets |
What This Does NOT Prove
This reading argues that the late angels are a formal project continuous with Klee's Bauhaus teaching, and that *Angelus Novus* is a stylistic outlier that has been retroactively treated as the series' anchor because of Benjamin. It does not prove that Klee himself would have described the angels this way; he left no manifesto for the series, and we should be careful about speaking for the dead. It also does not address the theological question of what Klee believed about angels, which is a separate argument and one we are not qualified to settle on a design desk.
We have also not covered the market history of the angels — who bought them, when prices moved, how the Zentrum Paul Klee assembled its holdings — because that is auction-room business and Bauhaus Walls is not an auction-room publication. Nor have we taken a position on the broader question of how Klee's late work fits into the history of modernism's response to fascism. Those are serious arguments. They belong in other pieces.
The Takeaway
Klee's angels are the Bauhaus preliminary course continued by a dying hand: forms reduced to the fewest marks that still carry meaning. Read them as line, not as theology.
FAQ
How many angels did Paul Klee actually draw?
The broadly cited figure is around fifty works that Klee himself titled with "Engel" or a close equivalent, almost all of them from 1939 and the first months of 1940. Counts vary by a handful depending on which catalogue you follow and how strictly you define the series. If you include winged figures Klee did not title as angels, the number grows considerably, but most scholarly treatments stay with the titled core because that was Klee's own labelling.
Is Angelus Novus part of the angel series?
Not really, in the sense the series is usually meant. *Angelus Novus* is a single 1920 oil-transfer-and-watercolour sheet made just before Klee joined the Bauhaus in Weimar, and it is stylistically unlike the 1939–1940 drawings. It became culturally central because Walter Benjamin owned it and wrote about it, but reading it as the template for the later series collapses twenty years of changing practice into one image. Treat it as a famous outlier, not as the key.
Were the angels made at the Bauhaus?
No. Klee taught at the Bauhaus in Weimar and Dessau from 1921 to 1931. The angel series belongs to his last Bern years, roughly 1939–1940, after he had been dismissed by the Nazi regime and returned to Switzerland. What the angels inherit from the Bauhaus is not location or workshop but the preliminary-course principle of reduction — building a figure from the smallest possible vocabulary of geometric marks that still reads.
Why do the late angels look so stripped down?
Two reasons together. Klee had scleroderma, a disease that progressively stiffens connective tissue, and by 1939 his hands could no longer reliably manage a fine, controlled line. At the same time he was producing work at a startling pace — over a thousand sheets in 1939 alone — on scrap paper, in simple media. The reduction is partly bodily necessity and partly a deliberate late-career choice to work in the same language of circle, triangle and arc he had taught at the Bauhaus.
What is the connection between Klee and Kandinsky's shape theory?
Klee and Wassily Kandinsky were both Bauhaus masters and close colleagues, and they taught overlapping versions of a basic-forms vocabulary: triangle, square and circle as the three irreducible shapes. Kandinsky formalised this in his 1923 questionnaire asking which primary colour belonged to which primary shape. Klee's own pedagogical sketchbooks run on parallel lines. The late angels use that same small vocabulary, assembled figuratively rather than abstractly — the discipline is identical, the subject differs.
Where can the angels be seen today?
The largest holding is at the Zentrum Paul Klee in Bern, Switzerland, which is built around the collection the Klee family preserved and donated. Individual sheets are held by museums and private collections around the world, and the series travels regularly in exhibitions. *Angelus Novus* itself is held by the Israel Museum in Jerusalem, having passed from Benjamin's estate through Gershom Scholem to its current home.
Did Klee leave any writing about the angels?
Not in the form of a statement about the series. Klee wrote extensively earlier in his career — the Bauhaus-era pedagogical sketchbooks, the diaries, the lectures on form and colour — but the 1939–1940 period is sparse in writing because he was ill and because most of his working energy went into the drawings themselves. The titles he inscribed on the sheets are our best primary source for how he was thinking about each figure, and they repay close reading.
How do the angels fit into Bauhaus Walls' own work?
Our generative compositions — *Foundation*, *Archway*, *Dessau Morning*, *Ascending Forms* — belong to the abstract-geometric side of the Bauhaus tradition, the preliminary-course vocabulary of circle, arc and bar in primary colour. Klee's late angels sit on the figurative side of the same discipline: the same small vocabulary of marks, used to assemble a figure rather than a pattern. Both are the reduction principle in action. You can see our compositions at browse the geometric prints.
From the collection
New compositions and 10% off your first print.
One email now with your code. No noise after.