Henry Moore: Sculptural Process and Public Identity

ISBN 978-1-84976-391-2

Henry Moore OM, CH King and Queen 1952-3, cast 1957

Moore suggested that the idea for King and Queen came from ancient Egyptian statuary and from fairytales he read to his daughter. Others have suggested that it might have been related to a photograph of him and his wife Irina or that it could have been inspired by the coronation of Queen Elizabeth II that year. Although some critics disliked the sculpture’s unusual combination of naturalism and fantasy, King and Queen became one of Moore’s most famous works.
Henry Moore OM, CH 1898–1986
King and Queen
1952–3, cast 1957
Bronze
1635 x 1385 x 845 mm
Purchased from the artist by the Friends of the Tate Gallery with funds provided by Associated Rediffusion Ltd 1959
Number 5 in an edition of 5 plus 2 artist’s copies
T00228

Entry

King and Queen comprises a male and a female figure sitting side by side on a bench. The male figure is slightly broader and taller than the female and when viewed from the front it is evident that they are both sitting at an angle, facing left of centre. The female figure sits further forward on the bench in a more upright position, while the male figure appears to be leaning back slightly, as though more relaxed.
The head of the male is delineated by an angular lower jaw that narrows to a sharp point at the chin and a thin blade of bronze occupying the position of the nose that connects the chin to the top of the head in a straight line (fig.1). In the place of cheeks are hollowed cavities that accentuate the sharpness of the central ridge, through which a circular hole has been drilled to denote expressionless eyes. The top of the head is flat but slopes steeply upwards from the front to the back. At the front, above the central nasal ridge, is a semi-circular band that extends out of the mass of the head and arches up and over to the other side of the face, and may signify a quiff of hair or an ornamental headpiece (fig.2). From some angles it appears as though the upper rear edges of the head take the form of goat-like horns. The thinness of the face contrasts with the two rounded protrusions at the nape of the neck.
Fig.1
Detail of heads of King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
Fig.2
Detail of King's head, King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved

The male figure’s shoulders are symmetrical and follow the line of an arc; the collar-bone and chest repeat the curve of the shoulders, and no bodily features such as nipples, muscles or a navel are visible on the torso suggesting that the figure is clothed. The arms are thin tubular limbs; the left arm is held slightly in front of the body and is bent at the elbow so that the left hand, with individually modelled fingers, rests in the figure’s lap (fig.3). The right arm is positioned slightly behind the body, with a slight crook in the elbow. The forearm tapers to a narrow wrist and the palm of the right hand rests flat on the side leg of the bench (fig.4).
Fig.3
Detail of King's left hand, King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
Fig.4
Detail of King's right hand, King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved

Seen from the side and the rear it is evident that the body is very thin, and there is a recessed, concave space between the shoulder blades that extends down the spine to the waist, which is narrow and serves to emphasise the figure’s broad shoulders. The figure is curved at the hips, and the legs extend horizontally off the bench, bending at the knees. The lower half of a tunic binds the two legs together and ends three-quarters of the way down the figure’s calves. Three pleats or folds of fabric can be seen running down the front of the tunic in between the figure’s legs, which lead to naturalistically modelled human ankles and feet placed firmly on the flat bronze base.
Fig.5
Detail of Queen's head, King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
The female figure sits to the left of the male figure, with a gap between them. Her head combines a thin, fin-like face with rounder forms at the rear. The central facial ridge consists of two angles that originate from the forehead and chin respectively, and meet at a point in the middle that could be identified as a nose. A circular hole pierces the thin face just like that of the male’s and denotes the eyes. Eyelashes have been incised on the left side of the face. The female also sports a semi-circular band that extends out and over the top of her head and could be deemed to represent a diadem (fig.5). Three bulbous forms are positioned at the nape of the figure’s neck and resemble bunches of pinned up hair.
Fig.6
Detail of Queen's hands, King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
The female’s body is akin to the male’s: it is long, thin and ribbon like, curved at the hips and knees and sits firmly on the bench. Overall she is smaller than her male counterpart, but the most conspicuous differences between them are the small, widely spaced, domed breasts on the female figure’s chest and her slightly wider hips. Her thin, elongated arms lead to her lap, where her hands interlock delicately (fig.6).
Fig.7
Detail of the feet in King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
Like the male figure, the female figure is wearing a tunic. Her legs are undefined underneath the fabric but clearly extend over the edge of the bench and bend at the knees. The tunic ends two-thirds of the way down her calves where five pleats or folds can be counted. Each of the concave, vertical folds is separated by a convex curved ridge, which cumulatively create an undulating surface. The figure’s ankles appear from underneath the tunic and lead to feet planted firmly on the rectangular base (fig.7).
Fig.8
Detail of the surface markings on King and Queen 1952–3, cast 1957
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
Although the two figures do not touch each other, and each one has been rendered individually, a sense of familiarity has been achieved by the ways in which their bodies replicate each other. Both have a similar spinal curve and concave spaces in their backs, and when seen from the front, the curve of their shoulders and the flatness of their torsos are similar. The shape and bend of the female’s right arm mirrors that of the male’s left. The surfaces of the two figures are highly marked, although not rough (fig.8). When seen from the rear, long vertical scratches run down the length of both figures’ backs and serve to emphasise the elongated proportions of the bodies.
In the early 1950s Moore slowly began phasing out the practice of making preparatory drawings for each new sculptural project. Although he continued to make sketches on paper, these were records of general ideas and were rarely, if ever, translated into specific sculptures. This change in working practice was probably the result of a number of factors, but principally due to his turn to bronze casting. The inherently malleable qualities of wax, clay, and plaster from which a bronze could be cast meant that Moore was able to add as well as subtract forms as he worked, thus avoiding the necessity of having to plan the form of a sculpture on paper in advance, as is the case when carving in stone or wood. In addition to changing his materials and techniques, by the early 1950s Moore had amassed a large collection of graphic designs for sculptures that he had drawn in the 1940s, to which he was able to refer at a later date.
Henry Moore 'Seated Figures' c.1948
Fig.9
Henry Moore
Seated Figures c.1948
Tate T00228
© The Henry Moore Foundation. All Rights Reserved
Between 1947 and 1949 Moore made numerous drawings of seated figures and family groups. One such drawing, Seated Figures c.1948 (fig.9), presents an array of sketches each depicting single figures or groups of sitting figures in varying degrees of naturalism. In the centre of the page is an upright male and female couple sitting on a bench. The woman, sitting to the left of the man, is identified with breasts, and holds her hands in her lap. The figures are notable for their thin, tubular limbs and the way in which their feet are firmly placed on the ground. Although there are differences in the pose and rendering of the figures in King and Queen, they nonetheless share a resemblance to those presented in this sketch. While it is unlikely that a direct causal link can be made between this drawing and King and Queen, it nonetheless demonstrates that Moore had been thinking about the pose of a seated couple prior to making his sculpture.1 In the late 1960s Moore recalled:
The ‘King and Queen’ is rather strange. Like many of my sculptures, I can’t explain exactly how it evolved. Anything can start me off on a sculpture idea, and in this case it was playing with a small piece of modelling wax ... Whilst manipulating a piece of this wax, it began to look like a horned, Pan-like, bearded head. Then it grew a crown and I recognised it immediately as the head of a king. I continued and gave it a body.2
Henry Moore 'King and Queen' c.1952
Fig.10
Henry Moore
King and Queen c.1952
Whereabouts unknown
© The Henry Moore Foundation. All Rights Reserved
If Moore’s recollection was correct, and the horned or beak-like head developed while he was modeling in wax, then a drawing of King and Queen made in c.1952 (fig.10) depicting the King with these features must have been completed after he made the initial maquette. A description of the creation of King and Queen provided by Moore’s biographer Donald Hall supports Moore’s recollection and goes into more detail:
The King’s head was the first thing that happened. Moore was idly fingering a piece of wax, with no idea of making such an image. He pinched the wax between thumb and forefinger, and the result made him think of the god Pan – or of a king. In a few minutes he virtually completed the head. Later the same day he made the King’s body. He dropped a flat sheet of wax in a pan of water to soften it, and cut it with a knife, and bent and rolled it to its proper thin shape. The idea of the couple crystallized overnight ... The next day he made the bench and a Queen to sit on it, space between her and the King. He added a thin wire-like frame, squaring the couple on all sides and the maquette was finished.3
However, in 2006 the sculptor Anthony Caro, who worked as one of Moore’s studio assistants between 1951 and 1953, described slightly differently how the wax maquette was made:
I remember watching Henry making the King and Queen maquette. The way he worked in wax intrigued me. He got attracted to working in wax while I was there. He set up a situation whereby we all partook in parts of his process. Alan Ingham [another studio assistant] and I were helping him in the big studio, the old cow shed ... First we boiled up the wax and poured it out on to flat sheets, so he’d got wax roughly the thickness of a casting. Then he’d cut it out, he’d cut out the shape of the figure he wanted – in this case the bodies of the King and Queen – and he did it exactly like a tailor cutting cloth into shape. He worked on a very small scale, the whole figure about six or eight inches [15–20 cm] high. He’d get this little figure and bend the warm wax into a seated position, then he’d make a little wax seat for it and give it legs and arms made up of thin elements of wax which he’d model. He was good at modelling hands or feet on a tiny scale. When it came to the head he’d make the head out of small wax parts, weld them together and dip the whole head into hot water, so the shapes would melt a little and cohere and the parts of the head would run together.4
Although there are slight differences between these three accounts, taken together they nonetheless provide a vivid description of the ways in which Moore experimented with forms by modelling in wax.
Henry Moore
Fig.11
Henry Moore
Maquette for King and Queen 1952
Private collection
© The Henry Moore Foundation. All Rights Reserved
The original wax design for King and Queen has survived in the form of a bronze cast, made in an edition of ten (fig.11). According to Hall the first example in the edition ‘was cast from the wax, and then subsequent bronzes based on the first bronze’.5 He went on to explain that of the examples cast in bronze, Moore preferred the first because ‘it retains the sharp edge in the knife left in the wax’.6 From Hall’s description, it would seem that the first cast of Maquette for King and Queen was created using the direct lost-wax process. This would have been a risky procedure as the original wax model would have been destroyed in the casting process: the wax would have been encased in a refractory material and placed in a kiln to melt the wax out of the casing, thus creating a hollow interior into which molten bronze was poured. Once the bronze had cooled and hardened the outer shell was removed to reveal the sculpture. It is likely that a mould of this first bronze was made so that the remaining sculptures in the edition could be cast. Despite Caro’s assertion that Moore was ‘good at modelling hands or feet on a tiny scale’, the feet of both figures in the bronze maquette are simplified, paddle-like forms. Features of the maquette that were not repeated in the final bronze sculpture include the King’s clenched left fist and the thin, square frame enclosing the figures.
Henry Moore
Fig.12
Henry Moore
Fragment of King and Queen 1952–3
The Henry Moore Foundation
© The Henry Moore Foundation. All Rights Reserved
Photo: Lyndsey Morgan
Having made the wax version, and possibly before casting the bronze, Moore also made a copy of the sculpture in plaster. A fragment of the queen’s head, torso and arms is held in the collection of the Henry Moore Foundation (fig.12). In this work, the plaster of the arms has been built over a wire armature. Since Hall suggested that the bronze cast was made using the original wax, it is unlikely that this plaster was created for casting purposes. However, its later role as a template for the creation of an enlarged version can be identified by the presence of numbers and crosses marked in pencil on the surface of the plaster.
Henry Moore 'Maquette for King and Queen' 1952–3
Fig.13
Henry Moore
Maquette for King and Queen 1952–3
Henry Moore Foundation
© The Henry Moore Foundation. All Rights Reserved
Having made the sculpture on a small scale, Moore then decided to create a larger version. Instead of working direct from a small scale to full size, he made a plaster working model at an intermediary size. Maquette for King and Queen 1952–3 (fig.13) is a fragment of the working model and was probably scaled up from a now lost plaster model or was copied from the cast bronze maquette; as in the bronze, the King’s fist is clenched. The surface of the plaster has been annotated with small crosses and numbers denoting the various measurements that the enlarged full-size version should take. According to the curator Julie Summers, the enlargement process would have been undertaken by one of Moore’s studio assistants, and was ‘a scientific rather than artistic process’.7 The plaster working model of the Queen has not survived but Caro recalled enlarging it:
I enlarged the Queen in terracotta for Henry. When I had been at the Royal Academy of Art Siegfried Charoux and Arnold Machin had taught me the technique of working in terracotta. I said to Henry, ‘Would you like me to enlarge it to three-quarters size in terracotta?’ and he said, ‘Yes, good idea, go ahead’. I don’t know why we never did the King, as he was pleased with the Queen. It was slow working in terracotta: you have to wait for the clay to harden and dry to a leathery consistency before you can go on to the next stage. Again, he would do a bit with it, he would push it here and there in the thorax for example, but by and large he mostly worked on the hands, feet and head.8
Henry Moore 'Seated Figure' 1952–3
Fig.14
Henry Moore
Seated Figure 1952–3
© The Henry Moore Foundation. All Rights Reserved
As an original sculpture in its own right this terracotta enlargement became known as Seated Figure (fig.14) and was subsequently cast in bronze in an edition of five. This version of the Queen, like the earlier Maquette for King and Queen on which it is based, is different from the final larger scale version; Seated Figure does not wear a headpiece and the hands are mitten-like, unlike the slender, naturalistic fingers of the final sculpture. Summers noted that once the enlargement process was underway Moore was able to ‘make changes to proportions or other aspects of the work’.9 In 1968 Moore explained that ‘hands, after the face, are the most obvious part of the human body for expressing emotion. That’s why I concentrated on the hands in the King and Queen in order to add to the sculpture an extra interest and meaning. This was missing in the little maquette, where the hands are rather rudimentarily and cursorily done’.10
In a letter written to the American lawyer and philanthropist Allan D. Emil on 21 October 1966 Moore responded to a query about his sculpture Hand Relief No.1 1952:
You are quite right in thinking it is the study for the hands on a larger work, actually of the KING AND QUEEN, 1952/53.
When I was making the large version of the KING AND QUEEN, and getting to working on the hands in detail, I got a little stuck with them, and contrary to any previous practice, I felt I needed to have a model of a real hand. And so I asked my wife to come into the studio and hold her hand in the position I required – she posed for about a quarter of an hour and then said she couldn’t stay any longer as the lunch she was cooking needed her attention. I then asked Mary, my daughter who was six years old, to come and hold her hand in the position I wanted, and in this way I was helped in completing the hands of the full-size sculpture.11
In addition to using his wife and daughter as models, the scholar Alan Wilkinson has noted that ‘Tam Miller, Moore’s secretary from 1950 to 1956, also recalls standing in Moore’s studio for two hours at a time with her hands on a little platform at elbow height’.12 Similarly, the curator David Mitchinson has recorded that Moore’s studio assistant Alan Ingham also modelled for the feet of the King.13 The remodelled Hands of Queen were cast separately and exist as distinct sculptures in their own right.
Fig.15
Full-size plaster of King and Queen 1952–3 in the garden at Hoglands
© The Henry Moore Foundation. All Rights Reserved
The naturalistic designs for the hands and feet were then incorporated into the full-size plaster model used to cast the sculpture in bronze (fig.15). This plaster version was constructed on an armature using the plaster working models as guides. It is known that Moore used iron rods in the armature for the 1951 sculpture Reclining Figure (Tate T02270), which was then coated with layers of plaster, and it is likely that he used a similar method to create King and Queen.
Moore made several different versions of the Queen’s head. In 1968 Moore recalled that ‘the head of the Queen was a problem because it had to be in harmony and I made two or three different attempts at it before being satisfied’.14 Since the head of the King was so distinctive it was important that the Queen’s head reflected the regality of her counterpart and thus convey a sense of unity, but also projected an identity of its own. Study for Head of Queen No.2 1952–3 (fig.16) appears to be similar to the head of the bronze maquette, while Study for Head of Queen 1952–3 (fig.17) is more naturalistic, with a clearly recognisable crown. The nature of plaster is such that Moore would have been able to remove the heads and limbs of the figures and replace them with alternatives. Although it does not appear to show either of the surviving designs, a photograph of the full size plaster taken inside Moore’s studio records the Queen with a different head to that of the final bronze. A seam running around the circumference of the Queen’s neck indicates where the head has been attached and it is notable that in this photograph the head of the Queen is looking towards the King (fig.18). However, Moore clearly found the Queen’s head and its orientation unsatisfactory because he reworked the design to present the face in a thin, fin-like form, like the King’s, and replicated his curved crown.
Henry Moore 'Study for Head of Queen No.2' 1952–3
Fig.16
Henry Moore
Study for Head of Queen No.2 1952–3
Henry Moore Foundation
© The Henry Moore Foundation. All Rights Reserved
Henry Moore 'Study for Head of Queen' 1952–3
Fig.17
Henry Moore
Study for Head of Queen 1952–3
Henry Moore Foundation
© The Henry Moore Foundation. All Rights Reserved

When the plaster was finished it was placed outside in his garden. Given the cramped conditions of Moore’s studio it is likely that Moore placed the sculpture outside so that he could look at it from a distance without other works blocking his view. At this time Moore experimented with the inclusion of the thin rectangular frame, as found in the original maquette. However, he ultimately decided against including this frame in the final version. When asked why he had omitted it Moore reportedly told Alan Wilkinson that, ‘oh, on a larger scale, they would look as if they were keeping goal in a soccer match’.15
Fig.18
Full-size plaster of King and Queen 1952–3 in Moore's studio with an alternative queen's head
© The Henry Moore Foundation. All Rights Reserved

The full-size plaster was completed in early 1953 and sent to the foundry for casting. Moore employed the Art Bronze Foundry in London to cast the first bronze example of King and Queen. He had previously commissioned this foundry to cast his maquettes for Madonna and Child in 1943 (Tate N05600–N05603) and he continued to use their services throughout the 1950s. It is likely that Moore first came into contact with the foundry when he was Head of the Sculpture Department at Chelsea College of Art in the 1930s; the foundry was located close to the College and was regularly used by staff and students. At the foundry, the technicians would use the plaster to create a hollow mould, into which molten bronze could be poured. King and Queen was originally cast in an edition of four, plus one artist’s copy. The fifth cast, which is now owned by Tate, was made in 1957. A second artist’s cast was made in 1985 and is now in the collection of the Henry Moore Foundation.
After casting at the foundry the bronze sculptures were returned to Moore so that he could check the quality of the casting and make decisions about the patination. A patina is the surface colour of a sculpture and is usually achieved by applying chemical solutions to the pre-heated bronze sculpture. In 1960 Moore explained:
I like working on all my bronzes after they come back from the foundry. A new cast to begin with is just like a new-minted penny, with a kind of slight tarnished effect on it. Sometimes this is right and suitable for a sculpture, but not always. Bronze is very sensitive to chemicals, and bronze naturally in the open air (particularly near the sea) will turn with time and the action of the atmosphere to a beautiful green. But sometimes one can’t wait for nature to have its go at the bronze, and you can speed it up by treating the bronze with different acids which will produce different effects. Some will turn the bronze black, others will turn it green, others will turn it red.16
Moore went on to explain that when working on his plasters he normally had a preconceived idea of what colour the sculpture would eventually take. However, he acknowledged that working with chemical patinas was not easy, and that it was difficult to replicate the same colours and effects across the edition.17 Tate’s King and Queen has a fairly consistent brown patina, although the surface of some high points, such as the Queen’s shoulders, has been worked to produce a golden-brown finish. Moore explained in 1960 that after patination ‘you can then work on the bronze, work on the surface and let the bronze come through again, after you’ve made certain patinas. You rub it and wear it down as you hand might by a lot of handling. From this point of view bronze is the most responsive and unbelievably varied material’.18 In contrast to Tate’s cast, the King and Queen in the collection of the Henry Moore Foundation, which is displayed outside, has turned a mottled sea-green, just as Moore predicted.
King and Queen is Moore’s only sculpture depicting solely a pair of adult figures. Although Moore said that he recognised the male head to be that of a king as soon as he made it, according to the curator David Mitchinson, ‘in one of Moore’s early sale record books this sculpture is rather prosaically called Two Seated Figures’.19 This title was occasionally used concurrently with King and Queen. For example, the sculpture was titled Two Seated Figures while on exhibition at the Musée Rodin in Paris in 1961 and when it was included in Moore’s exhibition at the Israel Museum in Jerusalem in 1966. Moore’s first statement on King and Queen was published in the exhibition catalogue for his solo show at Curt Valentin Gallery in New York in 1954:
The King and Queen group has nothing to do with present-day Kings and Queens but is more connected with the archaic or primitive idea of a King. The ‘clue’ to the group is perhaps the head of the King which is a head and a crown, face and beard combined into one form and in my mind has some slight Pan-like suggestion, almost animal, and yet, I think, something Kingly. How the group came about I don’t know, unless it may be that in the last year or two I have read stories to my daughter in which Kings and Queens have appeared a lot and this might have made one’s mind open to such a subject.20
Henry Moore 'Goat's Head' 1952
Fig.19
Henry Moore
Goat's Head 1952
The Henry Moore Foundation
© The Henry Moore Foundation. All Rights Reserved
The relation that Moore drew between the head of the King and Pan, the ancient Greek god of the wild, reinforces the sense that the figures in the sculpture represent archaic, ancient or otherworldly characters. Pan’s body was an amalgamation of animal and human form; he had the haunches of a goat and horns on his otherwise human head. Indeed, the art historian Will Grohmann has compared the King’s head with Goat’s Head 1952 (fig.19), which Moore made around the same time that he made King and Queen. Goat’s Head has a long thin snout, a pointed nose and horns that resemble the thin, elongated shape of the King’s head, beard and crown.
Fig.20
Limestone statue of a husband and wife c.1300 BC
British Museum, London
© Trustees of The British Museum
Moore’s reference to the ‘primitive’, a term that has been used to describe ancient, non-Western art forms, may also explain why he subsequently identified an ancient Egyptian sculpture from the British Museum as one of the sources for King and Queen.21 In the ancient Egyptian sculpture the two figures sit side by side in long, enveloping robes with their feet planted firmly on the ground (fig.20). A photograph of this limestone sculpture of a husband and wife was included in the publication Henry Moore at the British Museum (1981), which contained artworks from the British Museum’s collection that had influenced Moore during his career.22 In the book Moore explained why he admired this particular sculpture, stating:
This has always been a great favourite of mine. For me these two people are terribly real and I feel the difference between male and female. The sculptor had done it in an obvious way by making the man slightly bigger than the woman, but it works, and this influenced me when I came to make my bronze King and Queen. It is such a pity the hands are damaged for, after the face, I think the hands are the most expressive part of the body. But even damaged the arms have a superb sense of repose and serenity which is so characteristic of Egyptian sculpture. Notice too that there are no marks of aging on the faces. The pair are represented at an ideal age, one of full growth but before disillusionment has set in.23
Fig.21
Henry Moore with the ancient Egyptian statue of a husband and wife in the British Museum, 1980–1
© The Henry Moore Foundation. All Rights Reserved
Photo: David Finn, Henry Moore Foundation Archive
Moore’s biographer Donald Hall noted in 1966 that the stories Moore read to Mary and the ancient Egyptian sculpture were only identified as possible sources for King and Queen after the sculpture was made, remarking that ‘in the years since he made the first maquette, Moore has thought of two sources for the King and Queen; he was aware of neither at the time’.24 Nonetheless, the Egyptian sculpture has come to play a central role in accounts of the making of King and Queen. The subsequent importance accorded to the sculpture was encouraged by the existence of photographs of Moore looking at the sculpture admiringly (fig.21) and by Moore’s own statements. Indeed, in 1986 he stated that:
It was a big subject, the King and Queen. When I was working on them, I was reminded of an Egyptian sculpture in the British Museum that I had seen many times of a seated figure of an official and his wife. But somehow the sculptor had raised them above this status and had given them greater dignity and self-assurance, almost a nobility of purpose to make them appear above normal life. I’ve tried to inflect some of this feeling into my sculpture.25
Moreover, in 1998 Tate lent King and Queen to the British Museum where it was displayed opposite the Egyptian sculpture, as if the two couples were in conversation.
However, as early as 1960 Grohmann questioned the extent to which Moore’s King and Queen could be regarded as a modern counterpart to the Egyptian sculpture in the British Museum. Grohmann suggested that ‘to see references to Egyptian royal groups, such as the 18th-dynasty group in the British Museum, is incorrect; Moore could only have learnt from them what he did not want to do’.26 He went on to explain that unlike the ancient Egyptian couple, ‘there is nothing imperious about this royal pair; on the contrary, it is a very human couple’.27
Fig.22
Queen Elizabeth II and Prince Philip, Duke of Edinburgh. Coronation portrait, June 1953, London, England
Photo: Cecil Beaton
Although Moore claimed that the ‘King and Queen group has nothing to do with present-day Kings and Queens’, the filmmaker John Read has suggested that it may nonetheless be instructive to consider Moore’s sculpture, which was completed in the year of Queen Elizabeth II’s coronation, in light of the ‘patriotic fervour stimulated by the Coronation and what was often called the beginning of a New Elizabethan Age’ (fig.22).28 It seems an unlikely coincidence that the only identified couple in Moore’s output – a King and Queen – should have been made in the same year that Great Britain and the Commonwealth welcomed a new monarch. King and Queen responded to the same communal desires embodied by Moore’s Family Group, which expressed the post-war ideals of the welfare state. For Read, the figures in King and Queen ‘have an air of authority, but their grouping side by side emphasises their domesticity. The realistic modelling of their hands and feet illustrate their humanity. This royal family serves as a multiple of parenthood and of the hieratic aspect of a couple who are also the symbolic parents of a nation, at one and the same time, stern, protective and remote’.29
Fig.23
Henry and Irina Moore c.1952
© The Henry Moore Foundation. All Rights Reserved
Photo: Jitendra Arya, Henry Moore Foundation Archive
In 2000 the curator Anita Feldman Bennet noted that at various times Moore lent credence to accounts of the origins of King and Queen that referred to children’s fairy tales, Egyptian sculpture and the Coronation. However, she concluded that ‘perhaps the most convincing is that the poses were simply taken from a contemporary photograph of the artist and his wife Irina in Perry Green’ (fig.23), noting that Moore could take inspiration ‘from anything and anywhere’.30 While the poses of Moore and his wife do correspond to those of King and Queen, it seems unlikely that Moore would have copied the pose so directly. Many of Moore’s figurative drawings from the 1940s present their subjects in similar seated arrangements; for example, the presentation of a woman sitting with her hands in her lap might equally have stemmed from one of Moore’s Shelter Drawings titled Woman Seated in the Underground 1941 (Tate N05707).
The earliest cast of King and Queen was first exhibited at the Second Biennial for Sculpture at Middelheim Park in Antwerp in the summer of 1953. It was bought by the city of Antwerp and is now part of the permanent collection of the Middelheim Sculpture Park. Another cast was exhibited in London at the Leicester Galleries in February 1954. At this time, critical responses to the work focused on the form the sculpture took and the combination of abstract and figurative styles, which were unusual to find together in a single sculpture. Writing in the Observer, critic Nigel Gosling suggested that ‘the mingling of naturalistic limbs with only semi-representational heads is a surrealist gambit’, which he suggested was ‘disturbing’.31 In a review titled ‘Mr Moore’s New Bronzes: An Experimental Phase’, the unnamed critic writing for the Times suggested that the works in the exhibition, including King and Queen, ‘are prone to inconsistencies of style, which in spite of their many passages of subtle observation, give them the air of experiments which have yet to be resolved’.32 On a more positive note, Stephen Bone writing in the Guardian noted that Moore’s ‘strange inconsistencies’ in King and Queen were ‘plainly deliberate and carefully considered’, concluding that ‘the result is far more successful than might have been expected, though a faintly uneasy air of experiment hangs about this group’.33
Perhaps the most significant – but scathing – review of King and Queen was written by the critic David Sylvester and published in the journal Britain Today in March 1954. Sylvester had curated Moore’s 1951 exhibition at the Tate Gallery and was a close friend and collaborator. Although he started his review by noting the dichotomy of Moore’s recent work, which veered from the figurative Draped Reclining Figure 1952–3 (Henry Moore Foundation) to the geometric arrangements of the Time Life Screen 1952–3 (Time Life Building, London), when considering the union of abstract and naturalistic forms in King and Queen Sylvester proclaimed that ‘I, for my part, do not find these deliberate inconsistencies acceptable. I would like to see the King’s head existing as a work in its own right. Still more I would like to see the exquisitely modeled and proportioned feet and hands attached to bodies of an equally serene beauty’.34 Sylvester concluded stating, ‘Moore’s recent major works suggest that at present his aesthetic personality is a split one’.35 With these remarks in mind, it is telling why Sylvester chose not to include King and Queen in the retrospective of Moore’s work that he curated at the Tate Gallery in 1968.
A year after Sylvester’s review was published Moore explained why he consolidated different representational forms in King and Queen: ‘some parts of it are more realistic than others – the hands and feet particularly – to bring out the contrast between human grace and the concept of power in primitive kingship’.36 He reiterated his position in 1957 during an interview with the critic J.P. Hodin:
After Surrealism nobody should be upset that a head of one of my sculptures is different from the rest. Could not we say the same of Chartres, or of many primitive works? In Chartres Cathedral, the bodies are like columns and the heads are realistic. No one reproaches them for disunity of style. I willingly accept what I try to bring together. In the heads of my “King and Queen” ... some mixture of degrees of realism is implicit. But we got used to this mixture in Chartres, and we shall get used to it again.37
For critic Peter Anselm Riedl, writing for a German audience in 1957, it was the strange disparity between various component parts of King and Queen that made the sculpture so powerful. Riedl described how ‘the two beings are both familiar and unfamiliar to us at the same time’.38 In a discussion of the figures’ heads he noted:
The two figures have eyes like those of an insect, somehow dangerous and yet somehow noncommittal. In fact, they are not eyes at all, just holes, channels, that make the heads see-through. But this just creates a strange spell-like effect. One feels haunted by an invisible gaze that is actually wandering elsewhere – into the distance, into the void.39
In 1966 Donald Hall narrated that, ‘the large version, he [Moore] says, allowed him to carry further the human implications of the King and Queen. Working over it, he felt that he could add meaning, or slant meaning, by how he made the hands. The King’s hand resting on his knee could be clenched like a dictator’s, or floppy like a weak king’s, or in between’.40
Moore always paid careful attention to the depiction of hands and in 1980 reflected that, ‘hands can convey so much – they can beg or refuse, take or give, be open or clenched, show content or anxiety. They can be young or old, beautiful or deformed ... throughout the history of sculpture and painting one can find that artists have shown through the hands the feelings they wished to represent’.41 Like the hands of his earlier carving Girl 1931 (Tate N06078), the hands of the Queen have not just been modelled sensitively but suggest a sensitivity of touch. The fingers are gently interlocked but do not appear to apply any pressure to the skin.
Despite the ambivalence of some critics when it was first exhibited, King and Queen quickly cemented itself as one of Moore’s best known sculptures. As early as 1954 the critic Robert Melville claimed that ‘it is Moore’s finest achievement since the war, and probably the most graceful of all his works’.42 For Melville the sculpture demonstrated a ‘resolution of conflict between a static, hieratic approach and a dynamic, humanist one’, marrying a sense of primal longevity with an alert awareness of the immediate moment.43 Melville discussed the figure using phrases such as ‘magic and terror’, and contrasted the two figures, suggesting that the presence of the graceful Queen ‘is a kind of guarantee that the King will not peck out the hearts of his subjects’.44 The otherworldliness of King and Queen was also noted by the critic Herbert Read. In his introduction to the second volume of Moore’s catalogue raisonné (first published in 1955), Read noted that although the sculpture could be linked to the earlier family groups, ‘there is an equally obvious break, and it is a break with humanism and an advance into the superhuman realm of myth. This king and queen never reigned in our world’.45 Similarly in 1957 Riedl suggested that the ceremonial rigidity of the figures and their aimless gazing removed them from the activities of the everyday world, and placed them in a realm ‘beyond “doing”’.46
By 1960 Grohmann praised King and Queen ‘as a highwater mark in Moore’s creative work, a monument – for that is what it is – timeless and without specific purpose’.47 He went on to note that ‘it quickly won public recognition and higher esteem than the more naturalistic Madonnas at Northampton and for St Peter’s Church in Claydon’.48 Suggesting that this public esteem was a result of the multiplicity of styles found in the work, Grohmann argued that in combining human, animal and non-representational forms, Moore’s sculpture embodied ‘the totality of the world, sculpturally speaking of the unity of natural and supernatural, objective and abstract. Thus there is a synthesis here too, synthesis in the combination of the archaic with the contemporary, the unconscious with the spiritual’.49
This interpretation of the sculpture as somehow archaic, magical and mysterious was supported by the widespread reproduction of photographs taken by Moore himself showing King and Queen positioned outside in the Scottish countryside (fig.24). This cast of the sculpture was bought by W.J. Keswick in 1954 and installed on his estate, Glenkiln, in Dumfriesshire. In 1960 Moore described how the work came to be placed outside:
When Mr Keswick invited me to stay with him, I went all over his land, and there are not only the wild hills and moors but woods and lawns, and a lake with an island in the middle. I saw at least twenty or thirty sites for the sculpture which are separated from one another by hillside or trees, so that one could never see more than one sculpture at a time. There was such a large variety that I feel that every large sculpture I have done could find a suitable setting there. He showed me a scar on one hillside – a kind of scooping out, that formed a natural dais. We thought that this would be the right spot for King and Queen, and here eventually the two figures were placed. They look across forty miles of countryside into England.50
The placement of this cast of King and Queen in an isolated landscape has shaped interpretations of the sculpture more generally. In his book The Archetypal World of Henry Moore (1959), the psychologist Erich Neumann described the couple as lonely and desolate, an interpretation that is supported by the inclusion of a photograph of the sculpture at Glenkiln in which the two figures appear isolated on a rocky bluff.51
Fig.24
King and Queen 1952–3 installed on the Keswick Estate, Glenkiln, Dumfriesshire, c.1954
© The Henry Moore Foundation. All Rights Reserved

According to the photographer David Finn, ‘Moore has always loved photographs showing the piece on the hills of Glenkiln’.52 Indeed, Moore kept a selection of photographs showing the work from a variety of viewpoints pinned to a notice board in his studio. Some of these photographs of King and Queen in situ in the Scottish landscape were reproduced in Moore’s catalogue raisonné published in 1955, and regularly featured in articles and books about the artist, including the monographs written by Herbert Read (1965), Donald Hall (1966), and John Russell (1973) among others. The experience of installing this work and other sculptures at Glenkiln encouraged Moore to place more of his work outside. In 1978 he reflected on the importance of the open sky as a backdrop for his work, and concluded that ‘sculpture – at any rate, my sculpture – is best seen in this way and not in a museum’.53 In the foreword written for David Finn’s book, Henry Moore: Sculpture and Environment (1977), the former director of the National Gallery, Kenneth Clark, wrote that ‘the supreme examples of siting Moore’s work are Tony Keswick’s placements of four pieces on his sheep farm at Glenkiln in Dumfriesshire, Scotland ... everyone who sees the works in this setting comes away deeply moved ... The siting of the Glenkiln pieces is what one might call a Wordsworthian use of sculpture’.54
In a talk given in 1955 about sculpture in the open air, Moore noted that he had already begun work on the idea for the piece before it had been selected for the Second Biennial for Sculpture at Middelheim Park in Antwerp, making the point that he felt free to sell the same work to other clients: ‘although this Group has been listed as a commission from the Burgomaster of Antwerp, the maquette had already been produced before Antwerp decided to have a piece of my sculpture for the open-air museum in Middelheim Park. It was simply a sculpture that I wanted to do’.55 The second cast was bought by W.J. Keswick in 1954 for the Keswick Estate at Glenkiln and the third cast was bought by Joseph H. Hirshhorn through the Curt Valentin gallery in New York and is now housed in the Hirshhorn Museum and Sculpture Garden in Washington, D.C.. The fourth version cast in 1953 was bought for £3,000 by David Astor, a former editor of the Observer newspaper, from Moore’s 1954 Leicester Galleries exhibition. Astor displayed it in his garden in St John’s Wood, London, until 1976, when he sold it to the Norton Simon Museum of Art in Pasadena. The original artist’s copy was held in the collection of the artist’s daughter but is now in the collection of the Moa Museum of Art in Atami, Japan.
On 16 May 1957 the Director of the Tate Gallery, John Rothenstein, met with the Board of Trustees to discuss proposals to expand the gallery’s holdings of works by Henry Moore. Rothenstein explained that Moore had suggested that the gallery purchase eight sculptures representing the different phases of his work since the early 1930s. By this time the four original casts of King and Queen had already been sold, but it was suggested that an additional cast could be made especially for Tate should the gallery wish to acquire it. Although the trustees decided not to purchase all of the suggested sculptures immediately, they did agree to the acquisition of King and Queen. According to the minutes of the meeting, ‘the Director emphasised the extreme generosity and helpfulness of the sculptor’s attitude in making works available from his own collection for sums substantially below their market value ... and for agreeing to seek the approval of the authorities of the City of Antwerp for making an extra cast of King and Queen’.56 The owners of the first four casts had bought the sculpture on the understanding that no other casts would be made, and might have complained that a fifth example would diminish the value of their pieces. Moore therefore was obliged to seek their approval for this idea. In fact, the owners readily agreed as this additional cast would not enter the commercial market and would be acquired directly by the Tate Gallery. By 3 June 1957 Moore was able to write to Rothenstein to say ‘I am straight away fixing with the foundry for the casting of the King and Queen’.57
The casting and patination of Tate’s cast, however, took well over a year. In a letter to Moore dated 18 July 1958 Rothenstein noted the he had informed the trustees of the artist’s generosity in lowering the price to ‘virtually the cost of the casting, namely £1,200’.58 The purchase was also made possible due to ‘funds made available by Associated Rediffusion’, the first independent commercial television broadcaster in the London region, formed following the 1954 Independent Television Act of Parliament, which broke the BBC’s monopoly over television broadcasting. 59
In November 1958 the Deputy Director, Norman Reid, wrote to Moore asking whether King and Queen would be available for viewing at the meeting of the Friends of the Tate, which was due to take place on 4 December. Reid enquired, ‘Are you ready to part with it yet or do you think it should stay longer with you for the sake of the patina?’60 Moore evidently felt that the sculpture should remain in his possession for a little while longer because it was not until late January that arrangements were made for the transportation of King and Queen from Moore’s studio in Perry Green to the Tate. The sculpture was first exhibited at the gallery on 4 February 1959.

Alice Correia
December 2013

Notes

1
Another drawing by Moore dating from 1949 depicting a male and female couple sitting on a low bench was reproduced in Alan Bowness (ed.), Henry Moore. Volume 2: Sculpture and Drawings 1949–1955, 1955, 2nd edn, London 1965, pl.96.
2
Henry Moore cited in John Hedgecoe (ed.), Henry Moore, London 1968, p.221.
3
Donald Hall, Henry Moore: The Life and Work of a Great Sculptor, London 1966, pp.130–1.
4
Anthony Caro, ‘King and Queen 1952–53’, in David Mitchinson (ed.), Celebrating Moore: Works from the Collection of the Henry Moore Foundation, London 2006, p.241.
5
Hall 1966, p.131.
6
Ibid.
7
Julie Summers, ‘Fragment of Maquette for King and Queen’, in Claude Allemand-Cosneau, Manfred Fath and David Mitchinson (eds.), Henry Moore From the Inside Out: Plasters, Carvings and Drawings, Munich 1996, p.126.
8
Caro 2006, p.242.
9
Summers 1996, p.126.
10
Moore cited in Hedgecoe 1968, p.232.
11
Henry Moore, letter to Allan D. Emil, 21 October 1966, reprinted in Alan Wilkinson (ed.), Henry Moore: Writings and Conversations, Aldershot 2002, p.283. This letter is held in the Archives of American Art, Washington, D.C. See http://www.aaa.si.edu/collections/allan-d-and-kate-s-emil-papers-8502, accessed 29 January 2014.
12
Wilkinson 2002, p.283, note 22.
13
David Mitchinson, ‘King and Queen’, in Henry Moore: War and Utility, exhibition catalogue, Imperial War Museum, London 2006, p.48.
14
Moore cited in Hedgecoe 1968, p.221.
15
Alan G. Wilkinson, Henry Moore Remembered: The Collection at the Art Gallery of Ontario in Toronto, Toronto 1987, p.140.
16
Moore 1960, reprinted in Wilkinson 2002, p.234.
17
Ibid., p.234.
18
Moore 1960, reprinted in Wilkinson 2002, p.234.
19
Mitchinson, ‘King and Queen’, 2006, p.48.
20
Moore cited in Hedgecoe 1968, p.221.
21
As Moore’s biographer Donald Hall recalled: ‘An Egyptian Limestone Seated Figure is the sculpture Moore remembers especially. The official and his wife are rocky and solid.’ Hall 1966, pp.129–30.
22
In his introduction to the book Moore concluded that ‘It has been a wonderful experience for me to recapture the delight, the excitement, the inspiration I got in these pieces as a young and developing sculptor’. Henry Moore, Henry Moore at the British Museum, London 1981, p.16.
23
Ibid., p.38.
24
Hall 1966, p.129.
25
Henry Moore cited in Hedgecoe 1986, p.156.
26
Will Grohmann, The Art of Henry Moore, London 1960, p.147.
27
Ibid.
28
John Read, ‘King and Queen, 1952–3’, in Mitchinson, Celebrating Moore, 2006, p.238.
29
John Read, Portrait of an Artist: Henry Moore, London 1979, p.110.
30
Anita Feldman Bennet, ‘Rediscovering Humanism’, in Henry Moore: In the Light of Greece, exhibition catalogue, Basil and Elise Goulandris Foundation Museum of Contemporary Art, Andros 2000, p.65. A copy of this portrait, which was taken by the photographer Jitendra Arya around 1952, is held in the Henry Moore Foundation Archive, although it is uncertain when the image came into Moore’s possession.
31
Nigel Gosling, ‘Royalty’, Observer, 21 February 1954, p.11.   
32
Anon., ‘Mr Moore’s New Bronzes: An Experimental Phase’, Times, 15 February 1954, p.4. 
33
Stephen Bone, ‘Recent Works by Henry Moore’, Manchester Guardian, 12 February 1954, p.5.
34
David Sylvester, ‘Henry Moore’s Sculpture’, Britain Today, no.215, March 1954, pp.34–5.
35
Ibid., p.35.
36
Henry Moore, ‘Sculpture in the Open Air: A Talk by Henry Moore on his Sculpture and its Placing in Open-Air Sites’, March 1955, p.7, Tate Artist Catalogue File, Henry Moore, A23941.
37
Henry Moore, ‘Voice of the Artist 1: The Hidden Struggle’, Observer, 24 November 1957, p.3.
38
Peter Anselm Riedl, Henry Moore: König und Königin, Stuttgart 1957, p.4.
39
Ibid., p.7.
40
Hall 1966, p.131.
41
Henry Moore cited in Anita Feldman and Malcolm Woodward, Henry Moore: Plasters, London 2011, p.66.
42
Robert Melville, ‘Henry Moore and the Siting of Public Sculpture’, Architectural Review, February 1954, pp.88–9.
43
Ibid., p.89.
44
Ibid.
45
Herbert Read, ‘Introduction’, in Alan Bowness (ed.), Henry Moore. Volume 2: Sculpture and Drawings 1949–1955, 1955, 2nd edn, London 1965, p.x.
46
Riedl 1957, p.5.
47
Grohmann 1960, p.148.
48
Ibid.
49
Ibid.
50
Henry Moore cited in ‘Sculpture in Landscape’, Selection, Autumn 1962, pp.12, 15, reprinted in Wilkinson 2002, p.246.
51
Erich Neumann, The Archetypal World of Henry Moore, London 1959, p.115.
52
David Finn, Henry Moore: Sculpture and Environment, London 1977, p.298.
53
Henry Moore cited in Stephen Spender, Henry Moore: Sculptures in Landscape, London 1978, p.26.
54
Kenneth Clark, ‘Foreword’, in Finn 1977, p.20.
55
Moore 1955, p.7, Tate Artist Catalogue File, Henry Moore, A23941.
56
Minutes of a Meeting of the Trustees of the Tate Gallery, 16 May 1957, Tate Public Records TG 4/2/742/2.
57
Henry Moore, letter to John Rothenstein, 3 June 1957, Tate Public Records TG 4/2/742/2.
58
John Rothenstein, letter to Henry Moore, 18 July 1958, Tate Public Records TG 4/2/742/2.
59
Ibid. The company had previously been a radio broadcaster under the name Rediffusion. No details about how the company came to provide funds to the Tate Gallery for the acquisition of Moore’s sculpture are held in the Tate Public Records.
60
Norman Reid, letter to Henry Moore, 17 November 1958, Tate Public Records TG 4/2/742/2.

How to cite

Alice Correia, ‘King and Queen 1952–3, cast 1957 by Henry Moore OM, CH’, catalogue entry, December 2013, in Henry Moore: Sculptural Process and Public Identity, Tate Research Publication, 2015, https://www.tate.org.uk/art/research-publications/henry-moore/henry-moore-om-ch-king-and-queen-r1172098, accessed 25 April 2024.