10 Illumination of the Stage


THIS chapter explores the essential basis of lighting: how do
we see, and, because the subject is stage lighting, how can one
influence what is seen? The first essential is to repeat once
again that, although we talk about looking at an object, the
eye is a receiver not a transmitter. Light comes from a source
and is reflected, probably only partially in respect of intensity
and colour, and ultimately reaches the eye. Once the light
arrives there it is automatically focused on the retina and
transmitted to the brain. Only then does it become a picture
capable of rousing interest and possibly emotion. No light in
the first place and there will be no picture. Seeing is a matter
of external stimuli.
 If the light is poor, then the picture will be so incomplete
that the brain has to supply the missing bits, and this may give
rise to a continual strain and fatigue. It is not the eye that is
strained with too little or distorting light: it is the brain.
The eye will carry out instantaneously an adaptation to the
levels of light of some eight to one by opening the iris wide
to pass more when there is little, and closing it when there is
a lot of light. Certain chemical changes also go on at the
retina and, provided sufficient time has elapsed, the normal
eye can adapt itself quite easily to either the 0ú02 foot-candles
of moonlight or 10,000 foot-candles of sunlight. At very low
levels, like moonlight, the eye cannot perceive colour at all'
and at all levels of intensity it may be quite unable to perceive
form correctly. The fault in the first case is inherent in the
eye itself, but in the latter case it will be the lighting that is
wrong. The picture may be on the retina but cannot be
interpreted by the brain correctly. It is like listening to a con-
versation in a language in which one understands only some
of the words.
1. Black and white vision is known as "scotopic" and concerns rod receptors all
over the retina. These are very sensitive to light and one can literally see out of the
corner of the eye. Colour "photopic" and fine-detail vision needs cone receptors
which are concentrated in the centre. The angle embraced by this vision is very
small indeed and the eye has to scan the scene to bring it to bear.
 Personally, I consider optical illusion--the deliberate de-
ceiving of the brain--as a large part to play in the theatre,
and it is therefore considered later on.

Visual Perception and Lighting to See

Visual perception in the theatre suffers from the serious handi-
cap that members of the audience each get their own angle
of view of the stage. In the cinema the picture is presented
complete on one plate. It was lit for one viewer, the camera
lens, and yet even there, when projected, troubles may be
encountered of physical obstruction due to poor sight lines or
too distant or too close or too angular a view. In the theatre
the picture is three-dimensional, usually as deep as it is wide,
and the range of angular view encountered even in the most
orthodox of picture frame proscenium theatres is extremely
wide and varied. In the circle one looks downwards and sees
mainly stage floor, whereas in the stalls the view of the stage
floor may be minimal or non-existent. To those in the stalls
the actors may appear crisply lit against a dark night sky and
the producer congratulates himself on a lack of offensive
shadows there; but take a trip to the circle and the actors
will be seen against a confused collection of pools of light and
shadow on the floor. Of course, one accepts the idea that
not everyone will get a view of all the backcloth or of; perhaps,
the scenery on one side of the stage or the other, but not to
see the actors properly, that is quite another matter. After all,
any producer knows that their principal action must be
concentrated in that magic, and often all too limited, part of
the stage, the acting area to which all audience have uninter-
rupted sight lines. It is, however, no good doing this if that
area is not going to be lit properly.
 Where everyone looks down at the stage floor only from one
side the problem is eased, but in the old multi-tier house the
reluctant conclusion may well be that only the flat lighting
from the days when they were built--the battens and footlights
--is really appropriate. This is an extreme statement but has
more than a grain of truth behind it and may explain the
multiplicity of spotlights, the very negation or spotlighting,
encountered in West End theatres.
A very good place to begin our study of seeing is with the
geometric solids of the art school. In drawing either the cube
or tile cone, for example, one would not expect them to appear
solid without some shading on one side. This seems obvious
when trying to reproduce these three-dimensional objects on
a flat two-dimensional sheet of paper (Fig. 113), but applies
equally well when trying in fact to draw the same picture on
each of the two-dimensional retinas of the eyes of the audience.
It follows, therefore, that any lighting of these solids must be
arranged to cast some shadow. Too heavy a shadow could
become confusing if (Fig. 114) the objects are seen against a
dark ground. All becomes well again in the illustration (Fig.
115) where the camera angle provides a view of the baseboard
and the shadow of the object gives a further clue. This angle
is deliberately chosen to emulate the ideal theatre angle of
view, the baseboard becoming the stage floor.
 Essential key lighting is obviously that from the left side and
a second source (fill) from an opposing angle has been used to
lessen the contrast. The two shadows on the floor might then
conflict, particularly if sunlight were supposed to be represented,
and therefore one would try to use a softlight type of source
(Fig. 116) throwing little or no defined shadow at the second
position. On the other hand, if shadow-free soft light from all
directions was all that was available, then shape could be put
back in the cube by painting one of its sides in a darker colour.
The next step is to make the cube two-dimensional only and
 paint in the perspective and shadow. Under certain illumina-
tion as in the second photograph (Fig. 117), the eye would not
only be completely deceived but might take the two-dimen-
sional painted object as appearing more solid than the solid
itself.
However, the principal target for stage lighting, the actor,
is a solid and a very complex one at that. Lighting him, as
for a portrait, the key light would be placed to the left, but, as
heavy eye-socket shadows and nose shadows would be un-
desirable, this light could be neither too high up nor too
acutely to the side. Even with the most skilful placing, how-
ever, light is not going to turn corners and light the other side
of the face, so a second unit of less intensity will be placed at
the opposing angle. For photography, a third unit, a backlight
directed in reverse, would help separate the hair and shoulders
from the background. Thus, for one position and one member
of the audience, we have three lighting units (Fig. 118), and
if the floor were seen, as it must be in the theatre, then there
would be three conflicting shadows.
When, in addition, one remembers that there are in the
theatre several hundred other cameras all at different angles--
the eyes of the audience, so to speak--and that acting is not
static and usually involves several characters, an insoluble
problem appears to be presented. Fortunately, the eye of the
playgoer is only as perceptive as his brain and, generally
speaking, the latter is only capable of giving full concentration
to one visual idea at a time. Thus, if lighting provides some
acceptable dominant idea, the rest of the information gets pushed
into the background. The brain even seems to swallow the
notion of moonlight coming from two equal and opposing direc-
tions at once--the wings--as in most productions of the ballet Les
Sylphides, simply because the scene is painted as a moonlight
glade and the light is conventionally blue. Provided the show
is suitably gripping, many, but not all, of the various defects,
conflicting shadows and so on will pass muster. I have absolute
faith in a single dominant idea and no faith in lighting with
several ideas, for this will be interpreted as none. Lighting
whose sole purpose is to make the actor visible is vulnerable to
attack. The audience start subconsciously attaching their own
interpretations to what they see, and as in fact there is nothing
to interpret they are bound to find something to criticize.

Lighting an Open-end Stage

 Now let us consider the portrait lighting blown up to cover
the whole stage. It might be assumed that, if the three lanterns
giving key, fill and backlight are moved farther away and
made to cover the stage, this would be all that is necessary.
To some extent this is true and for end-stage work, the third
unit, the backlight, can usually be omitted. Colour denied to
black and white photographs or television may well provide
the necessary separation, so this brings the minimum number
of lamps for an end stage down to two.
 This really does work out; a spotlight out front either side
flooded to cover the entire acting area can provide a pleasing
effect, especially if one is a slightly different colour from the
other or somewhat less bright (Fig. 119).
 Of course, reliance on only two sources imposes limitations
even on an open-end stage, but on a picture-frame stage with
scenery it would make a nonsense of the setting. For this, like
the actors, requires lighting and what suits one is unlikely to
suit the other. However, it is worth while to pursue the study
of the "two source only" lighting rather further. Assuming
no scenery, the limitations would make themselves felt as an
addition to the movement problem well known to producers
already as masking. An actor might all too easily block the
light which should illuminate his neighbour. To some extent
this could be overcome by placing the lights relatively high
so that the beams would pass overhead, and by ensuring that
actors did not get too close to each other--rather troublesome
in a courtship scene! A further problem may arise in revealing
the actor's features as he moves away from the centre of the
acting area. It is going to be extremely difficult in practice to
light all his positions on the stage from the essential minimum
of two angles. Further, these two angles must not just be any
two angles; they must be the correct two angles for proper
modelling from the audience point of view.
 It will be noticed that when considering the lighting of an
end stage with but two lanterns two important principles arise.
The first is that the lanterns are automatically placed "out
front" and the second is that the beams are crossed. Each of
the two lanterns is directed to light the farther side of the stage
while lapping over the other in the centre. This provides an
effective compromise since action, being concerned with the
relationships of one character to another, tends to turn inwards.
One could say that the on-stage rather than the off-stage
aspect of the actors is lit. Furthermore, any shadows tend to
be projected to the outer edges of the stage where they are less
distracting. The light itself comes from the direction of the
audience and this helps its redirection back to their eyes and
allows them to see as much as possible with such slender
resources. There is a further bonus that, as the two spot beams
lap over each other centre stage, this gives rather more
prominence to this important area.
 When more lanterns can be placed at the two side positions
this will not only increase the amount of light but improve
the general pattern of illumination. Because spotlights
basically give a symmetrical beam they become wasteful when
spread to cover a wide area. As the width of the beam is
increased so is the height, and this latter may not be needed
and, indeed, may have to be intercepted by a shutter or barn-
door flap.
 With three spots a side, the beams can be arranged as in
Fig. 120, which not only is going to make more efficient use
of the lamps but will allow lighting to be localized into areas,
left, centre, and right, each being illuminated by the requisite
two sources: 1 plus 4, 2 plus 5 and 3 plus 6 respectively.
 Continuing with the simple open-end stage, in other words
a rectangular platform at one end of the hall with an audience
on one side only, the next step is to consider increasing the
number of locations for the lights. For example, instead of
one vertical boom each side there might be two, one farther
from and one nearer to the stage, allowing another set of three
areas upstage. But in this form of theatre, both booms would
still be in the auditorium area rather than the stage area.
Another method would be to use two horizontal bars at differ-
ing heights on each wall instead of relying on vertical booms
(Fig. 121). The next step is to put a bar or two across the
ceiling (still in the audience area) for frontal lighting and to
help light to get over to the far sides in a very wide hall. Indeed
it may be that the overhead position is essential to get the
correct illumination.
 All these arrangements allow a great increase in flexibility

since lanterns can be positioned at a variety or angles and, if
for no other reason than that scenery(1) often obstructs beams
from some positions, such provision will be justified. However,
these extra positions demand greater skill in choosing which
to use, and if one is not careful the more lanterns and angles
that are employed the less the visual impact.
  (1)Contrary to popular belief, a lot of very solid scenery is often used on end
stages--open and ill-equipped to receive it though they may be.
 This paradox arises from the loss of the essential modelling
and contrast without which the brain cannot interpret the
picture on the retina. So long as heavy painting of both
scenery and of actors' faces was the rule, flat lighting from all
directions did not matter, since shadows and lining could not
be destroyed. Today real materials and little make-up are
the rule and this, coupled with too many lanterns from too
many angles, means that although the picture is many times
brighter, we see it less well; it becomes bleached out. There
is an exact parallel in sound. Speech in an echoing and too
resonant room becomes unintelligible no matter how loudly
the words are pronounced. Clarity is essential to lighting as
well as to speech and to see better it may be necessary to switch
something off rather than to add something.
 I think the open-end stage an essential beginning, before
all others, to training in the use of light. A proscenium provides
greater opportunities but too many temptations for the
beginner, whilst thrust stages, or arena stages, with a large
amount of audience encompassment, demand so many dis-
ciplines as to discourage him at the outset.
 Working on an open-end stage it will soon be discovered
what a small proportion of lighting is actually placed over
the acting area. This arises not because there is no provision
there (The Mermaid, London, and Phoenix, Leicester, both
have excellent arrangements for this) but because these posi-
tions are primarily not useful for lighting. To take an extreme
example, vertical light except for some deliberate distortion
effect is virtually useless. Stand a character under a vertical
light and the result is not only unbecoming but areas such as
the eye sockets, under the chin, and so on, are so shadowed as
to become invisible (Fig. 122). A hat brim may remove the
whole face and a bald head or blonde hair may, by presenting
excessive contrast, do the same thing.
 It is true that this extreme effect will only be experienced
in the centre of the beam, for when set to diverge even slightly
its edges will be at an angle and the result in these positions
may be somewhat less objectionable. However, for every one
of these positions there will be an equal number where the
light will come completely from behind. Vertical light is
something to shun on the whole, but there is one application
of stage lighting where it cannot be avoided. Concert musicians
insist on lights pointing straight down in order to avoid glare
in their eyes. The only solution is to make sure beams lap
over each other, then each lantern automatically provide
some angular correction from the edges of its beam in the area
of its neighbour.
 There is a further objection to overhead lighting on the
open-end stage in that many halls suitable for this form
particularly those of historic interest, are lofty, and lantern
hung on wires or frames would look singularly ugly. In saying
this, one is all too conscious of the variation in sizes and shapes
of halls, historic and otherwise.
 It is one thing to declare "hang nothing overhead" and advise the use of the wall
position only to find in act that the hall may be excessively wide. I can only say
that by choice of long-range spotlights it is possible to achieve the desired
angle positions more often than one would think. It is necessary
to add that objection to lighting hanging overhead in the field
of view applies mainly to the usual temporary mess. When it
comes to a designed frame containing the equipment within
a boundary, there may be positive visual advantages as was
discussed in Chapter 2, but as a lighting position it can be
overrated. The one thing that defeats side-wall positions or
their equivalent at the ends of an overhead cross-beam is the
acutely hipped roof, for this brings the lanterns too low. In
consequence, ugly shadows become visible on the stage and,
worse still, the actor masking by shadow referred to earlier
will be aggravated.

Vertical Lighting Angles

 Several references to lighting angles not too high or too low
have already been made in this chapter, and it is time to be
more precise. I am a little reluctant to quote exact figures
because when I got the ABTT to propound this as a require-
ment several experts became scornful and refused to be tied
down (Fig.123). What in fact I had been doing for years was
to declare that lanterns must hang at 45 degrees and, at the
same time, point my arm dramatically in the air at probably
35 degrees proclaiming "up there." This, possibly, when all
is said and done, is about as accurate as one need ever be.
When set out as in the ABTT diagram it suggests to people
incapable of reading drawings a degree of rigidity that in fact
will not be there. Even if every hanging position were sited
to give only the exact 45-degree centre line a considerable
variety of lesser angles will be available because we are working
in three dimensions. Angling a lantern from one side of the
stage to the other could bring the original 45-degree centre
line angle of the diagram effectively down to 35 degrees or
even less.
 The truth is that the 35-degree or the 45-degree angle
provides a useful and memorable formula--a beginning, to
prevent people from positioning lamps too near vertical or
too low near horizontal. An added advantage if we stick to
the 35-degree image is that it also applies to the plan view
as our key light diagrams showed. Overhead lighting positions
at these angles are a must, and this particularly shows in the
case of audience encompassed stages, but other angles are, of
course, desirable when possible. In fact, as the proscenium
stage is the only one which allows an unlimited variety of
angles, at any rate backstage, this may prejudice some lighting
designers in favor of it.
 The great need in considering lighting angles for illumina-
tion, for visibility as distinct from dramatic effect, is to avoid
the common trap of imagining that because a beam of light
strikes an actor he can be seen. The fallacy was pointed out
at the beginning of this chapter, but it cannot be repeated
too often. It can literally be the case that an actor cannot be
seen as lit in certain positions on the stage--not that light of
some sort will be unable to reach him, but that having reached
him it cannot present any intelligible message to the eyes of
the audience. In these circumstances, the producer must alter
his moves; there will be no alternative. He would do so,
one hopes, if he found one actor masking another by actually
standing in such a position as to present a physical barrier.
This being so, why allow a long scene to be played in pitch
dark except for a beam of light from off-stage picking out little
more than a man's left ear?
 Opera is the worst offender in this respect--yes, even in
Germany. There is too much token lighting--a shaft of light,
probably a stage bridge follow-spot--coming from a random
direction to proclaim: "Yes, I haven't forgotten. I know he
is there." Except for his voice, the singer might as well not be
there, for his face cannot be seen.
 One reason for this kind of thing in opera is a preoccupation
with gauzes and transformation effects. One overhead spot
per Norn just because they are to appear in some sort of limbo
is not sufficient for the twenty-five minutes' introduction to
Gotterdammerung. Nor is Hagen's prompt-side shoulder and
half Alberich's cheek good enough for their goings-on some
hours later, even though it is supposed to be moonlight. Of
course, where these latter two were actually positioned in that
particular production, right downstage with a permanent
gauze stretched immediately in front between them and the
audience, no other means but a pin spot from the nearby wing
was possible to keep dark the rest of the set that was to be
revealed later. But, Herr Direktor, these characters need not
have been so positioned--a few extra feet upstage would have
made all the difference.
 Never let it be thought that bad positioning for lighting
applies to opera alone. It behoves every producer to under-
stand sufficient about the behaviour of light and its relationship
to seeing to appreciate the need to alter a move or to shift
the whole scene a little more upstage. This does not mean
that the lighting designer should not try his best before making
such a request. Indeed, if he was on the job early, as he should
be (see Chapter 9), he might have been able to point out the
particular difficulty long before the show arrived in the theatre.
 With the above warning in mind, one has to consider lighting
for seeing first and then dramatize the scene with lighting for
the dominant effect, what the Americans call "motivation."
This does not mean that the first stage takes place in isolation
and, once completed, some thumping super lighting is produced
and poured on top of the rest. In positioning every light for
illumination, one should have the ultimate dominant idea in
mind to such good effect that the final motivation lighting is
almost unnecessary. This is a counsel of perfection, requiring
great skill. In the world of compromise that is stage lighting,
even if attained, it will all too likely become lost in the melee
later on.

Lighting the Acting Area

 Let us consider a specific treatment over an entire acting
area but divorced from the problems imposed by any particular
form of theatre. In addition, the audience is still regarded as
being one side only. In effect, every important area of action
requires the minimum two spotlights of the original portrait
lighting. To what extent this is strictly localized to one area
or shared with several or neighbouring areas will depend on
the importance of the action there; but one thing is certain--
any character, wherever he or she stands, should be lit from
the requisite two directions. Only thus can a suitably rounded
portrait be presented to the eye. However, knowledge of the
ultimate motivation can even at this basic stage begin to exert
an influence.
 To state an extreme case: if in one scene the dominant
effect is to be sunlight from the OP side and, in another scene,
moonlight from the P side, it might be logical to warm all the
spots angled OP to P side with No. 52 Pale Gold and chill
those angled P to OP with No. 17 Steel. This does not mean
that all the illumination on one half of the stage is warm and
on the other cold, but rather that there is a sense of direction
in the lighting for each of many areas. Now this effect can be
overdone and scorn can be poured on it.(1) Nevertheless I still
think that Stanley McCandless, who probably made the first
authoritative statement of the principle (as distinct from
inventing it), was right to do so. Like my 45-degree angle
referred to earlier, one is provided with a bendable formula
which in this case acts as a reminder that modelling is not
achieved by equally balancing the lighting on both sides of
an object; also, that, where daylight is supposed to be con-
cerned, the whole stage is under one influence even though
for practical or dramatic purposes it is divided into a number
of separate areas.
  (1)In theory the actor will be seen from the front as having one side warm, one
cool with the facial configurations shadowed in the opposite colour. Features
are modelled or made plastic and three-dimensional in appearance, and colour
differences echo natural light, which is half direct and half reflected light. In
fact, members of the audience seated close to one source will see an actor
primarily
in the colour tone of that source and those close to the other will see that colour
tone. Almost half the house, then, will see pink actors while the other half will
see blue ones, if those colours are used."--Donald Mullin in Tabs, Volume 23,
Number I.

 If later in the same scene one has to deal with an impression
of artificial light, for example, the supposed sources of light
being electric brackets on the two side walls, then the distinction
of No. 52 from the OP direction and No. 17 from the P direction
in each area would prove an embarrassment. This is because
all the separate areas no longer have to appear part of a whole;
they are each, in fact, under the influence of the local moti-
vating source. Cold light may be wholly absent; in conse-
quence, the effect of modelling will be more appropriately
obtained by putting a still warmer colour on the other angle,
say double No. 52 or a pinker colour such as No. 54úú This is
all very well in a standing set for a run with a limited range of
changes and plenty of lanterns. For most of us these theoretical
questions will not arise; it will be a case of one colour through-
out and use of dimmers to reduce the intensity to obtain the
modelling and distribution appropriate to the particular
motivation at that time. It is a question of accenting some
spots by dimming the others somewhat.
 Thus far, illumination has been considered with only half
an eye on a dominant effect. This latter subject will have
to be considered in the next chapter which describes how to
paint the stage with light. Suffice it to point out here that,
provided the illumination of the acting area does not quarrel
with it, motivation is easy and may largely be confined to the
scenery or propsú For example, if there is now a dark sky
outside the window the same spotlights which served as daylight
for the acting area, when the sun's rays shone through the
window, with but a slight shift of emphasis on the dimmers
become perfectly appropriate to this very different time of day.

Lighting in the Round

 Up to this point, lighting for perception has concerned end
staging with or without a proscenium but with the audience
always on one side. In staging with a large degree of encircle-
ment, lighting would appear to present difficulty. How can
modelling be indulged in if the actors have to be lit from the
four points of the compass equally at once? There is no doubt
severe restrictions are put on lighting in these open-stage forms.
However, there is one compensatory effect which results from
the fact that the distances are much shorter. As theatre-in-
the-round relies on its intimate effect, no one is farther away
in a good example than five or six rows. Obviously seeing is
going to be much easier than in theatre where twenty-eight
 rows Or more are not uncommon.
  The usual practice in theatre-in-the-round is to consider
 four lanterns as absolute minimum and then repeat this
 formula in four separate areas or six areas or more, depending
 on circumstances and funds available. Spots for special effects
would be additional to this. Stephen Joseph, who should
know, declares that three lanterns per area will suffice if
absolutely necessary, and Fig. 124 is based on one in his book,
Planning for New, Forms of Theatre. Whatever the number used,
the angle is going to be important to avoid glare and a stage
at the lowest level is essential when the aim is to keep the
audience as close as possible. If, however, the raised Guthrie
form of stage is used, then although the stage itself still continues
to be intimate in area it has to be separated from the audience
by a no-man's-land or moat in which the beams of light
terminate (Fig. 125)ú

Following
Some unkind words crop up here and there about the simplest
form of illumination--the follow spot. Just because this is
basically so simple, it is used without imagination. The main
idea is a lot of light from the direction of the audience to pin
on the star and then hang on, following at all costs. This,
when used in light entertainment on a variety act, a comedian
or singer, is the equivalent of a close-up. There is a real place
for these hard beams stabbing the darkness at the London
Palladium and a strange beauty in the pursuit of the skaters
by the discs of coloured light in an ice show. This kind of
thing is out of place in any kind of serious work but this does
not necessarily put the follow spot out of court altogether.
Sometimes it may literally be the only way of getting the actor,
singer, or dancer lit in that particular part of the stage or of
making them tell.
 Except that someone is moving a spot, this kind of following
has no relationship to the other. The beam must be kept
soft-edged and at relatively low power. The pencil beam is of
itself no more important than the beam of any other spotlight;
it is the result at the end of it--the subtle lift in light--that is
important. When such following is properly done the operator
must have a means of regulating intensity to hand, preferably
as part of the spotlight itself. He can then match his follow
intensity not to tell too much against the level of the area of
the stage he traverses and, in consequence, the spot will not
draw attention to itself. All this can be done, as also can the
Palladium type of following, with great artistry. If the skill
is not available, it is better to leave following out. A follow
spot blundering after the dancer as he leaps into the air or
wandering around looking for the target before opening up or
steadfastly aiming at the star's middle instead of his face is
no help at all. It is a not uncommon fault to centre the beam
on the chest, or even waist, instead of the face. As the beam is
at its brightest in the centre and the reflection factor of the
face may be much poorer than that of the clothing, expression
can come off fourth best in these circumstances.
 There is another use of manned spots which is still very
common in Eastern Europe and Italy, namely their use steady,
with only the slightest movement for correction when lit but
when each man's spot is dimmed out he promptly re-positions
it for its nest entry. I saw extremely skilful work of this kind
in the National Theatre, Prague, recently, which reminded
me how impressed I had been in 1934 with the six Stelmars
on the: perches in Covent Garden used in the same way. This
is very much a carry-over from an earlier skill derived from
necessity but not to be despised even now. Having said this,
I must in truth confess it would not suit me, for I have always
liked to be master of my own console, whatever form it has
taken, and moving spots are therefore out.

Optical Illusion
So far we have been considering how to help the eye to examine
a plastic three-dimensional shape and to light it to appear as
the solid it really is. It is now necessary to think in terms of
downright deceit. The weapon most easily used for this
purpose is perspective and its concomitant, painted shadow.
With certain remarkable exceptions, these two essentials of
the scene painter's art have been curiously neglected in recent
years. It is significant that amid all Sean Kenny's scenic
complications for Blitz! at the Adelphi the sudden appearance
of a perspective cloth of Victoria Station was able to capture
applause.
 Scenically, there seems no limit to the deception that can be
practised. I defy anyone, when it is properly designed and
carried out, to state what is painted and what is constructed.
Indeed, it may well be that, as the late Ernst Stern relates of
Max Reinhardt, tile wrong choice is made.(1) In Mister Roberts,
at the London Coliseum in 1950, the battleship was so well
painted, rivets and all, that the deception was complete even
when standing on stage. I had literally to go up to a hatch
cover and take hold before it betrayed itself as the framed
painted canvas it was.

  (1)My Life, My Stage, by Ernst Stern, Chapter 9 (Gollancz, 1951)

 Fortunately, My Fair Lady is a stage production which almost
everyone will have seen. It presented a remarkable exhibition
of perspective scene painting. In the main, only the Higgins'
study scene was built ill relief. Even then, a large part of it
was in fact foreshortened and was made to look much deeper
than it really was by perspective. The big Covent Garden
Market scene was created entirely by clever perspective cut-
outs. The apse in the "Why can't women----" dressing scene
was perfectly flat. Several times during rehearsal I had to
go on stage to convince myself tile stuff was really flat. Per-
spective enabled an apparently lavishly built scene to appear
like magic as Mrs. Higgins' conservatory near the end of the
show. Just drop several two-dimensional cut-outs in and add
real furniture and tile result was absolutely convincing. A
permanent example of this kind of shaping by paint and
perspective is to be seen any day behind the altar of St. Brides,
Fleet Street, and this is viewed in daylight.
 The Royal Opera House, Covent Garden, has some excellent
perspective sets, particularly for its Zeffirelli productions. Of
these, Fig. 126 of the first act of Tosca is a fine example. This
set consists of a mixture of built pieces, notably the altar and
the dominant pilaster, actors right of it. The rest where built,
for example the pews and ironwork screen, is constructed in
perspective greatly to increase the feeling of depth. The whole
scene from setting line to backcloth was only 42 ft deep--no
great depth for the Opera House. The backcloth itself hangs
parallel to the footlights--the appearance of sloping away is
entirely the effect of painted perspective. The archway over-
head is a simple cut cloth and both its appearance of solidity
 and of changing direction, of being bent through an angle,
are wholly the effect of painting.
 Of course, we all know some painted cloths which look just
like the painted cloths they are. Many, especially those painted
as ballet decor, are not intended to look like anything else,
but others set out to deceive and yet they fail. Why? There
is a number of reasons, of which the most important is the
treatment of the edges. These must not appear as a floppy
piece of cloth. Either the vertical edges of a painted cloth
must go right off stage or they must be framed and profiled so
that the mouldings appear in correct silhouette. Another
method is to run the unframed cloth behind a built piece. In
such cases great care must be taken that light does not show
the joint. Not only this, side lighting must be avoided on the
cloth otherwise the bowed folds running from top to bottom
towards the sides of the cloth will show. These are quite
invisible under other conditions of light but this fault spoilt
the effect of the library cloth for Love for Love at the National
Theatre when I was there. It may have been an accident in
positioning that night, but it was there.
 Another tell-tale area is where the perspective cloth meets
the floor. Not only may an actual line show there but the
change of direction is too sudden. Built Pieces foreshortened
and constructed partly in perspective are helpful in bridging
the gap between the reality of the actors and the world of paint.
A steeply ramped floor (Fig.127) or Steps (Fig. 126) are also
a great help. These two photographs should be compared
with Fig. 161 (page 294), Where the whole set was constructed
in three dimensions, film or television fashion.
 The next deception that has to be practised relates to appar-
ent size or scale. In a way, perspective painting is concerned
with increasing size, certainly depth, but there are other means
available. Scale cannot be judged without some measuring
rod, and in the theatre this is provided by the actor, and it
might seem that as he is always present little deception can be
practised. However, it so happens that to use him as a measure
of scale we need to know exactly where he is in relation to the
object whose size we are seeking, probably unconsciously, to
know.
 The principle involved is never to provide a complete frame
and this applies at the proscenium opening just as much as in
the case of any archways or doorways which have to look
larger than they really are. At the proscenium it is highly
desirable that there should be no complete frame, not only to
avoid the barrier effect but also to prevent an accurate assess-
ment of size. (There may be cases where a frame is deliberately
required, but these are outside this discussion as it can easily
be provided when necessary. It is abolition that is difficult.)
If the vertical sides and the top of proscenium opening never
actually join but merely cut one another and at the same time
the stage floor runs through without a joint to form an apron,
it is going to be extremely hard to judge the exact position of
our human measuring rod in this area.
 Another curious optical illusion arises from the fact that any
line which is not actually seen to terminate seems to be auto-
matically interpreted as continuing when it goes out of sight.
This is an effect well known to both film and television
producers where judicious cutting of the edges of the picture
is all-important. A fireplace, part of a wall and a door becomes
a whole room; the brain automatically assumes the rest is
there. No effort is required to make this assumption; the
effort comes if one wants to remind oneself that there is little
beyond the edges of the picture but studio impedimenta,
cameras and lights.
 It is the use of lighting that can help us to achieve somewhat
similar results to the cameraman's choice of shot. "Stage
lighting must conceal more than it reveals," I wrote in 1948
for my first book, and this has become rather a trite observation
through over-use since. However, it still exactly conveys the
main force behind the use of spotlighting, the ability to put
light just where required and thereby draw the eye away from
something. Under certain circumstances, one may be quite
unable to see beyond a pool of light. Without the contrast of
light there will be no darkness.
 Perception of intensity and, as was described elsewhere,
colour is absolutely dependent on contrast. The eye adapts
itself over a period of minutes to accept new levels of light but
cannot do this when both are presented simultaneously. If a
scene is to appear really bright, it is no good piling on more
and more light in the entire field of view. The true over-
whelming of the eyes we get in nature requires thousands of
foot-candles where in the theatre we deal in tens. Suitable
contrast can overcome this difficulty and, as a beginning, there
is the darkened auditorium. The design and colouring of an
auditorium so that it will appear dark is obviously important.
An extreme use of contrast in the theatre are "blinders," a
row or cluster of low-power lamps facing the audience, beyond
which what is happening cannot be seen. This substitute for
a curtain might have its moments in open staging.
 Today's "island" sets, without a frame on a wide stage, as
for example in Fig. 128, may bring an unwanted isolation.
Irving Wardle, describing John Gabriel Borkman at the Lyttelton,
says ". . . the indoor scenes are played on a central pocket
handkerchief acting area which reduces their scale and pushes
them away into the distance."(l) I can confirm this as seen both
from the front row of the stalls and from the back rows.
 This effect is the result of the use of black in the audi-
ence's field of view. There is a common fallacy that to paint
an object or some area black is to make it vanish. It is true that
there will be little if any significant light reflected back to the
eye but the interpretation will be to see a hole, gap or chasm.
The more perfect the black the more the effect is aggravated.
Of course there are times, as in the Royal Shakespeare Com-
pany's Macbeth at the Aldwych theatre, 1975, when to play in a
limbo created by unlit black drapes gives a fabulous detach-
ment from dependence on scenery for time and place. In a
normal scenic context, however, to frame in with black legs
and/or black flooring is to detach the show visually from its
audience. This can even happen on a thrust stage. At the
Crucible, Sheffield, they have found, for instance, that brown
serves the purpose much better. Keith Green, for some time
technical director there, points out: "Without an architectural
feature similar to Stratford Ontario the designer at the Crucible
has the problem of how to cope with the gap up-stage. He
can either fill it up as if it were some kind of remote proscenium
--which is not an answer--or effect a "blend" with the
auditorium wall on either side to achieve a union of set and
building. In my experience this is rarely achieved. However,
with all vertical surfaces of the auditorium painted the same
dark brown one is not unhappily aware during performances
of this problem or of a point when the auditorium finishes and
the stage begins. Scenic elements are usually masked behind
by a kit of dark brown felt-covered flats. The choice of this
colour is admirable and infinitely preferable to black."(2) An

 (1) The Times, March 12, 1976
 (2) Sightline, Autumn 1975

extreme example of the lengths one has to go to, so to speak,
is, the way the mobile towers in Christ's Hospital Horsham
(Fig. 6), painted a strong red--of all colours-can remain
neutral because that is exactly what that particular auditorium
leads you to expect. However, red happens to have been the
undoing of the excellent Guthrie-type stage in the Octagon
theatre, Perth, Western Australia. Unlike the Crucible there
is actually a proscenium behind the thrust and all would be
well were it not for the fact that they have been landed with a
fine set of red plush tabs, simply because they were second-hand
and going cheap! These tabs unfairly rivet the gaze and
declare "proscenium" where otherwise it would not be noticed.
 In conclusion, another remarkable optical illusion-the
cyclorama--should not be forgotten. At the moment, this
device, like the use of house tabs, seems rather under a cloud,
but this does not mean it might not return. The trouble is
that so few people have seen a real cyclorama. The essential
is a perfectly smooth surface designed to reflect the maximum
light. It need not be a fixed affair of plaster, as I pointed out
in Chapter 2; cloth will do, but not any old cloth as most
theatre people seem to think. Provided the edges are masked,
the effect, properly lit, is of limitless space: there is nothing
by which the eye can judge distance. This is a true optical
illusion, every bit as convincing as good perspective painting
in its field and just as right and proper to the theatre.