The Tin That Reads as a Circle: Correcting a Timed Observational Drawing
The paper is A2 cartridge and the candidate has 180 minutes; 40 have already gone. The still-life group stands on a table whose surface is 15 cm below the candidate's eye level: a cylindrical biscuit tin (diameter 12 cm, height 18 cm), a clear glass bottle with a tapering neck, a plaster sphere resting on a small block, and a linen cloth falling over the front edge of the table. The only directional light is a window on the left, but the room's overhead fluorescent tubes are also switched on. In the drawing so far, the tin's top rim is drawn as an almost complete circle; the bottle's neck and body tilt in opposite directions; the cloth is described with outlines only and no interior planes; and the cast shadow under the tin is a single flat, evenly dark shape filling the whole footprint. The candidate says the drawing 'looks wrong' but cannot say where, and wants to start erasing and redrawing the tin from scratch.
1. Diagnose the ellipse problem and give the sighting routine the candidate should run before erasing anything.
Read the worked answer
The rim reads as a circle because the candidate is drawing what they know about tins rather than what the viewpoint produces. With the eye 15 cm above the table, the top rim is seen only slightly from above, so its ellipse must be compressed - the minor axis is roughly a third to a quarter of the major axis, not close to equal. The bottom rim sits further below eye level, so it is seen more from above and its ellipse is more open than the top rim's; a drawing in which both rims match is automatically wrong.
Before erasing, the candidate should run four checks, each answering a separate question:
- Proportion. At arm's length, hold the pencil vertically and mark the apparent width of the tin; rotate it 90° and mark the apparent height of the top ellipse. Compare the two marks as a ratio and transfer that ratio, not a guessed curve.
- Axis. Drop a plumb line from the top rim's centre and confirm the bottom rim's centre sits on it. Tilted ellipses usually mean a tilted axis, not a badly drawn curve.
- Symmetry of the ellipse. Check that the left and right halves of the curve mirror each other; a lopsided ellipse is the commonest tell of drawing the two arcs separately.
- Negative space. Look at the shape between the tin's rim and the table edge behind it; if that gap changes when the ellipse is adjusted, the ellipse is still moving.
Only after these checks should the candidate redraw - and then by correcting the axis and ratio, not by erasing the whole object, which would destroy the correctly placed verticals and waste the remaining time.
2. The candidate has 140 minutes left. Set out an order of operations and justify why detail must not come first.
Read the worked answer
The remaining time should be allocated to stages that get progressively smaller in scale: 20 minutes to re-establish the overall envelope and axis lines of all four objects and the cloth's main breaks; 35 minutes to place the shadow shapes as one connected value pattern; 45 minutes to model the three largest forms (tin, bottle, sphere) to about three values each; 30 minutes on the cloth's plane changes; the last 10 minutes to stand back, check alignments and restore only the sharpest edges. Hard detail - the bottle's highlight, the tin's label edge - must be held back deliberately.
The justification is structural. In a 180-minute timed drawing, everything drawn early is drawn at a large scale, so an error made in the first 20 minutes costs far more than an error made in the last 20. If the candidate starts with detail, the drawing gets a convincing eye-catching passage locked onto a wrong proportion, and correcting it later means erasing finished work, which candidates almost never do. Working large-to-small also keeps every decision testable against the previous one: the sphere's size can be checked against the tin only if both are still open shapes.
A second reason is that value structure, not detail, carries the read. The examiner will see the drawing at three metres before seeing it at arm's length. The cloth and the cast shadows - the largest areas - must therefore be resolved while there is still time to change them, which is exactly the reverse of the instinct to polish the glass bottle first.
3. Rework the cloth and the cast shadow. Explain what is wrong with the present treatment and what should replace it.
Read the worked answer
Two errors are present, and they have different causes.
The cloth. Outlines describe a silhouette, not a surface. Linen folds are planes; each fold has a lit plane, a plane turning away, and a narrow dark where it tucks under, so the candidate should look for the fold's ridge line first, then describe the value of the planes on either side of it. The number of folds should be reduced to three or four major ones - a common fault in timed work is putting in every small crease because each one is easy to see, which flattens the big form. The cloth also has a colour-and-value relationship to the table it lies on, and both are pale, so the cloth's near edge is best separated by a slightly cooler, darker passage rather than by a drawn line.
The cast shadow. A single flat dark is wrong on three counts. First, the window is an extended light source, so the shadow has a soft penumbra at its edges and a darker core nearer the tin. Second, the overhead fluorescents are a second source and will lift parts of the shadow, especially where it meets the object; that lift is diagnostic of the room, not a mistake. Third, no shadow in a lit room is uniformly dark, because light bounces off the table back into it - the shadow's darkest zone sits at the contact with the tin (the occlusion), and it lightens as it travels outward.
The replacement is therefore: a soft-edged, non-uniform shadow, darkest at the contact line, opening up at the far edge, with a reflected-light passage on the side nearest the viewer.