Watercolour and transparency

the watercolour conscience

For whatever reason, watercolour seems to attract more than its fair share of purist ideas and rules. We are told: ‘always use a fully-loaded brush’; ‘don’t use black’; ‘don’t use white’; ‘don’t use resists’; ‘don’t use opaque colour’; ‘don’t use more than one/two washes’; ‘use only the most transparent pigments’; and so on. While  there may be merit in a lot of this advice, when it’s presented as ‘never’ or ‘always’ statements, it can become something of a straitjacket for the beginner if he takes it too literally. It leads to a sort of watercolour conscience where we experience feelings of guilt if we fail to live up to the purist dogma – feelings which even someone of the stature of Rowland Hilder admitted to experiencing. It was only after studying the watercolours of Turner that Hilder felt comfortable with his ‘impure’ technique and realised that using body colour and Indian ink was not a violation of the medium. The idea that there is a right and a wrong way to paint in watercolour tends to be reinforced by the instructional style of some of its teachers: people like Edgar Whitney and Edward Wesson, both of whom were forceful communicators with strong opinions on how watercolour should be handled. Their facility with the medium could be both inspiring and daunting at the same time to the beginner.

To illustrate the problem, I’d like to look at one quality of watercolour – perhaps its defining quality – which has purist associations: transparency.

Every student of watercolour is reminded that the charm and beauty of the medium is down to the transparency of its paints which allow light to reflect from the paper’s surface to create the luminosity and delicacy of colour that characterises the medium. So, not surprisingly, a lot of instruction focuses on exploiting transparency, for example, through wash technique, glazing and choice of pigments.


transparency rules 

But what happens when you take this idea to the extreme? What happens if you feel you must always work transparently?

I fell into this trap (yes, it is a trap) during the early days of working with watercolour. I was exploring glazing technique (which is simply superimposing a wash of one colour over another) using a number of popular paints from the Winsor & Newton range including cobalt blue, French ultramarine, burnt sienna, burnt umber, yellow ochre, alizarin red and three phthalocyanines: Winsor blue, Winsor red and Winsor green. Naively, I expected to create glass-clear passages of jewel-like, vibrant colour, as well as discovering a whole range of subtle grey mixtures that I could use in my paintings. The results, however, told a different story. Among the many glazing permutations that I tried, very few gave satisfactory results in terms of colour vibrancy or transparency. In fact, as I remember it, only cobalt blue worked well as a glazing colour and didn’t obscure or muddy the underlying washes.

So what went wrong? After all, watercolour paints in dilution are supposed to be transparent so why were most of the results disappointing? It turns out, as is so often the case with watercolour, that there is no simple answer. In truth, my experiments weren’t very well controlled: there were too many variables involved. Take the paints themselves; they vary considerably in their inherent transparency. For example, the cadmiums, yellow ochre, light red, cerulean blue are all relatively opaque even in dilution. And colours such as French ultramarine and manganese blue have a strong tendency to granulate in wet washes which affects transparency and optical mixing with the underpainting. Then there are the staining colours: alizarin crimson, Prussian blue and the Winsor series fall into this group. Because staining colours are highly transparent in dilution and have a high tinting strength (a little goes a long way), you would expect them to be ideal as glazes; but, in practice, they often produce dead, muddy colour.

In addition, other factors that I didn’t fully appreciate at the time also played a part: in particular, the choice of paper, the brand of paint, and the wetness of the paint application. All of which leads to the conclusion that transparency is not a given in watercolour; it takes a pretty thorough understanding of materials and technique to be able to exploit it.


dark matters

Another difficulty for the beginner who tries to work transparently is the problem of how to paint the darkest values. I struggled with this for a long time. Even using low value paints with high tinting strength such as Winsor blue and Prussian blue I couldn’t seem to produce satisfying dark values by applying them as transparent washes. A wash would look dark at first but then lighten as it dried. Repeated glazing would eventually create a dark value but somehow it would lack the impact or finality I was hoping for. I assumed the problem was down to bad technique or poor choice of materials, so I persevered.

Eventually, by chance, I did find a solution; but only at the expense of the transparent approach. My answer was to use strong mixtures of relatively opaque paints and to apply them as a single wash. I found that mixtures of French Ultramarine and Light Red (both Winsor & Newton) produced a rich, velvety dark that suited my purposes. Thirty years on I still prefer this combination.

The question is, have I abandoned proper transparent watercolour technique? Have I ‘cheated’, like Rowland Hilder using Indian ink for his darks?

I’ve always felt there is something fundamentally wrong with the idea of using multiple glazes to create darks. What is the difference between a series of glazes and a single, heavy application of paint? I know the usual response will be that glazing will give a more ‘luminous’ result. But does it? To get the same dark value by glazing as you would with a single application means you must have the same concentration of pigment particles on (and in) the paper surface. And doing it by glazes has several disadvantages.

First, the more you disturb the paper surface the more absorbent it becomes because the sizing is washed off and this will allow paint to penetrate further which will tend to result in dull, uneven colour.

Second, there is the problem of registration – ensuring that successive glazes line up. If they don’t then edge quality may suffer (for example, if you want a crisp dark edge next to a very light area). Achieving registration tends to make brushwork tight and lacking in spontaneity.

Third, thin washes dry lighter. This is simply down to the degree of dilution of the paint: the more dilute the paint the lighter it will dry. You always seem to be one glaze away from achieving the intensity that you need. This is what I mean by finality and it explains the effectiveness of using Indian ink with watercolour (line and wash). Indian ink is not only a fixed value (black), it is also waterproof when dry so subsequent watercolour washes won’t disturb or weaken it.

Finally, glazing encourages monotonous, hard edges; and hard-edged, uniform, dark values often don’t sit well in a painting: they can create ‘holes’ and flatten forms. Dark values ‘fit’ much better when they are allowed to bleed into adjacent areas. This is very easy to do with a single application of paint either by working wet-in-wet or by softening edges with a damp brush.


opacity rules? 

So what is the message in all of this? What should we make of transparency?

In a sense, transparency does not exist! It’s a negative quality – an absence of opacity. And thinking in terms of opacity rather than transparency does have some advantages. First of all, it prepares us for the reality that all watercolour paints are more or less opaque – which at the very least may lessen the disappointment of glazing experiments. But it should also remind us that much of the beauty and character of watercolour can only be achieved with relatively opaque paints – granulation, for example, or vibrant wet-in-wet passages of saturated colour. Overemphasising transparency robs us of the textural possibilities of watercolour which allow us to suggest the roughness of stonework or timber; the foliage of distant trees; or the delicate atmospheric veil in a landscape or seascape.

In the final analysis, whether your technique leans towards transparency or opacity is largely a matter of preference – there is no right or wrong way. Subject matter may play a part. For example, there is a vogue for photorealistic still life paintings of flowers and highly reflective materials like cut glass and polished silverware. These are usually executed with a highly controlled glazing technique: lots of hard edges, very little mixing of colours on the paper. It’s a technique that’s well-suited to studio painting. In contrast, the en plein air watercolour paintings of Edward Seago and followers such as Edward Wesson and John Yardley show a more opaque, direct technique, analogous to the alla prima method of oil painting and much better suited to producing work outdoors often in challenging conditions. They work in a drier manner and achieve strong darks with single applications. Alvaro Castagnet and Joseph Zbukvic, who tackle similar subject matter (landscapes and cityscapes usually painted on site), also adopt a drier, opaque style with (often heavy) applications of bodycolour. It’s interesting to study the works of your favourite watercolourists to see how they  handle purity and transparency in the medium.

The real beauty of watercolour, for me, does not lie in transparency but in the contrast between transparent and opaque effects. Compositionally, contrasts are fundamental to achieving variety and focus in a painting and opacity/transparency is a useful textural contrast to explore.

One thought on “Watercolour and transparency

  1. great text. I’m a beginner so I find myself questioning my approach. but still there is something addictive in researching this purist approach

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.