Alla Prima painting process, my way

Here It can vary a lot , depending on the finish sought after , despite of that , I believe the the most important thing in a painting Is the START.

On the start there Isn’t too much to fuss about, It can’t vary too much as the finish , because It must be a systematic and logic way to put things on the canvas, and It must be done in a way that constitutes the skeleton for the work to held later on, so It must be also accurate.


Drawing with the brush, gesturely to capture motion, flow, and angular where it needs to be agular .straight lines as block-in(Bargue style), triangulation, thinned burnt umber. It must be exact in describing angles, masses, proportion but not too detailed or beauty, reserve all the effects for later, features indicated(just two dashes is enough) just to estimate correctness of the drawing of the skull, this drawing will be lost soon. Paint the shadows of the face with burnt umber in order to have a picture like a notan ( 2 Values) Fusing everything together. In this stage It’s crucial to use drawing also to find the correct placement of the subject on the canvas (this should be the very first thing to look after as you begin a painting. Try to spend a good amount of effort in placing the subject in the correct portion of the canvas, where it feels in balance. There Isn’t a really a rule, let your taste guide you). Work on that drawing and placement until It satisfies you, If you make a mistake erase with a rag and some mineral spirits, The drawing must be deadly accurate in describing big masses relationships, but It Isn’t important to carry this block in any further, detailing It. The accurateness in proportion of the masses and placement is what matters. When the actual colors will be layed in, The drawing will be corrected by the continuous manipulations of boundaries of background and flesh (this isn’t an excuse for sloppy drawing though, at least for me. If you have strong drawing skills you might begin with something more loose, but I think that it’s better to fight with the drawing at the beginning, because afterwards, slightly changes in the drawing will always occur since the manipulation of edges lead always to some sort of drawing restating). The drawing may also be approached with burnt umber a little bit stiffer, thus, when erasing with a clean sheet of paper a blunt trace will remain on the canvas allowing you to work again on that structure, refining the drawing . The drawing can also be done with a stick of vine charcoal  and then fixed later with a spray diffuser. This has the advantage to let you correct the placement of the subject as many times as you want, without dirtying up the canvas too much with paint, if paint would be employed instead to make the drawing. It’s totally up to you, and It will not yeld to no difference in the result, because much of the drawing will be done with the brush when the serious part of painting begins, no matter how much detailed or accurate the drawing in the beginning was, while painting it will be difficult as always, and very much of the drawing will be redone for sure. Furthermore, it is very like that a too detailed drawing might refrain you to take risks and to paint boldly since you will be afraid to lose the drawing. That’s the reason why I think that the initial drawing should concern only big masses. The initial drawing is useful only when it describes masses and nothing more. However It isn’t necessary to use a straight blockin approach, in fact I think that although it can provides fixed points where ones can accurately measure and judge proportions thanks to such straight lines, it can stiffen up and slow down the process a little and a loss in gesture may occur. Meanwhile,  a drawing done with long sweeping lines, is easily spotted since It has a better gesture, and a better connection and flow among the masses since everything is rhythmic and better related to one another, and the canvas is covered much quicker and the impression held and expressed more directly and faster, this approach, I think, can also allow you to explore faster the possibility to caricaturize the subject.

Color Block in

The second step might be the indication of a color block in, local color of each mass, scrubbed with the side of a big bristle brush. This way of putting down paint allows you to paint with the least amount of paint possible that will achieve a good coverage. Mix on the palette with a palette knife a middle tone for the BACKGROUND, another one for the MASS OF THE HAIR, another one for the MIDDLE TONE OF THE WHOLE FACE ( the general color that gives the impression and the character of the sitter, next in value to the high lights , not to confuse with half tones which are the transition values sandwiched between middle tone and the darks) and a general tone for the SHADOW SIDE OF THE FACE. These mixtures must be checked up against the model with the palette knife held with the same slope as the canvas, so that afterwards, the color checked on the palette knife will look the same on the canvas, since it receive the same amount of light. Or a color checker might be employed. You can easily create one by printing and plastifing a grey page, and by cutting a hole in It afterwards. This will isolate the color on a neautral ground, and facilitate the judgement on hue, value and chroma.

*how thick should the middle tone of the face be? Sargent advised to paint all the general passages quite thick, but how much? will that compromise the superimposition of the little color planes later? Or It will help in creating soft edges? I need to Experiment.*I think that in this local color block in, and in every start in general, is better to use thin color, since It is easier to control and achieve soft edges, thus creating the unity we require in the beginning to judge if we captured the whole impression or not. Afterwards we can begin laying in the little color planes. I think that the scrubbed color block in purpose Is just to eliminate the white of the canvas in a “smart” way, in zones, that will also aid the colors to be put upon later in the first painting, so it should be treated with importance.


I am always thinking of why should I do a thing or another, so I started thinking of why do I need to put a thin scrubbed color block in instead of painting directly with thick color planes. I try to answer to that now, but I know that this answer might change as time goes by. If we take into consideration the philosophy of general to specific,  a color block in, is the vision of the scene with the widest breadth of vision, the biggest and simplest, obvious shapes of color possible,  and , since that’s how we perceive  a scene from a distance, by big masses, our color block in should resemble the scene in colors and values very accurately, but with no details , as when we do color studies in small thumbnails  of other paintings. That’s what we must  put on the canvas, and since this is the biggest truth of all, there’s no reason to paint it with thick paint, it’s just a road map of values, a structure that indicate the range in value of the painting, where the details can be painted upon, later on, without going “off road”.

What’s the reason behind little color planes? I think that color planes exist to get the inner shapes, so it is a logic step after the big color blockin. Like a sculptor, we establish the big planes first of all, and then we chisel the and refine all the features and the forms progressively.  Color planes are thick not to be fancy ( I don’t think that sargent or other bravura painters used thick color strokes to be cool with friends) I think that the thickness of the color chips is due to permanency purpose, because oil paints will tend to be slightly transparent overtime, and the second reason Is that only thick paint can let you be deadly sure about the color you mixed on the palette, because it will have body, fully opaque. If I use an opaque color but I skimp on paint quantity, the color will not be as strong and and as true as I mixed It, and it will look different according to the background upon it is painted. For istance, gray scrubbed on a white ground Is warmer than a gray scrubbed on a dark ground.

Should this middle tone be of just a color? Or it should change according to facial zones? ( I can always be under the same value, even though I add some color variations). *I think that Is better to start in the most simple way, because color can also be scrubbed later on, the most important thing is to capture the true values and the general color and leave all the subtleties for later, in the refinement/scrubbing stage .

Should I superimpose the darks on this wet base spread all over or should I paint first the darks and then lay in the middle tone in order to not compromise the cleanness of the darks? *painting one thing into another always helps to achieve soft edges, so I think the only difference lies there, but I don’t think it is a big deal to do a way or the other way as long as you keep in mind to have all the edges soft until the end.

I try to answer that according to my last experience:

I’ve noticed that Sargent didn’t paint always in the same manner, that’s because of the time restraints(1 hour, 3 hours) and intentions(sketch,study).

Sometimes  He seemed to scrub a general color for each mass, light on the light side and shadowed on the shadow side, and on that he begun tiling with thick color chips. I figured out that these chips in order to retain the passage of the bristles of the brush must be quite thick, the brush must be well load with paint, and these chips must be put on the face to depict in a sculptural, mosaic way following facial planes all the time, and chasing after every little inner form that you see. The scrubbing of the thin color create first an arena of value that dictates the relationships between the masses and It is layed in without caring the inner forms , but thinking about the biggest planes relationships, the ones where everything may be superimposed  after with tiling. The general middle tone can be scrubbed with the side of the brush  just to be a sort of colored blockin underpainting where afterwards thick color chips will be layed upon, or can be spread with thick paint  since the beginning following facial planes, and upon that lay then the missing colours and values.

Use a limited color palette at the first as the 1 hour sketch of sargent shows.


Inner forms

The third step might be tiling of the whole face with thick color chips chasing after all the inner shapes, wood head man cut with an hatchet, the aim Is the amount of inner shapes and good amount of light ( recommending reading contemporary review, collier on Duran and other stuff). No blending obtained by brushing things together*I have noticed that It obliterates all the juicy brushwork that suggest facial planes, so I think that I should not blend the chips. Blending being allowed only for edges, mix every graduation. These chips should be in size more or less of that of an iris of the sitter, according to the dimension we are endeavor to paint, in order to capture all the inner shapes.

*my last question Is, how long this thing of color chips should continue ?I mean, after the 1st sitting’s done, should the 2nd painting be approached with thick color chips too? If the first painting came out more or less well, why obliterating again everything? Isn’t It better to continue with scrubbing drybrushing etc… using the work of the first painting and further refine it? I guess that the answer to this Is just If we reached the realism we want to put in our artwork or not. If the painting came out perfectly in the first or second sitting, and every part of It is true to life, there is no need to start over again. of curse, if we have to change something very wrong, we need a thick brushstroke insted, to cover the previous work.


At that point Sargent left us a lot of artworks left in that stage and this can be easily be the end of the first sitting, and the routes that I can follow can be vary, according to my taste. If I like to have a flawless surface, I might blend the color chips ( to make the transition from one another gradual) and add some more. Color chips/blending/restating /repeat until the piece Is finished. I can leave the painting just as patchy as It Is and soft all the hard edges once dry with some scumbling and dry brush( I don’t think that in the past they did that on purpose, I think that, if we found many paintings done that way, It was just because they were hurry to leave the painting place, maybe because the sitter got things to do, or maybe just because the painter forgot to soft an edge and he tried to soft afterwards once the paint was dry), or I can scrape down all the work I did in the sitting, soft everything with a brush in order to have no hard edges whatsoever  and use that as a suggestion for the next sittings ( I think this Is what sargent used to do in his early years, this Is done in order to retain the texture of the canvas and to make a painting that has not too much build up of paint, and also to control edges more properly), miss heynemen seems to describe this process in her writings on Mr Sargent:

“Should the painting not be satisfactory, the whole is ruthlessly fogged by brushing together, the object being not to allow any parts well done, to interfere with that principle of oneness, or unity of every part; the brushing together engendered an appetite to attack the problem afresh at every sitting each attempt resulting in a more complete visualization in the mind. The process is repeated until the canvas is complete.”


Solomon J. Solomon describe this Scraping method very extensively In his book.

It seems to be a common practice during late 1800. It was basically done in order to retain the texture of the canvas till the end, and to obtain a smooth surface quality and control over the edges, that were kept soft until the very end, because It is of common opinion that is hard to soft an hard edge, but relatively easy to sharp a soft one.

Notes on Lighting of the model

The lighting of a face reveals or subdue the inner shapes emphasizing big basses or scattering shadows all over destroying that strong light and shadow pattern. Sargent and Velasquez used a particular type of lighting called Loop Lighting. This kind of light reveals only big masses, resulting in a more pictorial composition since every value can be perfectly readable and Is perfectly stated.


Tell your friends!

Leave a Reply

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