(Updated 2-27-20) Fifteen years ago, after having painted for fifteen years, it began to seem that something was missing from the craft I had learned via 20th century painting professors such as Doerner, Wehlte, Gottsegen, and Mayer. This sense of not just a 'lost secret,' but a lost dimension, slowly but surely became a research project into the methods and materials of older painting, based on the available older texts on the one hand, and the findings of contemporary technical art history on the other. The full results of this research are available here in book form.
A condensed overview of the material in the book is below.
See also the video on how to make the fused damar and beeswax medium.
A quality painting process develops through sound, archival practice. This involves creating a multilayer structure - the painting - which holds up well over time. There are many variables involved in how this structure can be made. Because this structure is inherently complex, it's always best to attempt keep both it and its various components as simple as possible. However, this element of simplicity does not necessarily extend to purchasing ready-made materials if the hope or expectation is to create higher quality work: generic materials have a strong tendency to produce generic work. Investing in developing a relationship with the materials creates a much greater understanding of their working characteristics than any amount of work that begins and ends with branded bottles, tubs, and tubes, no matter what their supposed pedigree. Experience cannot be superseded by anything as the most reliable teacher. Quality art begins with quality time. Quality time begins with a quality process. A quality process begins with quality materials. But what constitutes quality materials? Here things get both personal, and complicated!
The craft, like the behavior of the light it represents, has a logic that is helpful and, if the work is to last, incontrovertible.
Relative humidity, unsupported canvas, and unquestioned acceptance of second rate commercial materials are the largest problems oil paintings face in the long run.
The ground is an area often ignored. If the ground is not at least slightly grippy or absorbent, long term issues with adhesion can easily result, especially on unsupported canvas.
The use of solvent became liberal in mainstream 20th century technique, often to the point of being incautious. Long term exposure to any solvent is likely to cause health problems, and solvent is only necessary in oil painting to work with resins such as damar or copal. Otherwise, brushes can be wiped clean on a rag, then cleaned more thoroughly with oil. Brushes can also be kept on their sides in a tray in a slow drying oil such as walnut or safflower. When used, solvents need to be protected from light and oxygen to avoid oxidation that leads to yellowing.
All resins yellow over time due to oxidation. It is easy to prove this to yourself with yellowing tests. The global use of resin as a medium component begins in general in the 19th century. Resin has not been shown to be the "magic trick" of any major painter in the 15th to 17th century. Resins can be used in small amounts without issue, especially on panels and with warmer colour schemes, but are best used to alter the rheology of the paint, not create saturation. Creating saturation is best left to the oil. The least yellowing resins in my experience are damar and larch balsam.
Specific materials have contributed to the paint film remaining bright over time rather than becoming deeper in value. The original material used for this on panels was egg yolk. Resins also work for this, but are best used in small amounts. Beeswax begins to be used in the 18th century and small amounts (2 to 5 percent) of wax are helpful for maintaining brightness and flexibility.
Alkyd "resins" are actually modified vegetable oils manufactured on a large scale. No alkyd dries quickly without an added metallic drier. The thicker the alkyd, the more drier it needs. Driers have been consistently shown to lead to darkening and a brittle paint film over time.
A cold-pressed oil yellows less than an alkali-refined oil. A refined cold-pressed oil, handmade or commercial, can be made non-yellowing. However, because all oils used by manufacturers are first commercially refined to remove the free fatty acids that cause rancidity through oxidation in an edible or cooking oil, these oils will never dry quickly (unless driers are added) compared to an oil processed specifically for painting. There are no oils on the market to my knowledge that have been processed specifically for fine art painting.
The simplest way for painters to access high quality oil economically is through the organic, nutritional grade oils. This is especially true for cold-pressed linseed (flax) oil, which tends to be dramatically overpriced when marketed for artists.
Through hand refining, linseed oil can also be made to dry quickly without lead. But this may or may not be relevant to a given technique. For example, a smooth surface approach does not need thus type of oil, whereas a more bravura or broken surface approach is enabled by it. Linseed oil can also be purified by hand to dry moderately.
Traditional mildly alkaline additives such as chalk, marble dust, and bone ash contribute to non-yellowing behavior in the medium.
A leaded oil must be preheated before using lead, the lead used in minimum amounts with minimum or no heat, the oil then aged in the light to be genuinely non-yellowing on drying.
Synthetic varnishes have improved to the point where they are being used in major national collections. Regalrez is inexpensive and marketed commercially, MS2A is expensive but has a less plastic look. However, in the past, all synthetics have been shown to encounter issues over time, and one key feature of these materials is their ease of removal. In the De Mayerne Manuscript, the opinion of Rubens is recorded that the best varnish is oil thickened on litharge. If the painter has access to oil that is non-yellowing and quick-drying, the lead can be dispensed with, a thicker oil applied thinned slightly with a non-yellowing solvent such as deodorized mineral spirits.
the most important component
...is the oil. No modern manufacturer, large or small, refines their own oil. The oils used by current manufacturers are refined according to standards developed for the food industry. The goal of this refining is to minimize oxidation that can cause rancidity. As such, without added driers, it is not possible to get a modern oil to dry quickly. This is not the case when beginning with an unrefined oil, because, in this case, without using lead or other metals, the potential for oxidation can be enhanced by a traditional refining method. Does this mean modern oil is "bad," or "wrong?" No, but there are some things it does better than others, and, from the perspective of older practice, some things it cannot do.
If you want to explore the foundation of the older process, try refining a liter or two of organic, cold-pressed linseed oil. Especially if you are an experienced painter, working with this oil leads to a logical explanation of how older painting evolved the way it did. Brands I've had experience with in America are Flora and Jarrow, there are others as well, all much less expensive in quarts online. The Allback cold-pressed organic linseed oil from Sweden is also very good, but the bulk organic linseed oil from Jedwards or Azure Standard is newer, probably more pure as it is edible, and will process more easily. The availability of oil of this quality is unprecedented in the history of oil painting, however, it must be refined to remove the impurities, and alter the structure of the fatty acids. The oil you create will dry quickly compared to commercial linseed oil. It will not yellow and have working qualities that are far better than any commercially processed oil, regardless of reputation, and at an affordable price. Photo above is of the Jarrow oil after a wash of one day.
A complete pdf file about the process of refining linseed oil is here. This is all text, no photos, for printing out. There are methods for four traditional processes, but the featured process is a faster adaptation of the water-sand-salt method in the De Mayerne Manuscript (early 17th Century) which Eastlake recommended in the first volume of his Methods and Materials (early 19th Century). This time honored process not only removes the impurities from the oil, but alters the fatty acid structure to polymerize faster and be less yellowing.
If you make this oil, then make paint or even just a medium with it, you will be in a position to understand the most fundamental "lost secret" of older painting practice: the significant rheological qualities inherent in hand-refined linseed oil itself. Much can be built on this foundation that cannot be built on commercial oil; this is especially true when using the putty medium, see also briefer explanation below. Commercial oil, at this point, is not necessarily "bad," but it is refined by a large scale, industrial process that is designed to maximize its shelf life as food and therefore inhibits its drying (oxidizing) potential. No paint company refines their own oil. Commercial paint contains a system of additives to make a lean paint appear rich or "buttery" and does not lead logically to the old craft, especially in terms of bravura or broken surface painting techniques readily available to paint made with hand-refined oil. It is, in a fundamental sense, incomplete. This is why, since the decline of painter-made materials, so much emphasis has always been placed on the medium: something is missing that was inherently part of the older system. It is important to view the oil, not as a mere placeholder for the pigment, but as the literal vehicle of the journey from blank canvas to art.
science and the washing method
Research has established that in a water-washed oil the fatty acids have been altered in three ways: by increased oxygenation, by increased dimerization (molecular linking), and by cis-trans isomerization (meaning that large molecules that were bent become straight, and therefore easier to link when polymerizing). These factors add up to an oil that is much more readily polymerized, i.e., dries more quickly, about 2.5 times faster than raw linseed oil. In other words, the efficacy of the traditional washing method, dating to the 17th century in print, is confirmed by the findings of current research. More on the science of this in explained in Effects of traditional processing methods of linseed oil on the composition of its triaglycerols (2004) by Jorrit D.J. Van den Berg, Nicoletta D. Vermist, Leslie Carlyle, Michal Holcapek, Jaap J. Boon. Aspects of this are relatively accessible, notably Dr. Carlyle's seminal contribution, and it can be downloaded free online.
the lack of meaningful standards
There are still no professional standards for oil painting products, and commerce is very good at telling half truths about what it markets. There is often a veneer of "science" involved at this point: the more science is used as a marketing tool, the more important it is to look at what is not being said, or claimed. It is difficult to get any manufacturer to talk, for example, in concrete terms about the oil they are using: where it is from, how it has been processed. Safflower oil is marketed as non-yellowing, but without added driers, safflower takes over a week to dry. Have driers suddenly become non-yellowing, or making the paint film brittle over time? This is doubtful. An alkali-refined oil will invariably yellow more than the more expensive cold-pressed oils, but one manufacturer -- who uses alkali-refined oil -- claims there is no "chemical difference" between the oils. Note how often "science" is roped in for support! That this is untrue can be shown by the different behavior of these oils in both yellowing and drying tests. How much this matters depends on the colour scheme involved in the painting, but it is a simply true that alkali refined oil yellows more. There is a tendency both on the part of commerce and painters to look for some sort of magical solution from science. Yet, there is a logical disconnect between modern industrial science and Old Master materials that is unbridgeable by any means. The only trustworthy science about painting materials comes from technical art history, not the marketplace. In terms of the oil, the older solution is a quality cold-pressed oil which is then refined to remove impurities, this has been proven by the condition of 15th to 17th century paintings. This oil is not available commercially but, just as it originally was, can be refined by the painter. This is not hard to do, and creates a qualitatively different foundation for the work.
A simple example of what became lost. Both of these bottles contain linseed oil. The one on the right was hand refined according to an ongoing recipe trail that begins in print in the 16th century. The one on the right was put up by a Well knowN international art materials company, probably in the 1960s, probably a hot-pressed, alkali-refined oil. The oil always darkens the most where it is not exposed to the light, under the cap. As you can see, one of these things is not like the other. Will age darken the oil on the left? No, darkening occurs as a result of the byproducts of the initial polymerization. This is why, when the varnish is removed, well-made older paintings such as Bacchus and Ariadne or The Night Watch look like they just left the easel. Cold-pressed, hand-refined linseed oil does not darken over time. This is why a company like David Davis in New York was able to sell a hand-washed, cold-pressed oil for so long, for so much money, to knowledgeable painters. These painters had experienced the alternative. Are things better now? Yes, but there's still oil being used in paint by some manufacturers that gets quite dark, and no modern oil dries well on its own. Can you buy an oil like the one on the left, which dries overnight and doesn't yellow? No, no one would pay for it, it has to be handmade. And if you make it, you will not only have a great oil to work with, but a vital clue about what has been lost.
A broad spectrum of commercial materials are available now, so broad it can be confusing. As has always been the case, it is best to do research into the actual quality of a product, rather than accepting a company's marketing strategy as the truth. One manufacturer is even calling chalk an "adulterant." Yes, the adulterant of Vermeer, Chardin, and Rembrandt. The loudest voice on the forum is typically not the smartest. Marketing is becoming increasingly involved in demonstrating how "scientific" and "committed" the company is. However, it is not 1845 anymore, there are 7 billion people on the planet, and the economic pressures on commercial art materials -- from above and below -- are unprecedented. With regard to oil paint, there are also significant pressures from general consumer expectations about what paint "should" be. All of this means that commercial paint, in the official artist grade, is all relatively similar, some being thicker or thinner, some containing chemical additives, some relying on the older mineral additives such as chalk. But no commercial paint uses a cold-pressed, refined oil, this would simply be too expensive. What type of oil were all paintings in the 15th to 17th century made with? Cold-pressed and hand refined. The oil is the foundation of the material, but it is the last thing the larger producers want the painter investigating. Does it matter how fine, or how evenly dispersed the pigment is? Commercial paint is typically two to three times as fine as the median particle size in the Van Eyck's iconic Adoration of the Mystic Lamb. Does this make it better? Commerce sells the sizzle, hoping no one will look at the actual steak. This is not to say that commercial materials are necessarily bad, but that they are based on a set of modern assumptions dictated by the marketplace, and have built in limitations. Once all additives are removed from the process and the oil is hand-refined, it is possible to address, in general terms, what happened five and six hundred years ago at the easel. Building the craft into the process at this level allows for much more in the way of reliable creativity with the materials.
Resins are often mentioned in the "secret ingredient" category. Early researchers such as Eastlake or Merimee in fact felt that resin -- hard resin varnish such as amber or copal in their case -- was the official "lost secret" of older painting practice. We know now, through the research in over thirty National Gallery Technical Bulletins, that resin use was never global for any major painter from the 15th to 17th century. The most typical use was of pine resin was in "very small amounts" in order to make a red lake glaze dry. Mention is made of this because the Old Master shell game -- who used what "special medium" when and where still goes on -- and much time and capital can be lost getting involved with it. If you encounter resin extolled as "the secret of the Old Masters," look carefully for concrete research evidence. There won't be any, because there isn't any. There is simply a lot of smoke and mirrors. Contemporary research is "all wrong," only a few, select people truly understand the "right" way to paint, the secret of which which just happens to be for sale. Beguiling literary detective work sells the aura of the fabled past in a tube or bottle. I have made all the hard resin varnishes, and have worked with all the others. They are interesting materials, but only one aspect of an equation that is essentially multi-dimensional. What about the oil, the refining of which is the cornerstone of the well-known 17th century De Mayerne Manuscript? It is ignored. Yet, when hand-refined, oil is capable of a great many rheologies that modern oil is not. This isn't to say resins should be excluded, but it is important to keep in mind that all resins yellow over time due to oxidation, resins tend to involve the process with solvent, which also tends to yellow over time, and neither of these materials are necessary. The adhesive element they provide can be created many ways without them using traditional materials and processing methods.
Mediums based on mastic varnish and black oil continue to be marketed as the "special" Maroger medium. The most special thing about this product is that it is a sure ticket to a much more fragile paint film that will darken significantly over time. Maroger was not a painter, but a technician at the Louvre who concocted a fantastic series of "authentic" recipes based wholesale on the technology in Merimee"s The Art of Painting in Oil and Fresco, published in 1837. Maroger's book, Secret Formulas and Techniques of the Masters (1948), contains no actual research and is betrayed by a "logic of the medium" that occurs over many centuries and locations, from Leonardo to Rubens, while the paint itself is not mentioned at all. Maroger was first exposed by A.E.A. Werner in The Vicissitudes of the Maroger Medium published in Studies in Conservation (1957), in which Werner points out that Maroger uses the De Mayerne Manuscript as a source for the mastic gel medium, when no such reference exists in the manuscript. The history of the mastic gel medium is traced in perspicacious prose and great detail by Dr. Leslie Carlyle in The Artist's Assistant, and it is interesting that this medium, while possibly having its origins in Venice, is primarily an English phenomenon, and is in fact referred to as "Vernis Anglaise" by M�rim�e. Mastic gel mediums are routinely condemned by conservators and painting professors in the 20th century, but do occur in 19th century work occasionally which has not fallen apart. Unfortunately, this is sometimes paraded as evidence that the medium "works", when in fact it is much safer to say it will always darken, but may last over time if used with great awareness of its fundamental limitations. A quality overview of the perennial mastic issues occurs in Turner"s Painting Techniques (1993) by Dr. Joyce Townsend. Paint and Purpose (1999) by the Tate also gives a good overview historically of what happens to paintings made with mastic over time. Problematically, the medium is always sold as a kind of magical panacea, which is analogous to saying that a daily dose of opium will make all your aches and pains go away. It is also important to note that Dr. Carlyle's book explores the history of English painting materials in the 19th century, and, while a great effort is made to point out quality technique when it occurs, the book is technical art history, not a manual of technique approved by conservation. So, whether Maroger was an accidental or premeditated charlatan, recent analysis of older paintings by technical art history contains no confirmation of his specious system, rather the reverse. The book A la recherche des secrets des grands peintres, published in 1986 after Maroger's death, gives insight into where his research ultimately led, notably to emphasis on the original quality of the oil -- cold-pressed, water-refined -- and on smaller proportions of mastic and leaded oil in the paint film.
Alkyds are often called resins but are actually modified oils and are made industrially from many different vegetable oils. Because alkyds are made in great quantities, they are typically inexpensive, and are promoted endlessly as "just fine" by the larger manufacturers. They make a strong paint film but take a long time to dry unless they have added driers. As such, all commercial alkyd mediums have added driers, and all driers cause yellowing and brittleness in the paint over time. Cracks from brittleness are a definite possibility, for example, with a fast0drying commercial alkyd ground. Commercial alkyd mediums are definitely prone to yellowing, this has been reported by many correspondents. Instances of delamination associated with the use of alkyd resin over an acrylic ground are also common. But, just as there are qualities of oil, there are qualities of alkyd. Kremer Pigments, for example, carries a castor oil alkyd (#79240) that is thin and light in colour. It is expensive, takes several days to dry, but does not yellow. This could be used in small amounts in a fine technique with a long open time, but so could the logical traditional material, walnut sun oil.
the oil defined
While quality modern commercial linseed oil can dry without yellowing, this is at the expense of the drying rate of the oil. From left: Hand-refined cold-pressed organic linseed oil, hand-refined cold-pressed organic linseed oil exposed to oxygen for three weeks, cold-pressed linseed oil put out by Holbein. It is sometimes claimed that there is "no technical difference" between a modern cold-pressed linseed oil and an alkali-refined linseed oil. This may be more true than not in the context of commercial oils, but a hand-refined organic cold-pressed oil is a very different case, drying twice as fast as ANY commercial oil, even faster if pre-thickened as the middle oil in the photo. If your oil dries quickly and doesn't yellow, why do you need anything else?
Various finished versions of refined organic linseed oil aging in the light. The cold and the addition of a small amount of chalk or lime tend to precipitate out further small particles of break. One would think that the lightest oil would be "the best" but this is not necessarily so. The most important thing is not the color the oil begins, but what color it is after being dry for six months.
A painting is made up of layers, beginning with the canvas or panel, proceeding to the final varnish. The ultimate strength of the painting technically depends on the strength of the structure as a whole, not on the cumulative strength of the ingredients themselves. This is important to consider in using any given material: is it appropriate for its place in the structure? A material which is too strong for it's position, such as glossy commercial acrylic gesso, may create resulting structural weakness: the applied oil paint ultimately flakes off. The use of a strong material like sun oil or copal varnish too soon or in too great a concentration may result in the beading of the paint in subsequent layers: the painting will need to be oiled and sanded down to accept further paint. It is important to think in terms of designing the structure of the layers of a painting so that the ultimate strength of the whole is maximized. This means using the right material in the right place.
Three great aids to structural flexibility from the history of oil painting are the use of panels, the use of chalk or other calcium carbonate as a buffer to the long term acidity of the oil, and the greater film strength over time of lead carbonate as a white pigment.
The manufacture of canvas on stretchers has become a large industry, with many different levels of quality available. There are also many levels of quality of stretchers themselves. While work on panels predates work on stretchers by centuries, stretched canvas has come to be expected. This is unfortunate, because unsupported canvas will always be less permanent, more subject to various forms of potential damage, than that same canvas mounted archivally to a panel. It is always best to stretch the fabric oneself, and to use linen if at all possible on stretchers. If using commercial canvas, beware of fabric which has been given a shiny surface by the ground, either an oil ground made flexible by the addition of Stand Oil or an acrylic emulsion ground. Also beware of hardboard panels, these are not necessarily archival or acid-free. It is simple to make highly archival panels oneself using cabinet grade plywood covered with cotton or linen canvas. A more recent addition to this family of products is Medite, a particleboard bound with a synthetic resin. This is receiving high conservation marks at this point. An excellent quality plywood for small panels is the Baltic Birch plywood used for furniture making. Larger panels can be made with 1/4 inch plywood cradled at the back.
linen, cotton, paper
Linen is the stronger fiber and should be chosen for unsupported canvas if possible. If cotton is chosen, the reverse surface should be protected from possible damp by sizing it with PVA or by the older method of a tightly fitting panel insert, this goes back to Mantegna. Cotton on panels is fine because it is permanently protected both front and back. Small paintings on prepared paper -- usually outdoor sketches -- have fared surprisingly well over the last few centuries, but should always be mounted to panels or framed behind glass in the long run.
The traditional size was some form of hide glue, rabbit skin glue is often said to be the most flexible of the affordable glues. When combined with glue gesso on panels, this system has survived in some cases for over five hundred years. On unsupported canvas it is less permanent because of the movement of the canvas and the hygroscopic nature of the size. The most important thing to understand abolut glue size is that it begins to degrade at a relatively low temperature, about 150F. This means it is possible to make glue in a double-boiler and overheat it significantly. It is really best to make glue with a thermometer and keep the temperature low. There are alternatives available now in acrylic emulsion size and PVA size, both of which are more flexible and resist moisture better. Both of these are also stronger, so the questions become: how strong to make the size, and what to put on top of it.
It is important thing to keep the ground lean and at least somewhat porous or toothy so the paint will adhere well. A commercial oil ground on canvas presents potential issues with both adhesion and yellowing. This is because these grounds are made to be rolled and typically contain stand oil. Although the stand oil creates more flexibility for some time, it will darken over time and creates a ground whose surface is relatively slick, with the potential for the paint not to adhere well.
The danger of a commercial acrylic ground is the typical level of gloss making it difficult for the paint to penetrate the surface enough to safely adhere. Commercial acrylic grounds have often been the cause of delamination, because the oil paint cannot stick to their glossy, sealed surface. These can have chalk or marble dust added and thinned a bit with water to create a matte surface that will hold the paint better. Some acrylic grounds now are being made moderately absorbent, these will have no shine and a drier feel.
Because they are rigid, it is possible to put a quality ground on a panel commercially. Still, this is an area where the painter will be able to create a much more personalized and interesting painting surface with ease.
On panel, the logical solution remains hide glue and glue gesso, this gesso can be enhanced with titanium white, and made in a variety of ways that then affect the paint applied on top. It is also possible to make a more flexible oil emulsion gesso, as discussed by Kurt Wehlte in "The Materials and Techniques of Painting," for use on canvas. For more information on this, go here. All glue gesso grounds can have a thin white lead imprimatura layer applied, this is surprisingly consistent throughout the history of panel painting, although even a thin layer here needs to have a few weeks to cure hard.
The quality of linseed oil in America during most of the 20th century was not high, texts from that period take significant yellowing of the oil as a given. While modern commercial oils are better, the oil used for painting should be tested for yellowing, because some oils are better than others. Drying tests with oil take about six months to complete. If the oil is cold-pressed, this is often a key to higher quality. At another level comes an unrefined cold-pressed oil, usually organic as well in the case of linseed oil. However, this must be refined before use. The various low-tech processes available to the painter all take time, but are not that time-consuming or difficult, and open up a range of possibilities not possible when starting with commercial oil. This is especially true with linseed oil: because of the time factor involved, and the original high price of the oil, no approximation of the linseed oil of older painting exists commercially. While commerce offers refined linseed oil which yellows minimally, this oil has also lost all verve or snap to the refining process, it is essentially generic, remaining linseed oil in name only.
Artist refined cold-pressed organic linseed oil is a different product. How much this matters depends on how of much of a role the craft plays in a given painter's conception of the work. But if someone tells you categorically that it doesn't matter, or is not worth the effort, take a moment to stop and consider the source. It is ironic that, while highly processed oils with high price tags are sold in the art supply store, the interest in the nutritional qualities of unrefined linseed oil have made oils of incredible quality and potential available at the health food store. These can often be found more cheaply online. However, these oils must be refined to dry quickly and remain safely non-yellowing over long periods of time. Adding bread crumbs to clear a cloudy oil, then placing it in the sun for two weeks, is not enough.
The oil used in a painting medium should also not be raw. This is especially true if raw oil commercial tube paint is used. Preheated oil offers a very simple method for decreasing yellowing, and increasing the overall strength and potential longevity of the paint film. This is more true for linseed oil, although artist processed linseed oil is significantly less volatile than low quality commercial linseed oil.
More information about working directly with the oil can be found here.
An extensive pdf file about the process of refining linseed oil is here. This is all text, no photos, for printing out.
New for 2020! Download a detailed PDF selection from Living Craft about linseed oiland how to keep it from darkening .
Well, painters often groan when tests, but they're the most reliable way to put the process beyond the uncertainty of opinion -- including one's own -- and the endless diablerie of commerce. Once the habit is formed, it is actually pretty interesting to see what happens over time with a given set of materials. You can even test a medium incrementally, ingredient by ingredient. Tests take time, usually from three to six months to find out how much an oil will yellow, and longer for a medium containing a resin. But you end up with reliable information, and lots of it. Ah, but why take my word for it? How about making your own tests? Come to your own conclusions! Start thinking in the materials for yourself! Of course, this is EXACTLY what our good friends who manufacturer painting materials don't want us to do. We are supposed to be good consumers, pay attention to the sizzle, not the steak, and buy, buy, buy. Photo of a recent test panel, taken after the humid six months, so I'm pretty happy with this one so far. The ground on the panel is glue gesso, the same gesso I use to paint on.
June 2021: I make these test panels repeatedly, they're all over the place, and sometimes one gets buried. That's what happened to this one here from 2017, it was literally in the dark for a few months. That's the photo on the left. So then I put it in the light for two weeks, that's the photo on the right. A little bit different! This test panel featured lots variations of commercial pre-polymerized oils: I wanted to see how much they darkened uncut, then cut with various other oils I knew yellowed less. It eventually became clear that these oils needed to be used in small amounts, if at all, for maximum brightness over time. At this point I'm trying the opposite approach, letting go of all commercially heat-polymerized oils, and like this better. But the point is, if linseed oil is involved, the amount of light is important: darkening can often be almost completely reversed by lower humidity and higher light levels. This is also why it is important to do your own tests, to be in control of the variables involved.
Tests are so helpful that some manufacturers themselves are conspicuously involved in tests. But, for more complex reasons than might first be apparent. In fact, I'd suggest caution about interpreting tests from a commercial source. It's not that anyone would lie, especially when things look so organized and scientific. It's just that they might arrange the outcome, very scientifically, at least on the surface, so that it's easy for you to come to the conclusion they want you to. The facts actually do tell half the truth, but in order to promote a larger lie. With commercial tests, pay attention to what is not being said. Notice where you're being directed, then look in the opposite direction. I'm not saying a commercial interest would rig a test. Gasp! But a commercial interest just might gently, articulately, and oh-so-scientifically, misdirect your attention. Thus gently, articulately, and oh-so-scientifically allowing you to fool yourself completely.
The simplest way to thicken an oil is to leave it open on the palette, or closed in half full containers. This is especially true of painter refined organic linseed oil, which becomes tackier and more glutinous in a matter of a few days in a thin layer in the open air. This oil can be used at any number of viscosities, and offers a simple, high quality way of working from lean to fat.
The earliest known recipes for painting oil all direct that the oil be heated for a reasonable length of time. Heating an oil to 150 degrees Celsius with constant stirring for an hour or two results in a slight increase in density with little increase in color. This oil dries faster and is more leveling than raw oil, and is still thin enough for undiluted use.
The other traditional method of producing a thicker oil is to allow it to thicken in the sun. If you are in a climate where this is easy in the summer, sun oil is an excellent and simply made material. It can be made for undiluted use in a few weeks, or allowed to thicken to the point of being a taffy. True commercial sun oil is always very pale, relatively expensive, and smells quite sharp. If you have found a surprisingly cheap "sun oil" that has a denser, cooked smell, this is in fact modern burnt plate oil, see below. When making sun oil from any oil, this oil should be refined, either commercially or by the painter. While an unrefined organic linseed oil placed in the sun for several weeks will dry somewhat faster, it will not dry as fast as that same oil, previously refined, and still contains a full complement of water-soluble fatty acids which can yellow significantly as the paint film oxidizes over time.
Stand oil is a more modern thickened oil, usually made with low quality linseed oil which has been heated in a vacuum. Stand oil is produced by many manufacturers, and from many different grades of oil. It can be, but is not always, a reliable material. If the working qualities of stand oil fit with your style, please make sure to test several different brands for yellowing, these tests take about three to six months. The best one I've found is the Kremer 73201, this is relatively quick-drying and non-yellowing.
Another modern commercial thickened oil is the burnt plate oil of the graphic trade, used to alter the viscosity of printing ink. The original burnt plate oil was literally oil that had been heated until it caught fire. Modern burnt plate oil is an evolution of stand oil which involves heating the oil in a kettle which sparks off the rising volatile elements. The result is a thicker oil which has a more mobile or slippery quality, and which has less of a tendency to yellow than stand oil. Burnt plate oil is also very slow drying, and imparts a great depth of saturation to the color. For any type of realism, it needs to be used in very small amounts in a medium or putty, 5 percent is enough. If working with the product from Graphic Chemical, the most versatile grade is 5, about the consistency of stand oil but lighter. Photo above is of grade 7, somewhat thicker but still on the light side.
The very thick Kremer stand oil, this is made from Swedish linseed oil and does yellow a little more than the thin version in an unpigmented test, but it also dries much faster.
Mediums became increasingly important once paint began to be engineered to be put into tubes and kept for long periods time. Modern conservation research has made it clear that, while many materials were additionally used in older painting in decided moderation, the oil itself was by far the most prevalent ingredient in any medium. It's very important to keep any resin to a minimum. A small amount of hard (cooked oil) resin varnish such as amber or copal has more to offer protectively for the paint film than a soft (spirit) resin varnish, but also has more potential to yellow. The research in the National Gallery Technical Bulletins suggests that older painters may well have used a small amount of soft resin fused into the oil as a simple, readily-made solution. Damar can be used for this, it fuses just above the boiling point of water. Another quality resin for this purpose with a long history is Larch Balsam. This is what Ralph Mayer refers to as Venetian Turpentine. The Venetian Turpentine on the market is a mixture of Larch and Colophony, which will not dry as quickly and yellow more. Canada Balsam is popular, but has recurring drying issues, and Gottsegen, backed up by the CAMEO database, feels that it yellows over time. Mediums are a place to avoid complexity. This is especially true in terms of maintaining paint film consistency from layer to layer by using the same medium throughout the course of a painting. The medium should always be leaner or used more sparingly in the beginning layers. It can be used more richly at the end, or in an alla prima painting which will dry in a single layer, but the spot use of different additions or underlayer treatments for specific effects need to be avoided; these have proven to cause significant difficulties with paint film integrity in the long run. The Tate book on the Pre-Raphaelites illustrates several examples of what happens when painters overthink the process without understanding the structure of the materials in the first place.
Although simple, this medium has no historical reference in any older text. It's origin is the result of modern conservation research, which has found additions of chalk or other forms of calcium carbonate in the paint of painters such as Rembrandt, Chardin, or Velasquez, and ground silica in the paintings of various older Venetian painters. The various types of ground calcium carbonate and silica can be coupled with the stability of preheated oil and the quick drying nature of sun or unsun oil to produce a family of mediums with many different working characteristics. The strength of the medium lies in it's technical stability, solvent-free ease of use, and protean versatility. This last quality also means that using the medium effectively can involve a bit of a learning curve as the painter becomes acquainted with it's endless possibilities and adapts them to personal taste. Older painting practice contains many subtle elements. Next to the craft of the oil, the use of inert stone dust as an extender, brightener, stabilizer, and impasto creator is at once both the most prosaic and the most profound.
At it's most basic, this medium is simply ground stone dust used in conjunction with oil and the paint on the palette. Oil to make the paint move more, stone to tighten it once again. From the highly irregular surface impasto of some of his later self-portraits, this may well be one way, if not the only way, Rembrandt used this method.
The medium can also be premixed into a putty. This putty can be mixed into the paint before painting in any amount, giving access to a wide range of working characteristics. Putty can be used to make impasto and internal texture but can also be used in smoother surface styles. Conservators have found evidence of Rembrandt's use of translucent glazes using chalk, chalk has also been found in Vermeer's Girl With a Pearl Earring (c.1665).
Other forms of calcium carbonate can also be used and have slightly different rheologies as putty. Chalk is the most absorbent and, unless it is very fine, also has a certain mobility. The finer chalks, such as Champagne Chalk, have a somewhat stickier quality in oil but are also suitable for the finest detail work. The more crystalline forms of calcium carbonate - marble dust, calcite - make a putty which is more adhesive but with less potential for the glutinousness of the finer chalk putties. As Vel�squez used ground calcite and Rembrandt used ground chalk, the difference isn't one of quality, but of personal preference, or perhaps simply availability, coupled with the all important element of long experience with a given material. Pure white and finely ground marble dust is readily available and works well as a point of departure. Ground silica presents a different case depending on particle size. Larger particle silica - 200 to 400 mesh - will tend to a make a putty feel drier under the brush. The smaller particle silicas - such as fine cristobalite - tend to produce the lubricating effect, sliding more than the other finer particle stone dusts. While the calcium carbonates tend to accelerate drying slightly, silica tends to retard drying slightly. Any ground silica presents a long-term respiratory hazard and should always be used with a serious particle mask when in the dry state.
The putty medium is especially effective at altering the relative lack of body and tendency of commercial paint to slide due to the common modern addition of aluminum stearate as a gelling agent. By using only high quality, preheated oil in the putty, the instability of raw oil commercial paint - its tendency to shrink and crack - can also be significantly diminished. The putty can be engineered to any level of final gloss as an addition to the paint, again overcoming the tendency of raw oil paint to dry matte and sink-in. It is ultimately possible for the painter to create a tube -- or several different tubes -- of putty which, when added to commercial paint, consistently give it a number of desirable characteristics it does not have on its own. This presents a viable alternative to the longer learning curve of making one's own paint. More on the putty medium can be found here.
silica gel mediums
This is a much more reliable gel than a mastic gel based on adding fumed silica to the oil. When using tube paint, the oil used should be preheated to avoid any yellowing potential. Many varieties of silica gel can be made by varying the amount of silica and the mix of the oil. The typical way of using this is as a glazing medium or in conjunction with the putty medium. The silica gel moves, or slides, easily, while the putty has more of a tendency to stick or grab. A fumed silica gel can also be used in a freer manner in alla prima painting.
Fumed silica is very light and should be handled with great care, using a quality respiratory mask. In America, Cabosil M5 is readily available as a thickener for epoxy used in boat building. In Europe, Aerosil 300 is transparent in oil. Once combined with the oil, it presents no further risk and can be tubed or kept in aluminum foil wrapped with masking tape or duct tape.
fused damar and beeswax mediums
This a medium I made in a lot of variations starting in 2016, a mix of thicker oil, crushed damar, and beeswax. The idea here is to get paint that dried brighter, and held its chroma more over time. So, this medium gives the paint more resistance to humidity than an oil-only approach. The damar is fused into the oil on low, repeat, low, heat, the heat is turned off, the beeswax is then added, stir until it melts. You can add chalk or marble dust. Very versatile, definitely creates more sequestered or bright paint either alla prima or in layers. Just keep track of the proportions of the ingredients to one another, and the proportion of medium you're mixing into the paint. A medium that is fifteen percent damar, for example, sounds like a lot, but if it is used at one part medium to three parts paint, this becomes a little less than four percent damar in the paint film.
fused damar and beeswax video
The quality of commercial paint is improving, the days of dark orange oil oozing from under the cap are on the wane. But it is still an excellent idea to compare brands, do research and yellowing tests of your paint, regardless of the brand's pedigree. Artist made paint will always have the potential advantage of being made with higher quality oil. No manufacturer, large or small, uses cold-pressed oil refined in a low-tech, non-invasive way and then aged in the light, as was the common older practice. In terms of linseed oil especially, this matters. On its own, commercial paint does not develop a strong enough film to resist oxygen or the subsequent sinking in of later paint layers that well. This is why the element of medium has become so important in modern painting practice. The pigment needs to be suspended and sealed from oxygen in a way that it cannot be using raw oil tube paint alone. Commercial paint is moving away from linseed oil to safflower oil. It is important to note that, while safflower oil is relatively non-yellowing, it is also a very slow drier unless driers have been added to the paint. The wiser small manufacturers are returning to the simplest paint possible, with minimal or mineral binders.
However, if the paint is made with preheated oil, as in the illustration, a very different rheology is available from the start. Preheated oil is found routinely in older paintings by the researchers at the National Gallery, as reported in their yearly Technical Bulletins. The use of preheated oil creates a stronger and more stable paint film which is less likely to yellow. The increased body of preheated oil makes it less likely that the pigment will sink in, as well as altering the rheology away from "long and stringy." The oil used to make this paint was preheated to 100C for 48 hours. It also works to preheat the oil to 150C for an hour. This paint contains nothing but oil and pigment, it is short and dense, but mobile and bouncy.
fat over lean
The structural rule of painting fat over lean is very important to observe overall within the structure of the painting, especially if raw oil is used with raw oil tube paint. It becomes less important the more the paint and medium are modified in the ways that are standard from sources such as the De Mayerne Manuscript and the findings in the National Gallery Technical Bulletins. This is because these methods -- principally preheating the oil, and introducing lead into the oil -- produce oil which is stronger, more flexible, and significantly less volatile in it's drying characteristics. Another great addition for long term stability is the addition of calcium carbonate to the paint through the use of the putty medium.
Many painters take fat over lean very seriously because of the emphasis given to this principle by Ralph Mayer. It is important to realize that Mayer was born in 1895, and lived through a period of the worst commercial linseed oil imaginable. If a painting does not become relatively fat at some point, one is always underpainting, and these layers are always sinking into one another, producing a leaden opacity, especially in the midtones. The key to avoiding this situation is to understand that commercial paint, for all its "buttery" texture, is quite lean, and begin to introduce elements of preheated or otherwise thicker oil to the layers in small, discrete increments. At the same time, the ground itself must be kept lean, or there can be adhesion issues over time. The classic recipe for delamination in modern materials is a slick acrylic ground painted over with an abundance of alkyd medium.
A more complete discussion of fat over lean, excerpted from Living Craft, is here .
There are several different potential causes of yellowing in a painting. While lower quality linseed oil in commercial paint is on the wane, this has been a major cause of darkening in 20th Century paintings, and it is unfortunately possible to see this readily in major museums. Another standard culprit is turpentine which has been oxidized by exposure to light and air, the residue of this can yellow badly and affect everything it was a part of. If used, turpentine should always be high quality, and protected from both light and air by being stored in small size amber glass bottles. A thin application of a given oil may dry without yellowing, while a thick application of the same oil will darken. Oil which has been preheated or aged in the light has less of a tendency to yellow than the same oil, new and raw. Humidity is another factor often overlooked. An oil which dries without darkening in low humidity may dry with significant darkening in high humidity. In some climates, a dehumidifier is a studio necessity during the summer. Setting paintings to dry in moderate sunlight is a traditional remedy for short-term darkening, north or east windows work well for this. Paintings will darken naturally if not exposed to enough sunlight, this is especially true of work made with linseed oil. However, subsequent exposure to light will brighten the work once again. The darkening associated with resins is longer term. It is wise not to use a hard resin such as amber or copal in any but the most minimal amounts, and never as a final varnish. It is wise to keep all resin use to an absolute minimum: oil paint is so sensitive and these materials so strong that between 2-5% of a hard resin varnish in the paint layer will make a significant rheological and optical difference. This is enough. More may well be inviting trouble down the line. Oil paintings made with permanent pigments do not ever get lighter or brighter over time. Painting on a white ground is therefore a good idea, as is painting the values "up" somewhat, especially when working in layers. It is much easier to continue to darken a painting, than to lighten it once again reliably.
It is becoming more common for painters to pay attention to conservators, and this is by and large a good development. But sometimes conservators become obsessed by yellowing. In oil painting, there is a profound difference between normal, visually comfortable mellowing and an overall darkening which is fact disfigures the work. It is important to keep in mind that the dire warning of the conservator about a material "yellowing over time" might be a little overstated, some conservators simply like to get the attention of painters. Conversely, it's possible to see before and after photos of paintings where old varnish was removed which show that the painting clearly benefited from the warming effect of the varnish layer�as the painter may well have intended. So, the issue is to know one's materials. A lower chroma earth color palette, such as used by Rembrandt or Velasquez, can be used safely with quality linseed oil, as was in fact the case for both painters. Raphael's higher chroma palette may well have benefited from his use of walnut oil. The more the color depends on a cool, neutral light, the more the color is focused on pure color in the midtones, the more the painter needs to be concerned with yellowing in the choice of the oil. However, as will be seen below, it is possible to paint in such a way that the relative warm-cool dimensional distance of the painting remains constant or in fact increases as the painting ages.
Color is the great chameleon, it's behavior is always contextual and relative. Modern painting practice tends to feature more color as opposed to wise color. Older painting practice often featured great discipline with regard to the use of white and the separation of warm and cool tones on the palette, resulting in the illusion of brighter color and greater dimension from a relatively to extremely limited palette. This can produce as astonishing sense of chromatic harmony. For more details on this extremely powerful tool, go here.
Color is organized into triads. For realistic painting a primary triad that is warm and a primary triad that is cool are all that is necessary in addition to white. There are many possible triad combinations that can be mixed and matched for various situations, rather than resorting to a greater variety of colors. Exploring this helps to assure harmony and a convincing spacial envelope.
A relatively warm, low chroma triad composed of three earth colors: Yellow Ochre, Venetian Red, and Ivory Black.
A cool, high chroma triad: Primary Yellow, Primary Magenta, and Primary Blue.
White lead is capable of a great many different rheologies or working characteristics depending on how it is made and how it is modified. Paint made with pigment and oil only is more responsive than commercial paint.
Solvent is unfortunately often considered necessary to oil painting when in fact is is only necessary to the type of painting which uses a soft resin medium such as damar. Older painters kept their brushes in oil, not solvent, and the use of solvent by Rubens was discussed by De Mayerne as an innovation. Long term solvent use in an unventilated studio leads to significant health issues, Denmark has in fact labeled this "Painter's Dementia." Solvent should always be kept in small amber glass bottles which are full, protecting the contents from both light and air. This is especially crucial with turpentine to avoid yellowing, it may be best to forego turpentine altogether at this point. Both the traditional putty medium and the more recent silica gel medium offer a variety of solvent-free painting options. Solvent is also unnecessary when using a small amount of hard resin varnish in the medium or a fused soft resin and oil medium. Odorless mineral spirits is sometimes promoted as "safe," but it is still evaporating into the room. In other words, a solvent that works is a solvent to beware of. Sometimes citrus thinner or spike lavender are promoted as "healthy" because they don't tend to give people headaches, but this is irresponsible and dangerous: a solvent that works is still a solvent to beware of, and both of these have very high aromatic content. If you find surprisingly inexpensive spike, possibly in Brooklyn, it has been cut with mineral spirits.
Here is a PDF containing an overview of nine different spike lavender products. More detail about the spike lavender situation is available in the paragraphs below. A PDF file with an overview of the Art Treehouse-Chelsea Spike Lavender Fraud is available here
the spike lavender fraud
Have you seen this ad? It conceals a fraud. Spike lavender has a long history of use in oil painting, we are told that it was distilled by Durer and Leonardo, probably because it is a strong enough solvent to dissolve sandarac for use in a medium or as a varnish, similar to the way damar varnish has been used more recently. Spike lavender has been marketed recently as a "safer" alternative to more toxic traditional solvents. Yet, the price of spike lavender varies widely. Spike lavender sold to painters tends to be inexpensive, whereas spike lavender sold as an essential oil is about 500 dollars a liter, when it comes with a GC/MS analysis that proves it is actually spike lavender. And this is the crux of the problem. The MSDS of "spike lavender" from companies in Europe, compared to the ingredients that are in real spike lavender, shows that these materials contain little, if any, genuine spike lavender. It is chemically impossible. The older method, documented since the 17th century De Mayerne Manuscript, of cutting real spike up to 75% with turpentine has given way to more sophisticated methods of concocting spike. There are three ways: 1) from various other solvents and a little cheap lavandin, 2) from synthetic versions of its major components, or, 3) from other essential oil components that are less expensive. For example, if you mix camphor and eucalyptus oil with a small amount of the cheaper lavandin, this produces a strong solvent that has the characteristic camphoreceous lavender smell of spike. Some variation of this last method is the one in use by the recent American companies that are marketing spike as a "safer solvent". While these companies have all produced elegant MSDS sheets to go with their product, there is no actual analysis of the product in them. One of them summarizes importantly that spike does not cause cancer. This is like saying that assault weapons do not cause cancer. In America, an MSDS available to the consumer is not required. So, what is being sold as spike lavender in this situation? The buyer has no way to know. Note too that, if you look at the MSDS provided, the sellers are quite careful not to actually claim that their product is 100% spike lavender. No analysis, no mention of lavandula latifolia.
So, there are several points here. First is the basic fraud of calling something a false or misleading name. But, unfortunately, this is something that is endemic in our culture at this point. We need to require genuine MSDS files with these products. Second is the misleading way the "safer alternative" approach markets their product as Mother Nature's gracious solution when it is in fact thousand of times more concentrated than the flowers from which it is made and comes to them in a 55 gallon drum. Is this natural? It is true that the neural toxicity of VOCs (volatile organic compounds) from solvents like turpentine or D-lymonene is considered greater than the toxicity of FOCs (floral organic compounds), but this is like saying that wine is safer than cognac because it contains less alcohol, when the health issue is how much alcohol is ingested, and for how long. The same is true of FOCs. And, as the recent report on lavender essential oil causing the growth of breasts in adolescent boys suggests, there are lots of things about essential oils that we still don't know very much about: Was this actual lavender, or sophisticated lavender? Finally, who needs a solvent this strong anyway? The only real reason to use it is to dissolve resins like sandarac or Manila copal, and specialized resin varnishes like these, while interesting historically, are not exactly necessary to make quality oil paintings. Everybody is different, and every body is different, but this hopefully gives you more reliable information with which to make decisions about spike lavender. (Genuine spike lavender with a detailed GC/MS sheet is available from Lotus Botanicals in Maine for 26.00 an ounce.)
This is a table from a scientific paper showing a comparison of the major and minor constituents in lavender, lavandin, and spike lavender. The major constituents of spike are starred in blue. These constituents are also available as synthetics, or in less expensive essential oils, making the smell of spike lavender easy to fake.
An earlier Kremer MSDS for "spike lavender" dated 2015. The exclamation points are compounds that are not in genuine spike lavender. The MSDS also shows 5% terpinyl acetate (not in actual spike), and 9% camphene (0.2-1.9% in actual spike), which were listed on the preceding page. Compared to the proportions of the ingredients that are officially in spike lavender, this list seems to be so skewed, with linalol itself missing altogether, that this version does not contain any actual spike lavender.
A more recent MSDS for "spike lavender" marketed by Kremer, with the official proportions in red. This is a much more sophisticated sophistication, but, between the skewed proportions and the lack of the ingredients that occur in smaller amounts, there is also no reason to suppose it contains any spike lavender.
The first version of selling a lab-produced solvent as spike lavender was conceived by Robert Maynord, the owner of the "endearingly folksy" Art Treehouse website. The entire system of this fraud was pioneered by Maynord. Say it is spike lavender when it is not, say spike lavender is non-toxic when it is not, back the whole thing up with a phony scholarly MSDS written by an academic, and link it with the romance of older painting practice. Yes, everything in the MSDS is true, even the grand conclusion that spike lavender does not cause cancer. But this is like saying handguns don't cause cancer. The real issue here is damage to the kidneys and cns through prolonged exposure. And of course, just because a PhD collates a lot of research on spike lavender does not mean Maynord is actually selling spike lavender. Again, as with Chelsea, the price point makes this impossible. As with Chelsea, there is no analysis of the product in the MSDS, and a smokescreen of romantic prose on the website. This situation is made more complex by the fact that Maynord presents himself as a healthy painting crusader. If so, why is he selling a synthetic solvent of unknown composition as magically non-toxic? I disagreed with Maynord years ago about the toxicity of spike lavender on the Huffington Post, -- he simply insisted that it is non-toxic and that he had proven it -- but I had no idea at that point that he wasn't even selling spike lavender. Nothing is in fact known about the toxicity of the solvent Maynord sells as spike lavender, because its formula is a mystery. But if it is formulated along the lines of the Chelsea solvent, ALL the ingredients should be listed as hazardous. This is before we even get to the issues of the increased toxic potential of synthetics versus their natural counterparts, or the increased toxicity potential of the oxidation of the solvent's components. Yes, Chelsea has marketed this fraud more effectively at the national level, they are probably selling more "spike lavender" a year at this point than Spain produces. But if we're looking for the origin of the amazing calculated deviousness behind this scheme, we have to give the credit to Robert Maynord, owner of The Art Treehouse.
Whether or not to varnish a painting and how long to wait before varnishing it are complex subjects, depending on on how much paint was put on, the medium used, and the amount of time the painting took in total to complete. It's important to understand that the atmosphere in houses and urban areas is now generally much cleaner than it was during the time when houses were heated with wood or coal: varnish then made the inevitable cleaning safer and easier. Not all paintings are seen the way their maker intended: Monet liked the matte quality of the surface of his paintings, almost all of which have now been varnished. At one point a high-gloss, enamel-like surface was desirable for a painting, no matter how large. Now larger paintings are often finished with a more matte overall surface. The safest varnish is one which is easily reversible, so that it can be removed if anything goes wrong: mastic varnish, in spite of it's inevitable yellowing, was actually preferred for this quality. The final varnish is an area where modern products from quality manufacturers may perform better than older materials, almost all of which have been proven to yellow significantly over time. Possible exceptions are discussed below. It is always best to test the final look of a varnish first on a painting which is unimportant: someone else's idea of the perfect look may not exactly be yours.
A painting made using the all oil system can, with experience, be engineered to any desired level of shine. The more paint, and the richer the medium that can be used safely, the more the painting will be resistant to the destruction of the oil by oxygen. This is an advantage offered by painting on panels over unsupported canvas.
An early varnish used in Italy was made from sandarac, either dissolved in spike lavender or used an an oil varnish. The National Gallery has found specimens of both these in good condition, although the varnish in oil has yellowed more. The varnish used on most 17th Century paintings was mastic, a soft resin dissolved in turpentine. While this was known to yellow, it also aged to the point where it crumbled easily away, making it simple to remove. When damar came into the picture in the 19th Century, mastic was still preferred due to it's lower level of shine. Writing on the experience of older varnish removal from the National Gallery makes it clear that copal, because it both yellows and dries irreversibly, should not be used here.
During the 20th century, damar was thought to offer a final solution. It has now been found that damar will darken over time, although some conservators still feel that it is the resin about which the most is known, and use it with an addition of Tinuvin, a UV absorber that extends its useful life significantly.
There are many modern synthetic varnishes available now which may well offer a more reliable final finish, or have at least been engineered to be easily removed should something go wrong. These include the MSA (Mineral Spirit Acrylic) family, Regalrez (Gamvar), and the older Paraloid B-72, although the latter involves the use of relatively toxic solvents. At this point MS2A, while quite expensive, is considered to be superior in some circles, and is in use at the National Gallery in London. Regalrez is less expensive and is favored by the National Gallery in Washington. MS2A has a somewhat lower level of shine, and brushes out better. Regalrez may appear to be somewhat plastic without an addition of wax, and is so reversible that only one coat is possible using a brush. These resins are always used with an addition of a UV absorber. While these perform better than damar in theory, a look at contemporary conservation practice suggests that not all conservators feel that enough is known about these alternative resins.
Genuine Silver Fir Turpentine, or Olio d'Abezzo, is quite light in color and can be used as a final varnish when diluted with solvent, 1 part resin to 3 or 4 parts solvent. Sandarac can be dissolved in pure spike lavender and used in a similar way. While these materials are traditional, dry quickly and yellow minimally for natural resins, they are brittle and may well cross-link over time in a way that makes their removal difficult. They offer an alternative to the look of damar, and need to also be used with an addition of a UV absorber.
the real secrets
The search for answers with regard to older painting practice began officially when Reynolds set out to emulate Rembrandt. He kept notes, some of which are reproduced in Eastlake, and these notes detail a great many common errors that have been made with mediums ever since. The fundamental conceptual problem has always been the same: a search for a lost material, rather than a lost technique. This shell game still goes on at the commercial level, especially with small companies offering unusual or hard to get materials with historical backgrounds, or featuring the names of famous, long dead painters. The problem here is that there is no working manual of older painting practice written by an actual oil painter. While Pacheco's Arte de la Pintura has some interesting information attached to its highly cultured rhetoric, much of it is technically obscure, and some of the fundamentals are obviously unsound, as the condition of his paintings suggests. This lack of reliable written explanation makes sense since the training was hard won, often paid for, and made one's living in a fiercely competitive environment. What does exist is a great literary jumble sale of odds and ends across centuries of technique. Reading it all is difficult enough, exploring it adequately would take a large research team a lifetime. However, thirty years of research into older paintings at the molecular level, documented in the National Gallery Technical Bulletins, suggests strongly that the materials themselves tended to be simple: oil, pigment, perhaps ground stone, perhaps a small amount of egg or pine resin. This is logical because the last thing a working craftsperson could afford was technical failure. When Leonardo's great gamble, the Battle of the Anghiari wax-fresco, failed, he left Florence.
The way to avoid the complexity of the boutique manufacturers on the one hand, and the grand literary snipe hunt on the other, is to work with the materials at a fundamental level until you understand them well. This takes time, and patience, but is not rocket science. It is a matter of learning to pay attention at a different level. In America we turn out "Masters of Fine Art" in two years, but Chardin said it takes thirty years to make a painter. Which assessment of this situation is more accurate? The underlying problem is that our way of educating the brain relies on the left brain and linear thought. Not only is painting image or right-brain oriented, it is also multi-dimensional. All modern color models are three dimensional without the added dimensions of relative opacity -- transparent, translucent, opaque -- not to mention the effect of layering the colors over time. Without a willingness to undertake this equal and opposite form of education, it is simply not possible to access the more subtle ways color was used and paint was applied in older practice. Thus we often end up with painting which uses an increasingly superficial version of color, and an exaggerated version of style, in order to be heard above the din of all the other painting doing exactly the same thing. In his wonderful Oil Painting Techniques and Materials, Harold Speed says that a culture gets the art it deserves. Having abandoned all classical or ethical -- to say nothing of spiritual -- precepts about the purpose or meaning of creativity, it is no wonder that a kind of arbitrary commercial chaos has ensued.
If, on the other hand, you are interested in pursuing painting for its own sake, or, as the Taoists say, as a method of cultivation, you will find the real secrets where they have always been: hidden in plain sight. Ongoing experiential commitment to a working dialogue with the craft easily allows access to aspects of it that are overlooked by painters in a hurry, too intellectual or aesthetic to bother, or with fame and fortune as a primary goal. The materials can only explain themselves to someone who is paying attention. It is that simple, but also that complex.
collingwood & co.
In the past few years, partly as background research for Living Craft, partly because it turned out to be interesting in its own right, I read a great deal of writing about painting. Most of this has been wondrous, but, given our immersion in duality, some of it less so. Contemporary academics can write cleanly and purposefully, such as Laurinda Dixon's book on Bosch, weave a personal but fascinating tapestry of commentary in the manner of Diderot, such as Norman Bryson's Looking at the Overlooked, or create a variety of glass bead games that really have little to do with the subject, such as, well, never mind, let's just say that, if everyone who wrote about painting were required to know how to paint, a lot of folderol and tail-chasing could be avoided. Technical art history is more consistent because it is necessarily more grounded in the facts. The level of prose is high, people like Joyce Townsend or Leslie Carlyle tend to get a certain spin on the science. Mention can again be made here of the extraordinary book by Susie Nash about the Northern Renaissance, an introductory text that is basically a work of art in itself.
I ran across R.G. Collingwood while researching what Plato had actually said about art in The Republic. His way of writing was immediately sympathetic and I got a copy of the book the excerpt came from, The Principles of Art, first published in 1938. The aim of the book is to answer the question, "What is art?" Collingwood was a philosophy professor at Oxford, but, interestingly from the perspective of current academic writing, is not particularly devoted to large words. In fact, quite the reverse: he makes his points in simple declarative sentences. Now, if the subject is complex, reducing it to its essence is the ultimate challenge, but also the most fruitful approach in terms of testing the sense of the arguments at the most basic level. Similarly to Nash, The Principles of Art stands alone in terms of an author tackling and taming a complex subject with a minimum of waffle. This is especially amazing considering that Collingwood must cope with the infinitely murky waters of centuries of aesthetic theory. His genius is not to wade into them, but rather, part them, and walk right through, taking the reader along with him. If you have ever waded into aesthetic theory, (Elizabeth Prettejohn's Beauty & Art offers an accessible overview of the history of the subject and asks some interesting larger questions as well) you know that this is pretty miraculous. Collingwood accomplishes this by establishing a series of logical distinctions that serve to clear the air. He refuses, for example, to let art be defined by technique or style, distinguishing several categories of pseudo-art before ultimately exploring the concept of art as applied emotional integrity, and why this is culturally so vital. He does not attempt to separate art and life, but unite them, and, as a result of an extraordinary world view grounded in Greek philosophy, makes some incredibly prescient comments about art and the life of a society. As with T.J. Clark in Manet: The Painting of Modern Life, he understands that the Industrial Revolution involved capitol in the spiritual disenfranchisement of the lower classes, and goes so far as to say that the education of the English rural population in the later 19th century robbed it of it's sense of magical connection to the earth via its traditional culture. This from a professor at Oxford writing before Munich! But, unlike Clark he does not use this as a platform for an incipient Marxism, rather, he simply says this does not work, citing the way ancient Rome tried to permanently distract it's proletariat with bread and circus, only to crumble via a way of life that had no meaning for those who were living it. This is perhaps the most unnerving thing about this book from the current perspective, since, similar to Delacroix complaining that trains and clipper ships introduced too much speed for its own sake, the situation Collingwood found alarming in 1938 has multiplied exponentially since the time the book was written. This search for aesthetic definitions in the best Platonic tradition is highly recommended if the contemporary battlefields of style seem to entirely miss the point of what art is about.
The craft is a lot like a swimming pool: it has a shallow end and a deep end. Everyone has a place where they feel most comfortable, or are having the most fun. It's not really possible to assess someone else's experience if you're in a different part of the pool than they are. It's the same pool, but we all have a different experience in it. Sharing these differences is a big part of what life, and art, is about. A great deal of the technical information about painting available now is not only written from the shallow end, but assumes that the deep end does not exist. Or that, if it does, it is a "bad place," or inherently dangerous. But, as in life, the deep end is only dangerous for people who don't know how to swim. It is one thing to make things foolproof for beginners, or consensus consumption, but the real danger of oversimplification comes when the deeper end is demonized, or its existence denied, especially by someone who is calling themselves an authority. In this case, it is easy to begin to mistake a discussion of buoyancy and water molecules for swimming, when these are very different activities. The substitution of theory for practice is unfortunate, especially when it is not possible to even begin to understand older technique by reading the modern painting professors -- any or all of them. The craft teaches the painter how to swim, and naturally provides safe access to the deep end of the pool. Anyone who denies this always does so from the shallow end. I don't have judgments about the shallow end, it serves an important training purpose, many well-known painters from the 20th century never left it, and it is where most contemporary viewers feel comfortable. But art history itself proves it is by no means the entire pool.