Skip to content

What can rotten tomatoes tell us about moving pastels?

28 December 2015

FlavellaWell, we might want to throw them at the author (John Hawkesworth) of this drivel which appeared in the Gentleman’s Magazine as the Poetical Essay for May 1750 (I have spared you the lengthy text). But it serves to illustrate how long it has been known that moving pastels is risky. I’ve written about this in far greater detail (and with copious references to the literature) in Chapter V of my Prolegomena, but it may be worth summarising the key points of the debate between the risk-deniers and the nay-sayers. We are united in wanting to find the answer.

Everyone knows that pastel is fragile and that moving it is difficult. 2015 has seen a number of international pastel exhibitions (Karoline Luise in Karlsruhe, Vigée Le Brun in Paris, soon to move to North America, and Liotard in Edinburgh and London come to mind) which, to judge from press reports, may have left some people thinking that the problems have been solved – although in fact pastels have been moving to international exhibitions for many years.

As I recount in the Prolegomena, an important first step in the scientific research arose with the 1989 Degas exhibition in Liverpool and Glasgow where significant concerns were raised. An international association of most major museums concluded in 1995 that “unfixed pastels are usually too fragile to travel.” But by 2004, when a number of exhibitions were held to celebrate La Tour’s tercentenary, pastels were on the move again; only for another report to conclude that all loans of pastels should be prohibited, without exception. That remains (with very rare exceptions) the policy of the Louvre and the museums with major pastel collections in Saint-Quentin, Amsterdam, Dresden and Geneva. As it happens it is also my policy since 2004 (but not before). But not everyone agrees, and I can see that there is a defensible position to take in arguing that some exhibitions are so important that the risks – maybe even the certainty of some minute level of damage – are worthwhile. But that’s a somewhat different position than suggesting that the problems have been solved. (And I am of course aware of the lengths to which organisers go to minimise risks.)

A great deal of research has gone into the logistics of moving art, but I want to give you an idea of what the issues are for pastel and how far we are from having comprehensive answers or a general consensus. Specifically I want to highlight some counterintuitive results from the published research. Do please contact me if you are aware of any relevant work I haven’t covered (in the Prolegomena – this blog post is only an informal summary).

Perhaps the most obvious point is that art can be damaged, even by surprisingly small levels of shock or vibration. When the British Museum carried out construction work for the Great Court development, careful monitoring found that vibration could be transmitted far further than the standard decay model predicted. And it also found that objects with painted surfaces could lose pigment with shock levels as low as 0.2g – far lower than anyone had imagined. Admittedly these had pre-existing weaknesses – but of course it is the nature of pastel to have what insurers call “inherent vice” (which means that the question of shipping them safely may have a financial angle for owners as well as the purely artistic question in which we all have a share). You simply can’t lift a pastel off the wall and package it however carefully without exceeding that level.

But of course you will say that this is nonsense: if that were true, no eighteenth century pastel would survive today. And (if you are in the trade) you ship them all the time, and no damage is ever visible…well, of course, there was that case… – but that was in poor condition. And better not raise…

So yes: pastels can suffer a higher level of shock than 0.2g with no visible damage – sometimes (maybe even most times). (Those that can’t are already ruined, and probably disposed of to release the frame. A great many eighteenth century pastels have vanished.) And of course there are the cases no one wants to talk about (me neither: I’d win no favours for identifying specific works as seriously compromised – so there are no photographs in this post): there is a built-in cognitive error compounded by an anti-disclosure bias surrounding the information.

But also the fear is that you can’t tell from the appearance what’s going on underneath. Damage may not be reported simply because it isn’t visible. Your picture comes back from the exhibition: it looks a little strange, perhaps, but when you compare it with the photograph you can’t spot any area of fallen pigment, so you must just be mistaken. Better not make a fuss: no insurer will pay up.

Here’s where the fruit come in. The great thing about a bruised peach (a real one, not a Liotard nature morte) is that it will reveal its hidden abuse fairly rapidly. And the supermarkets have invested a vast amount of money in research in how to measure and minimise shock and vibration in the transport of perishable goods. So we know for example that vehicles with air-ride suspension are smoother than those with traditional leaf-spring suspension (unless the air-ride mechanism isn’t working properly, in which case the ride is worse: the compressors used to drive the suspension can themselves cause vibration). But we also know that the grapes at the top of the box fare worse than those at the bottom, even though the latter suffer far greater compression: so repeated small impacts can be worse than fewer stronger ones.

We also know that the shock and vibration behaviour of different forms of transport is counterintuitive. To take an example where opinions remain strongly divided, most studies show that, at least as regards measurable vibration, road freight is astonishingly far worse than air (and marginally worse than rail, but the rail figures aren’t consistent). But the big problem with air travel is that you aren’t allowed simply to place your packing case directly in the hold: it has to go through the cargo handling areas, where it will drop off the end of a chute, experiencing shock levels in the range 6–10g (and much higher if it is a case small enough to be lifted by hand, when dropping is highly probable). (Of course good packaging will reduce the impact on the pastel itself, perhaps by 60%, maybe more.) And that’s not to mention the shocks on the “dollies” which carry the boxes out over the tarmac – and this to owners who carry out experiments in which the difference between pneumatic and solid rubber tyres on trollies used to move pastels within a museum is thought significant.

So if you have to choose between air and road transport, you have to balance the risk of say half a dozen shocks of 8g with say 50 hours of vibration at a mean acceleration of perhaps 0.5g at a frequency of say 10Hz, i.e. occurring ten times a second. Obviously distance matters: the hazard from road transport is proportional to it, while if vibration during the flight is as low as suggested, “ce n’est que le premier pas qui coûte.” But how do you calculate the trade-off?

Fortunately mechanical engineering again has a methodology. Here is a so-called S-N curve, in which you plot on the y-axis the stress which causes failure for a specific number of cycles of a repeated shock. If there is just a single or a few cycles (e.g. at N1), the stress required will be high; but repeating a lower stress many times may also cause failure (e.g. at N2): hence the general shape of the curve.

SN curve

The key question for any material is how the long tail behaves. Does it level out above zero stress, so that if the stress is below that level (the so-called fatigue limit), it doesn’t matter how often it is repeated? And if so, are you sure that the fatigue limit itself will not be exceeded?

This is how many engineering problems (notably in air crash analysis) are approached. They are dealing typically with simple materials such as homogeneous metal alloys. The same methodology is being applied in painting conservation, and to pastel by a group currently working in the Rijksmuseum, but as you can easily imagine the work is immeasurably more complicated. We don’t yet know if there is a fatigue limit for pastels (nor whether they are really like wings or pears), and so we can’t really answer the “road or air?” question. (Except by: no.)

A good deal of information has been gleaned from mounting accelerometers in different locations in various types of lorry. (I’ve also done this in my car, which I can report produces figures considerably better than air-ride lorries: but I’m not offering a competing service.)  We can identify which types of packaging minimise certain frequencies of vibration – and note that some unexpected resonances can occur which mean that solving one problem may create another. We can establish (and most people agree) that for road transport the vertical shocks will be higher than the horizontal ones (either in the direction of travel, or from side to side). But does it follow that a pastel is best travelling flat? Most (but not all) museums think so; but I’m not sure there is much evidence comparing the forces required to lift pigment from a pastel shearing it sideways rather than propelling it at right angles to the surface (one paper that investigated this used modern samples that were not mounted on canvas, a vital part of the problem). And horizontal travel for any support that has lost tension increases the risk of the pastel touching the glass. And so on.

What is clearly important is for more information to be gathered and shared openly. If someone comes up with a magic carpet, let us gain consensus for its use. If a practicable fatigue limit for all pastels against all risks can be established, it will only be generally accepted when the research is published in scientific journals. Until then few will want their original eighteenth century examples tested to destruction; but even when that is done, my fear is that we will find that no two pastels are the same. All metal alloy specimens resemble one another, but each pastel is inherently vicious in its own way.


From → Art history

One Comment

Trackbacks & Pingbacks

  1. Reviewing the Liotard reviews… | Neil Jeffares

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: