The Big Question: How many of the paintings in our public museums are fakes?

Why are we asking this now?

The National Gallery is just about to mount a major exhibition about faking in art. It reveals in the course of the exhibition that the gallery itself has been duped again and again in the past. A work said to be by Botticelli, for example, one of a pair purchased in the third quarter of the 19th century, was later discovered to be the work of a pasticheur, painted in the style of Botticelli. Works by other old masters have been proven to be by studio assistants, friends of the artists, or even by fakers who may have lived hundreds of years later.

Is it easy to fake a masterpiece?

Not as difficult as you might think, though some would be more difficult than others. A faker should concentrate upon the present – or the relatively recent past. Tom Keating, the most famous forger of our times, did precisely that.

It is, for example, much easier to replicate a work by Picasso than a major work by a painter of the Renaissance. Picasso is much less fiddly. What is more, he often worked very quickly, painting a single painting – or even several paintings – in a day. His paintings are relatively easy to characterise and to caricature. Technically, many of the 10,000 or so that he produced during his inordinately long life would be relatively easy to replicate. They seldom get bogged down in onerous detail. What is more, the kinds of paints he used – or their equivalents – are still widely available today.

Does the gallery world's eagerness to promote culture as a leisure pursuit encourage fakery?

Undoubtedly. And especially so in America. This is where forgery and faking begins to merge with, and meld into, the world of restoration almost seamlessly. There are few things more dazzling than a well restored painting, and the Met in New York and the National Gallery in Washington are full of them.

That is how people want to see them. The public, having paid its money, expects perfection. So dingy old paintings, little by little, are re-painted, with painstaking lovingness. It happens all the time. By the time in 1994 when the Sistine Chapel was restored to the condition in which it would have been when Michelangelo painted it, was it still by Michelangelo? Or is it by now a modern fake – or as near as damn it? This is a moot point. The Chinese are much more honest about these things. If you go to the Temple of Heaven in Beijing, you will find that the plaque on the wall of this splendid wooden structure will say: first constructed in 1420. Why pretend otherwise?

Why does it matter if a painting is a fake?

It matters hugely to the institution who owns it. In the case of a painting newly re-attributed to Michelangelo at the National Gallery in Washington – Saint John the Baptist Bearing Witness – it matters the difference between £400,000 and £150m. That is the difference between its value when attributed to Michelangelo and its value when attributed to his slightly older friend, Francesco Granacci.

And why did this mighty leap in value suddenly occur? Because Everett Fahy, former head of European Paintings at the Met, had a "eureka moment" as he put it, when he looked at the painting again. It's as simple as that.

Does it matter so much to the average gallery-goer?

Much less so. For the most part, we know most of the paintings that we love by looking at them in reproduction. It is images of these reproductions that we carry around in our heads, and if being face to face with the originals gives us a qualitatively different experience, all to the good. What is more, the very fact that a painting may have been restored and re-varnished repeatedly, means that the authentic historical object has withdrawn from us. What we see is an object willed into renewed life by loving restorers.

So are there many fakes in our museums?

This is too crude and unseemly a question; better to ask the same question in a more nuanced way. As the National Gallery's forthcoming show will demonstrate, many works were bought because they were, at the point of acquisition, believed by the harumphing experts of the day to be by painter X, without a doubt my dear boy.

When the next generation's experts come along, they may be inclined to disagree. So the real question is this. Of the paintings which are owned by, say, the National Gallery, how many of them are no longer by the person they were first thought to be by? What is more, it will be easy for a major gallery to pull the wool over our eyes because most of these paintings of doubtful attribution will now be in store, so they will not be on the walls of the gallery for us to scrutinise them.

But exactly how many are fake?

A reasonable estimate might be that at least 20 per cent of the paintings held by our major museums, some up on the walls, many others in the vaults, will no longer be attributed to the same painter 100 years from now. Another matter worth bearing in mind is this: the less money the institution has to spend on restoring its paintings, the more likely it is that the paintings will be in a condition which resembles how they would once have looked. You see proof of this in many of our smaller provincial museums. Suddenly you find a dingy Blake. It seems barely worth a second look. Do look again though. This may be Blake untouched and neglected, as authentic a specimen as you are ever likely to see.

Is faking on the increase?

Undoubtedly. It is thriving, especially in the field of architecture. But it is not called by the gross name of faking. Go to St Petersburg or to Dresden where great buildings were destroyed during the Second World War, buildings which had helped to define a nation – the Frauenkirche in Dresden, that great Baroque church which was rebuilt stone by stone; or Catherine the Great's palace, with its newly recreated Amber Room, just outside St Petersburg. They look so brash and so brazen, these monstrous, gleaming edifices, that they must be fakes. And so they are. But in a sense they are not. They are also the past, brought back to miraculous life, a past that a populace is desperate to rescue from oblivion and the terrible depredations of war.

Is it possible to limit fakery?

Yes...

* Museums should be more stringent in their attribution policies

* If in any doubt, don't buy. That may discourage pedlars of forgeries

* Educate the public to buy new art and to hate copies and the problem will be solved

No...

* Fakery has been an easy way to earn money through the ages. It will go on being so

* There will always be a demand for copies of masterpieces, whether regarded as fakes or not

* The copying of masterpieces has long been regarded as a form of apprenticeship