Composite Names Three Different Objects, and Only One of Them Is a Fake

One word is carrying three claims that do not follow from each other

Composite, in a catalogue entry, can mean that an object was built out of many parts, that it acquired its parts at different dates, or that somebody put old parts together to sell a thing that never existed. Those are three separate findings and only the third is a reason to doubt anybody.

The vocabulary is slippery in a specific direction. A reader who meets composite in the entry for a reliquary and remembers having met it in the entry for a discredited antiquity will import the suspicion, and the entry gives no signal to stop that. The house habit of separating words that get conflated applies here as much as it does to the five things the word signed can mean.

Fifty-seven records here carry one of these words. Sorting them gives three groups of very different sizes and very different consequences, and the test that separates them is not technical. It is a question about when the parts came together and whether anybody was meant to be deceived.

Assembled can be a fact about manufacture and nothing else

The Ardagh Chalice is a two-handled beaten and spun silver vessel assembled from 354 separate pieces, made in Ireland in the eighth century and commonly dated to about 750. Every one of those parts was made for this object, in one campaign, by people working to a single design. Composite here is a statement about ambition.

The detail that settles it is on the underside. The decorated foot carries a polished rock crystal at its centre, and the surface a congregation kneeling in front of the altar would never see is finished to full standard. Nobody assembling parts to deceive a buyer finishes the bottom.

This is the largest of the three groups and the least interesting to a sceptic, which is exactly why it should not share a word with the other two. An eighth-century chalice of 354 parts, a Renaissance nef and a sixteenth-century automaton are all composite in this sense, and in none of them does the word bear on authenticity at all.

The Louvre dates one sword to four different periods, and says so itself

The second group is objects that grew. The Epee du sacre des rois de France, MS 84, has been used at French coronations since at least 1271 and is traditionally identified as the sword of Charlemagne, who died in 814. The Louvre's own collections database dates its pommel to 900-1100, its quillons to 1100-1200, its grip to 1200-1400 and its blade to 1700-1900, and states that the sword is composed of very diverse elements, much modified, particularly as to the blade.

Not one of those four ranges includes Charlemagne's lifetime. The object is nonetheless completely authentic: it is a real coronation sword with a continuous documented function, and the scabbard carries velvet remade in 1804 and again in 1825 because it was still being used. What the component dates destroy is a claim about physical continuity with one man, and the Louvre publishes the evidence against that claim in its own record.

This is the case that shows why a composite needs its dating rule stated. If the object is dated by its latest part it is an eighteenth or nineteenth-century sword. If by its earliest, it is tenth-century. Neither number describes the thing, and the honest entry gives all four and names which part each belongs to.

Museum armour is assembled, and the museum did the assembling

Armour is the field where the second group is standard practice rather than an exception, and where the assembler is often the institution. The Metropolitan's garniture of George Clifford, third Earl of Cumberland, is dated 1586 and made under Jacob Halder's direction at the royal Almain Armoury at Greenwich, on evidence as good as this field offers: Halder annotated two of his own designs in the workshop's record book with the words that they were made by him, and the corresponding design for this exact garniture survives.

The object on display is still not what left Greenwich. The garniture as now shown was assembled by the Museum from two separate acquisitions, made across two different funds four years apart, rather than received complete in one transaction, and its left gauntlet is not original to the 1586 commission. The Museum's own photographs of the armour of Henry VIII, dated 1527 in three places and accepted as the earliest dated product of the same workshop, likewise carry an associated piece that is not from that harness.

A visitor sees one suit of armour. The record describes a display, and the difference between those two things is a hundred per cent of the question a specialist would ask.

Pastiche adds an inference about who intended what

The third group is where the word stops describing and starts alleging. The Tiara of Saitaphernes was sold to the Louvre in 1896 as a Greek gold tiara presented by the city of Olbia to a Scythian king of the third or second century BCE. It was made in Odessa in 1894 or 1895 by the goldsmith Israel Rouchomovsky, who came forward in Paris in 1903 and proved his authorship by reproducing a section of the work in front of a commission.

What he had produced is described in this archive as a competent and learned pastiche, assembled from the visual vocabulary of Black Sea Greek gold that was being excavated and published in his own lifetime, in a city where that material was arriving. The subject matter was assembled from published excavation reports supplied to him by the dealers who commissioned the piece. It is seventeen centimetres high and eighteen across, worked by hammering, repousse and chasing and put together by soldering, and the Louvre's catalogue now gives Rouchomovsky as its author.

Note where the deception sits. Rouchomovsky made an object and did not sell it as ancient; the dealers who briefed him did that. Pastiche is accurate about the object and silent about which person intended the fraud, which is the one thing a reader most wants to know.

The Rospigliosi Cup shows what a museum does when it stops guessing

Four positions in a hundred and thirteen years is the record of this object. The cup was catalogued as Benvenuto Cellini until 1969, reattributed by Yvonne Hackenbroch to Jacopo Bilivert in that year, reassigned by the same scholar to the Aachen goldsmith Reinhold Vasters in her Metropolitan Museum Journal monograph of 1984-85, and is now carried by the Museum with no maker at all.

The Museum's current record is worth reading as a piece of writing. The title is Cup. The date is probably about 1840 to 1850, or somewhat earlier. The medium is gold, partly enamelled, with pearls, the object is 19.7 by 21.6 by 22.9 cm, the accession number is 14.40.667 and the credit line is the Bequest of Benjamin Altman, 1913.

That entry contains no allegation against anybody, names no forger and asserts nothing it cannot support. It also tells a reader everything material: the object is a nineteenth-century thing that was bought and bequeathed as a sixteenth-century one. Withdrawing a maker is a finding, and it is a stronger one than the three confident attributions that preceded it.

The strongest word was published by a museum against its own object, and it was wrong

In February 1968 the Metropolitan Museum of Art Bulletin carried an article by the Museum's Vice-Director, Joseph V. Noble, titled The Forgery of Our Greek Bronze Horse. The object had been bought in 1923 and shown for more than forty years as Early Classical, of about 480 to 470 BCE. It was condemned in December 1967.

A technical re-examination lasting thirteen years reversed that. The bronze alloy, the corrosion products, the ceramic core with its iron armature and chaplets and the repair patches are all consistent with casting in antiquity, and thermoluminescence dating of the core returned 2250 plus or minus 210 years. The Museum now catalogues the statuette as Greek, late second to first century BCE, bronze, 40.2 cm high, accession 23.69, Fletcher Fund, 1923.

Two things follow. The first is that the allegation was withdrawn. The second is less often noticed: the object is genuine and the date it was bought as is gone, because the work that rescued it also moved it about four hundred years later than anyone had claimed. A composite argument was made, refuted, and in refuting it the museum lost the thing it had actually been proud of.

Three meanings, and the question that separates them

The three groups can be told apart by asking when the parts came together and whether the join was meant to mislead. Nothing technical is required to ask it, and the answer changes what the entry is reporting.

Four of these five are ordinary facts about objects. The fifth is a charge, and the bronze horse is the standing proof that a museum can bring it against its own holding and lose.

How to read a composite entry

Ask which part the date belongs to. An object described as composite and given one date is under-catalogued, and the Louvre's four ranges for one sword are the standard to hold other records against. Where a single date appears, assume it is the latest part until the entry says otherwise.

Ask who did the assembling and when, and treat the museum as a candidate. A display assembled from two acquisitions is not a fault, and a record that does not disclose it is.

Then read the word itself as evidence about the cataloguer. Composite describes, pastiche infers an intention and forgery alleges one, and a catalogue that reaches for the strongest available word is telling you about the confidence in the building as much as about the object. In February 1968 that confidence was total, published and wrong.

Land the point

An eighth-century chalice of 354 parts, a coronation sword built up across a millennium, a display armour put together by its own museum, a learned fake made in Odessa and an ancient bronze once denounced by the institution that owns it are all called composite somewhere in the literature. Four of them are exactly what they claim to be.

The remedy is not a new word. It is to say when the parts came together, on whose authority, and whether anybody was meant to be misled, in the entry, every time. A catalogue that does that can use composite freely, because the sentence beside it will carry the meaning the word cannot.

Why this archive is free to read

We do this for everyone who loves art. The people who have studied it for years, and the people who just stopped to look. Iconotheca exists so that what we learn belongs to all of them.

Nobody pays us for this. No ads, no sponsors, no commercial interest. These works belong to all of us, and we would rather share them with as many people as we can than keep them to ourselves.

If it gave you something today, tell us to keep going. Follow us, leave a like, or write a positive comment. We read every one, and they are what keeps us going.