How to Spot Fake Wine: A Bottle Inspection Checklist
Updated
Learning how to spot fake wine starts with a question no forger can answer: did the estate ever make this wine, in this vintage, in this size of bottle? Every famous counterfeit came apart on that arithmetic rather than on taste. Rudy Kurniawan offered eight magnums of 1947 Château Lafleur across his two 2006 auctions, and David Molyneux-Berry, the former head of the wine department at Sotheby's, told the 2007 TASTE3 conference in Napa that only five were ever produced. Laurent Ponsot stopped a 2008 consignment of Clos Saint-Denis carrying vintages from 1945 through 1971, because Domaine Ponsot had never made a Clos Saint-Denis before 1982. Lead prosecutor Jason Hernandez argued that six bottles of Domaine Georges Roumier Bonnes Mares 1923, sold for $95,000 at a 2006 auction, had to be fake because Roumier produced no wine before 1924. Check the production record first. A bottle that fails it is finished before anyone opens it.
Does the bottle pass the production record?
Ask four things before you look at the glass. Did that producer make that wine? From that vineyard? In that vintage? In that format? Each one has ended a real sale.
The Ponsot case is the cleanest. Kurniawan consigned bottles attributed to Domaine Ponsot from the Clos Saint-Denis Grand cru with vintages ranging from 1945 through 1971, and Laurent Ponsot told the auction house the domaine had made nothing from that appellation before 1982. The lots came out. In April 2007 he consigned several magnums of 1982 Château Le Pin at Christie's in Los Angeles, prominent enough to sit on the catalogue cover, and representatives of the château contacted the house to say the bottles were fake. Christie's withdrew the lot after review.
The same test caught Hardy Rodenstock from the other direction. David Peppercorn and Serena Sutcliffe questioned six-litre Imperials of Château Pétrus from the 1921, 1924, 1926, 1928 and 1934 vintages, poured from Rodenstock's collection at tasting events in 1989 and 1990. Their concerns were never proven and Rodenstock disputed them, but Christian Moueix, who manages Château Pétrus, confirmed the estate holds no records of producing Imperials in those vintages.
None of that required a laboratory. It required someone to call the estate, which remains the cheapest authentication step available and the one most buyers skip.
Why is bottle format such a fast tell?
Because fakes concentrate where the real population is small enough to count. Eight magnums against five produced is the Lafleur case in a single line, and nobody needed to open a bottle to see it.
Format also shows up as a statistic. In a March 2010 lawsuit filed against Christie's, Bill Koch alleged that Rodenstock had distributed 818 bottles in the United States through Daniel Oliveros and Jeff Sokolin of Royal Wine Merchants between 1998 and 2008, virtually all rarities, of which 87% were magnums. Oliveros and Sokolin deny selling fake wine. In May 2012 the U.S. Court of Appeals for the Second Circuit ruled the claims against Christie's were time-barred and affirmed the dismissal.
Treat that 87% as a screening rule rather than a verdict. When a consignment leans towards large formats of scarce old vintages, the pool you are drawing from is usually small enough that the estate can tell you exactly how many exist.
What should the seller be able to tell you about provenance?
Who owned it, where it sat, and how it reached them. If the seller cannot answer, you are buying the story rather than the wine, which is the argument for treating provenance as a priced attribute.
The Jefferson bottles are the case study. Rodenstock said he received a telephone call in early 1985 about old bottles found in a walled-up cellar in Paris, engraved with late eighteenth century vintages and the initials "Th. J.". On 5 December 1985 one of them, a 1787 Château Lafite engraved "1787 Lafitte Th. J.", went up at Christie's in London with its catalogue value listed as "inestimable" and sold for 105,000 pound sterling, which as of 2007 remained the worldwide auction record for a single bottle.
What never appeared was the chain behind it. Rodenstock never revealed the name of the person who sold him the bottles, never gave the Paris address, and never stated how many were found, with figures quoted over the years ranging from "a dozen or so" to thirty. When the German magazine Stern offered in 2007 to have his remaining bottles analysed at its own expense by the Federal Institute for Materials Research and Testing, he declined. A German court had already found in 1992 that he had "knowingly offered adulterated wine" for sale, a case settled out of court on appeal.
Bill Koch only started asking because the Boston Museum of Fine Arts wanted provenance for an exhibition in 2005. His staff found nothing beyond Michael Broadbent's authentication, so they contacted the Thomas Jefferson Foundation at Monticello, whose curator answered that on Jefferson's own records the foundation did not believe the bottles had been in his possession. Tracing back through the Chicago Wine Company and Farr Vintners showed all four of Koch's bottles had originated with Rodenstock. He later claimed evidence that nine more bottles in his possession, dated from 1737 to 1936, were also fakes or "highly suspect".
What does the physical bottle give away?
Period detail that the forger had to manufacture with modern tools. Koch's investigators concluded that the Jefferson engravings had been cut with a high-speed electric drill, similar to a dentist's drill, using technology that did not exist in the eighteenth century. Benjamin Wallace's 2008 book The Billionaire's Vinegar reported a tritium test and two carbon-14 tests dating the wine circa 1962.
Label fraud done well requires the bottles, corks and packaging to be manipulated to match, which is exactly what the FBI found when it arrested Kurniawan at his home in Arcadia, California on 8 March 2012. Agents recovered inexpensive Napa wines annotated to show they would be passed off as older vintages of Bordeaux, along with corks, stamps and labels. Later investigation established the method: buy old but cheap Burgundy, then relabel it with a prestigious producer name and vintage.
The cruder end of the trade uses the same components. Methods used to dupe consumers include photocopying labels and creating phony château names on the capsule and the label, and sometimes an authentic bottle is kept and another wine is added with a syringe. Christie's has taken to smashing empty bottles with a hammer to stop them re-entering the black market, and at one 2008 sale a French vintner was "shocked to discover that '106 bottles out of 107' were fakes." That is also why empty fine wine bottles have a resale market worth knowing about.
Who is consigning it, and does the price make sense?
The consignor is evidence. In February 2012 Spectrum Wine Auctions withdrew several lots worth an estimated $785,000 from a London sale when allegations emerged that they had been consigned by Kurniawan through a second party.
Spending is evidence too. By 2006 Kurniawan was buying auction lots at as much as $1 million a month, and was described at the time as possessing "arguably the greatest cellar on Earth." That reputation is what let him sell. Reading what a house does and does not assert about a lot is a skill in itself, covered in how to read a wine auction catalogue.
The checklist
| Check | Question to ask | What it caught |
|---|---|---|
| Production record | Did the estate make this wine at all? | Clos Saint-Denis vintages from 1945, none made before 1982 |
| Vintage | Was the estate producing that year? | Roumier Bonnes Mares 1923, no wine made before 1924 |
| Format | Were bottles of this size ever filled? | Château Pétrus Imperials, no estate records for those vintages |
| Count | How many exist against how many are offered? | Eight magnums of 1947 Château Lafleur against five produced |
| Chain of custody | Can the seller name the previous owner? | Jefferson bottles: no seller named, no address, no count |
| Physical bottle | Do glass, engraving, cork and label suit the period? | Jefferson engravings cut with a modern high-speed drill |
| Consignor | Whose wine is this, and through how many hands? | $785,000 of lots withdrawn in London in February 2012 |
How much fake wine is actually out there?
More than the famous names account for. Wine Spectator has noted that as much as 5% of the wine sold in secondary markets could be counterfeit. The 2016 documentary Sour Grapes states that as many as 10,000 counterfeit bottles created by Kurniawan may still sit in private collections.
It is not confined to trophy Burgundy either. In 2002, bottles of the weaker 1991 vintage of Château Lafite Rothschild were relabelled and sold as the acclaimed 1982 in China. In 2000, Italian authorities uncovered a warehouse holding nearly twenty thousand bottles of fake 1995 Sassicaia. And in 2022 a Europol and Interpol operation called OPSON XI, led by the European Anti-Fraud Office, seized nearly 14.8 million litres of illicit drinks along with counterfeit bottles, packaging and equipment for making sparkling wine.
What is being done about it?
Producers are marking bottles with engraved serial numbers on the glass and taking more control of distribution, and for older vintages laboratory methods such as stable isotope analysis and isoscapes are likely to grow in importance. France set up the Service de la Répression des Fraudes to police fraud among its AOC wines, and appellation systems across Europe exist in part to stop a wine claiming an origin it does not have.
None of that helps a bottle already sitting in a warehouse with an unverifiable history. What helps is the sequence above, applied before the money moves. For the full case-by-case account of how each fraud was exposed, read famous wine counterfeit cases, and for what the newer tools verify, see wine authentication technology. If you are buying Burgundy, the Burgundy map is a quick way to place a producer and an appellation before you trust a label.
Join the newsletter for the 40/60 score on every wine, so you can see how a bottle is rated and what it should be worth before you decide whether its story holds up. Track where the money is moving on the fine wine market index.