The Red and (or is) the White

Are you sure you could always, even blindfolded, tell a white wine from a red? We tried out this, and some other manipulative tricks, on some experts. Ingrid Motteux reports.

The term chef de cave took on new meaning for a small Grape team, as we burrowed our way into a wine laboratory, aiming to produce an array of adulterated wines with which to confuse a wary panel of expert tasters. Deploying beakers, pipettes, filters, scales, a variety of (mostly legal) additives (including a plump green pepper, of course) and a case of finished base wines, we set about concocting some pretty believable (and also some pretty unbelievable) wines.

The next day we asked the panel to write down their impressions and comments and to score each wine on the 20-point scale. We had let them know we were attempting to catch them out – some or all of the 12 wines would be doctored in some way. Our aim was to observe how easy or difficult it would be for this bunch of professionals to be foxed by them. Their good nature in willingly taking part in this attempt to catch them out must be stressed.

A spate of newspaper articles not long ago, prompted by a well-reported study made by Frédéric Brochet in a Bordeaux study, claimed (smugly and rather too triumphantly) that wine ‘expertise’ is just a lot of linguistic hot air and snobbery, and the experts really knew little more about tasting wine than the lay person. The winning revelation was that not one of Brochet’s 54 experts had detected a tinted white wine masquerading as a red. Sacre Bleu! How could zees happen?

Actually, more easily than you think. If you’re one of those wine buffs who takes great pride in your ability to sniff out a molecule of bound sulphur at 500 paces, don’t assume it could never happen to you. After all, most grape pulp is colourless, and in adding skin-derived colourant anthocyanins to a white wine, one is merely including what a red skin would have provided anyway. If the base wine has been through malolactic fermentation and had a lick of oak, all the more challenging for the unsuspecting victim. Brochet claims, for example, that the molecule that causes tasters to perceive the flavour of blackcurrants and raspberries in a red wine is identical to that which imparts apricot or peach characters to a white. The description varies according to the colour. You taste what you’re expecting to taste.

After all these additions, what about spotting the subtractions? We included a pair of samples from the same barrel – identical wines except that one had undergone reverse osmosis (see box) because of its fairly high level of volatile acidity (essentially, vinegar). According to the deployers of this increasingly popular technological tool, in-house studies have shown no significant taste effect on wines that have been through the process. We asked our panel to pay particular attention to comparing the pair of wines, and found ... well, read on.

Wine 1: Peachy White
Recipe: one third off-the-shelf chardonnay peach cooler plus straight white wine.
This shameless blend started our panel off on a positive note, confident that they had spotted a fake. All commented on it being confected and artificial or manipulated; no mention, though, of low alcohol (approx 9%). Average score: 11

Wine 2: Spier Sauvignon Blanc 2001
Undoctored. Most thought this wine ‘natural’ but dull, even with its obvious sauvignon character. Score: 12.5

Wine 3: Green-peppered chenin
Recipe: The pepper was liquidised with some base wine, and a small amount was added to a bottle of chenin blanc. (This trick has, in other experiments, caught out some of the most highly regarded local pros – no names may be mentioned.) ‘Quite good sauvignon blanc if it’s Stellenbosch; a fraud if it’s from Robertson’ was one comment. All but one taster thought it really was sauvignon. The most plausible of the fakes. Score: 12.5

Wine 4: Red chardonnay
Recipe: A wooded chardonnay (2000 vintage; four stars in Platter) was transformed by anthocyanin into a deep and believable red. (An earlier attempt at using cochineal for colouring failed miserably – a terrible strumpet colour, some of which stuck to the side of the glass.)
This effort fared poorly, and comments included ‘unidentifiable red fruit’, ‘mulberry jam’. and even ‘black pepper’. All tasters found the wine a bit odd. One thought it a light red with added tannin and colour, and another identified it as a white wine (though this taster also claimed two true reds were white). In retrospect, it would have been interesting to also put the original chardonnay in the tasting, to test Brochet’s point about descriptions being colour-dependent. Score: 9.5

Wine 5: Rouge Blanc
Recipe: Oak essence; anthocyanin for colouring; red wine (20%); glycerol; a pinch of one of the tannins used in quite a few local reds, on a well-established Australian model. Added to a mix of pretty ornery chenin and sauvignon blanc, and laboriously filtered, leaving it brighter than the red chardonnay.
This work of vinicultural art was rated half a point higher than the chardonnay. Though its appearance was more natural, the acidity was searing and the wine did not hang together at all well. All tasters identified it as a ‘fake’. Where fruit was found, it was described as raspberry and cherry, and the oak essence was subtle enough to avoid more than a passing comment. Score: 10

Wine 6: Diemersfontein Pinotage 2001
Perhaps because tasters were a bit wary by this stage, this Absa Top Ten wine of 2001 scored a low 11. It was found over-alcoholic, jammy and spritzy, with a bitter finish. One taster thought it was doctored.

Wines 7 and 8: Treated cabernet
Wine 7 was a decent cabernet, with an only-just-legal level of volatile acidity. Wine 8 was the same wine which had gone through a reverse osmosis (RO) machine to lower the level of volatility. The wines were markedly different. The untouched sample was slightly darker; all tasters found the post-RO wine excessively bitter and tannic, with various uncomplimentary descriptions (waxy, bubblegum, floor polish, wax crayon, Playdo) – it scored terribly, although the pre-RO wine was the tasting’s second-highest scorer at 13.5 (and no-one complained of volatility). A very different story to the claims of the RO operators.

Wine 9: Blend X control sample
A ‘natural’ blend of three red wines, also used as a base for wines 10 and 11.
All tasters agreed that this was a young wine of good quality, not obviously manipulated. (In fact, the blenders of the wine, who knew its components, thought their random blend very much better than any of the wines by themselves, especially after a night’s aeration!) Score: 15 (highest of the tasting)

Wine 10: Adulterated blend X (1)
Recipe: 3 grams of castor sugar added to a bottle of the blended base wine.
Strangely different from the control sample – in the lab the day before it had markedly improved what was then a rather harsh red wine. The wine was found to have markedly reduced aromatics (shy, muted, dumb, inert) compared to the control, but increased (‘cherry’) juiciness on the palate. Score: 13

Wine 11: Adulterated blend X (2)
Recipe: added cassis liqueur, arabic gum and glycerol.
Judging by the score being lower than the control’s, this further proves that additives do not necessarily improve a wine (although the failure may have more to do with our lack of lab-skills.) The tannins seemed more obvious, and although found to be rich, resinous and fuller bodied, it was well described as ‘all front-palate with no finish’. The additions probably nudged the wine into this state of disharmony. Score: 13.5.

Wine 12: Leasingham Classic Clare 1996
Undoctored Australian shiraz.
A couple of tasters had the impression that this wine had added sugar and essence. It was opulent, perfumed, rich and overripe. Rod Easthope offered ‘ripe, eucalypt, mint and warm alcohol’, and Angela Lloyd hit the nail on the head, describing it as a ‘popular Ozzie style’. Score: 14

Some tentative conclusions

Rod Easthope commented that if any of these wines (bar the first) were placed in a competition line-up of 100-odd wines, none would be picked up as strongly ‘fake’. In fact we could probably also conclude that at least some of them might expect to score rather better than they did here, where tasters were probably rather guarded with their generosity, knowing of the possibilities.

Legal manipulation of wine, with legal processes and additives (tannins, anthocyanins, etc), is widespread, throughout the world – perhaps particularly in the New World. Many commentators would say that illegal manipulation also happens (everywhere) to some or other extent. If so, it would be done more cleverly than our amateur trickery. Ours was an admittedly amateur attempt (though we did have some strictly anonymous professional help), and perhaps this is why none of the wines were greatly improved by our manipulations. Done more judiciously, in the hands of an experienced and knowledgeable (in this field!) winemaker, with additions made more carefully, and at specific stages in the winemaking process, no doubt the deliberate enhancement of the resulting wine would be maximised.

How many manipulated wines do we drink, and even decorate with gold and silver? Perhaps more than we realise.

A NOTE: REVERSE OSMOSIS
This process (which can separate out  various components of wine, such as water and alcohol) is increasingly widely used throughout the world, for various purposes.  The three most important of these are: removing water from grape must (to concentrate it), and removing alcohol or volatile acidity from a finished wine. The last of these is probably the most widespread use to which reverse osmosis machines are put in South Africa. It is an expensive process – but volatile acidity seems to be an increasing problem for various reasons, and the machines are reportedly extremely busy.

Tasters
Margaret Barker, Cape Wine Master
Greg de Bruyn, Cape Wine Master
Rod Easthope, winemaker and judge
Angela Lloyd, wine writer and judge
Jean-Pierre Rossouw, wine writer