“Saint-Emilion classification
Hubert Boidron’s defense at the International Wine Academy’s winter symposium. December 5, 2013.
Mr. Chancellor, Mr. Chairman, dear colleague, dear friends,
Hello,
I’m Hubert Boidron, oenologist and head of production at Maison Boidron, a winegrower in the Saint-Emilion region. It has always been an honor to be among you. Today I’m proud to finally be one of you. Many thanks to all of you. From an early age, I spent my weekends following my father into the vineyards. I was seven years old when he taught me how to distinguish a cochylis from an eudemis and that his black hands were not dirty, but tinged with anthocyanins, the mark of nobility of winegrowers. Today, I’m going to give you a very short presentation on the “Saint-Emilion classification”.
Historical background.
In 1841, Monsieur Lecoutre de Beauvais, editor-in-chief of the newspaper “Le Producteur”, established a classification of Saint-Emilion wines into three classes:
First class: Belair, Fontplégade, Canon, Couspode, Balestard, Pavie, Larcis, Mondot, Ausone, Beauséjour, Berliquet, Magdelaine, Soutard.
Second class: Cheval Blanc, Figeac.
Third class: Corbin, Jean Faure.
This classification was probably established in relation to the selling price. From 1850 onwards, Cocks (Édition Féret) published “Bordeaux et ses vins” in which he gave the public, along with a few local descriptions, a list of owners, ranked by the author according to their merit. As early as 1900, in the absence of any official classification, many owners in the Saint-Émilionnais region were unceremoniously giving themselves the title of Premier Cru. Absolute anarchy reigned in the Saint-Emilion region. This disorder caused great confusion among consumers and did a disservice to the reputation of the Saint-Émilionnais.
In 1934, Robert and André Villepigue, owners of Château Figeac and president of the cooperative winery, wrote an “essai de classement scientifique” based for the first time on a study of the nature of the soils (appendix 1). It was not until October 7, 1954, that an official classification was obtained by decree. It gave rise to much procrastination, which ended with the publication of a decree in the Journal Officiel on October 31, 1958. At that time, there were 75 estates divided into three classes: Premier Grand Cru Classé A, Premier Grand Cru Classé B, and Grand Cru Classé, which were reviewed every ten years. It was in fact revised in 1969, 1986, 1996, 2006, 2009 and 2012.
The 2012 ranking is clearly distinguished from other rankings by the official publication of a set of rules that are therefore known to all (appendix 2). Unfortunately, the terms remain extremely vague. During the preparation of the file, any request for clarification is declined. This was the beginning of a titanic task, providing as much information as possible on our respective vintages. Some dossiers exceed a thousand pages. It is only several months after all the candidates have submitted their dossiers that the I.N.A.O. sends a letter to over 70% of them, stating that they have not been admitted to the classification. This letter includes the various marks obtained. It is accompanied by instructions on how to “construct” the wine quality and consistency score, and enfinds four tables containing the criteria used by the I.N.A.O. to assess the reputation, characterization and management of the winery, together with their instructions for use (appendix 3). These candidates then have fifteen days to prepare for their oral exam, which they will present for twenty minutes before the classification commission, whose members have been appointed by the INAO. The final result will be made known to all by the homologation of the classification on October 29, 2012.
Let’s take a look at some of the criteria used in this ranking.
Sampling methods
No sampling protocol is provided. An agent of the I.N.A.O. visits the candidates to obtain samples. On this subject, in 1934, Monsieur Villepigue wondered about “those who will fill their bottle with an unquestionable Premier Cru bought in the first grocery store that comes along”. Like him, I think it’s more appropriate to pick up samples unexpectedly at a well-stocked wine shop rather than at the candidates’ homes.
On wine quality and consistency
About tasters:
No taster selection protocol has been announced. A call for candidates was launched by the I.N.A.O. Candidates selected will undergo three months’ training by a professor emeritus, against a standard sample. Two groups of tasters are formed and will share the thousand or so samples to be tasted. No distribution protocol has been communicated.
On tasting.
Wine tasting will establish the level of quality and consistency. For Grands Crus Classés, it will account for 50% of the final score; for Premiers Grands Crus Classés, it will account for 30%.
No tasting protocol is provided. There are no eliminatory marks. The rules stipulate that a second tasting is excluded. Only an explanatory note on the construction of this score is provided (appendix 3a). Wines are tasted on the basis of seven descriptors, each assigned a coefficient by the I.N.A.O.. Detailed scores by descriptor are not provided.
To determine the consistency of wines over ten years, I.N.A.O. will assign a bonus/malus called i.
(Figure1) The propensity to obtain bonuses (40%) is greater than the propensity to obtain maluses (30%): the distribution of i is not equal.
(Figure 2) In the case where the absolute value of i is equal to 1, the bonus is attributed to a greater number of notes than the malus: the distribution of i is not equitable.
The value of i is arbitrary. What’s more, its inequality and unfairness increase the divide between the scores distributed on either side of 14/20.
As the title suggests, the quality and consistency score is a construct, and I regret this.
Conclusion
The fog surrounding the tasting procedures prevents any investigation and fuels suspicion. This is most regrettable. “It is indisputable that at this level, tasting corresponds to expertise” and that this “sensory analysis can only be carried out by tasters of a high level who have international knowledge of wine” (AIV, JP32). Jacques Puisais intends this for noble wines. Would it be ridiculous to wish it for the classified wines of Saint-Emilion? But why choose experts if we’re going to put blinkers on them and restrict them with arbitrary weightings? In fact, it would be easier to choose neophytes who would be trained to prefer one type of wine without asking themselves too many questions about all the great wines they haven’t yet drunk.
Consistency is insipid. It can be evaluated by a statistical processing method, but not arbitrarily.
On the other hand, if a standard sample is to be chosen, two questions arise: who is going to choose it and which one is it going to be? If it were me, I’d be in a quandary: the one made from 60% Cabernet and 40% Merlot on sandy-gravelly soil, or the one made from 20% Cabernet and 80% Merlot on clay-limestone soil? I love both and consider them to be great wines, each with a different expression of excellence. And if I do make a choice, it’s certain that the other will be less highly rated than a third wine, less good, but with characteristics more like those of the stallion. I’m convinced that this method leads to standardization, in other words, a denial of the expression of terroir.
A minimum eliminatory score must be required afin order to avoid exaggerated compensation through non-sensory parameters. Could it be that a small wine rated 8/20 in a tasting will win the classification?
Topographical and geo-pedological analysis (appendix 3f)
To carry out this work, I.N.A.O. will rely on the soil map of the Saint-Emilion vineyards (Appendix 4), drawn up 23 years ago in Cornelis Van Leeuwen’s thesis.
On the change of scale
The I.N.A.O. produces a new 1/6,500 scale map (3.8 times more accurate than C.Van Leeuwen’s). Cornelis Van Leeuwen points out that certain “indications can only be global, due to variations that are sometimes very localized” (p.9 and p.25). This remark is fundamental to cartography, and explains why no-one can seriously afford to draw up a 1:6,500 map from another 1:25,000 map without first carrying out additional, compulsory and necessary observations, such as coring and soil pits. This would have enabled us to increase the density of information and thus to localize most of the “sometimes very localized variations”, to restrict or enlarge an area, to modifier its boundaries and, more generally, to perceive a problem no longer in its entirety but with greater precision and thus accuracy. Unfortunately, I.N.A.O. did not carry out this work. It simply zoomed in, confusing computer graphics with cartography.
On hydromorphy
The purpose of this map is to delimit and qualifier the various hydromorphic zones in the Saint-Emilion vineyards on a parcel-by-parcel scale. Apparently, the I.N.A.O. only considers soil hydromorphy as a limiting geo-pedological factor. But the term hydromorphy is used incorrectly. Let me therefore clarifier here the exact meaning of this word. Etymology of the word: hydro from the Greek hûdor (= water) and morphê (= form). A hydromorphic soil is one “in the formation of which the principal factor has been water; that of submersions or that of the water table” (Plaisance and Cailleux, 1958).
Today, an official definition exists:
“Soil hydromorphy is the visible result of previous waterlogging of a soil, but can also characterize the process of formation or evolution of a soil class in the presence of prolonged excess water”. It is stipulated that this term should not be confused with “soil waterlogging”.
Baize and Jabiol (1995) insist: “in the strict and original sense, hydromorphy is the morphological manifestation of waterlogging (past or present) in the form of stains, concentrations, colorations or discolorations, resulting from the dynamics of the two colored elements in an alternately reductive then reoxidized environment: iron and manganese, or in the form of an accumulation of organic matter (peaty soil). So be careful not to confuse cause and effect, and don’t use the word hydromorphy to describe waterlogging. What’s more, if there’s no iron or manganese in a horizon, hydromorphy won’t occur (as in the case of certain materials made up almost entirely of limestone).”
We therefore understand that the words hydromorphy and hydromorphic are often used incorrectly, as there is confusion between the morphology of certain soil horizons showing aspects linked to fossil or present-day waterlogging (pseudogley) and the waterlogging itself. Unfortunately, I.N.A.O. makes this mistake. It takes Cornelis Van Leeuwen’s legend: soil with “presence of a pseudogley”, and transforms it into: soil “with temporary water table”. However, no observations or pits correlating these two legends have been made. The I.N.A.O.’s conclusions are therefore false.
Pseudogley soils (appendix 5), which can be found in over 50% of the Saint-Emilion A.O.C., are indeed hydromorphic soils.
These soils have the property of keeping most of their redox features (pseudogleys) intact even in dry periods, whether these are temporary (summer) or permanent, resulting either from artificial drainage (agricultural drainage) or definitive climatic change.
As a result, in a large number of cases, soils with a distinctly hydromorphic morphology no longer experience significant waterlogging, which could be detrimental to the quality of the grapes grown on them, and hence the wine produced from them.
On the author’s recommendations
Cornelis Van Leeuwen accompanies his map with a leaflet that reads:
“As a precaution, we would like to point out the limits of use of this document: it is a pedological study, intended as an inventory of the region’s soils. Any extrapolation with a view to attempting to classificate the crus would be meaningless, as the hierarchy of wine quality is based on the conjunction of several parameters which, when they concern the soil, do not always fit into the rigid and schematic framework of cartography” (p.9).
“Human action can, through the cultivation of land, profoundly modifier the initial profil of the soil: modification of the depth of the water table and of the conditions of soil aeration through drainage.” (p.24).
“In rehabilitated soils, pseudogley is evidence of ancient hydromorphy” (p.25).
“The aim of this work was never to rank vintages according to soil type” (p.50).
The I.N.A.O. does not find it useful to take this information into account, and makes a serious error in interpreting the term hydromorphic.
Conclusion
Now that everyone understands what a hydromorphic soil is, we can affirm that it’s not the hydromorphy of the soil that’s responsible for its so-called inferior quality, but rather its waterlogging. A soil classification based on a morphological study is either false or incomplete. I think what we need to look at above all is the soil’s water regime. We can measure the kinetics of the sinking of the water table according to the season, and its maximum height at the right moment when the vine should experience a hydric deficit, triggering the growth arrest necessary for quality ripening.
We could, for example, supplement these data with the percentage of slope as well as the orientation of the plot (determining the rate of sunshine), vigor, vegetative expression, wind dynamics but also the density and diversity of underground fauna and flore, not forgetting most certainly the state of the clay-humus and ectomycorrhizal complexes.
General conclusion
Each time, it’s a thunderclap that provokes a scandal with international repercussions. These reactions are amplified by the lack of transparency in the method used. For example, in 1985, Beauséjour Bécot was demoted. The I.N.A.O. gave the reason: “the land originally belonged to two crus not figurant in the officielle list of Premiers Grands Crus Classés”. Would that he had been warned that he could not do so. Apparently in 2012 this rule changed (modification of the land base ranging from simple to double, sometimes of unclassified origin). Unfortunately, not all applicants will be made aware of this new opportunity. In 2006, the classification was annulled on the grounds of inequity of treatment between candidates. In 2012, a request for annulment to the administrative court is underway. Each time, the ranking loses credibility. Is it the fault of those who defend themselves against injustice? No!
If “the fundamental objective of vine-growing, which is to satisfy man’s taste for wine” (J.Branas 34, AIV) had been the objective of the classification every time, we wouldn’t be where we are today.
In any case, whatever the objective, it’s important to choose real experts and make sure they don’t make any mistakes. The economic and human consequences are too far-reaching. Arbitrariness must be ruled out and impartiality respected.