Which French vermouth belongs in a dry mixed drink? With which, for example, should one prepare a Dry Martini Cocktail? What answers do historical recipes give to this question, and how should we understand their instructions?
We were asked whether a dry French vermouth must always be used in historical recipes that call for French vermouth, or whether a sweeter variant could also be used. Specifically, the two alternatives Noilly Prat Dry and Dolin Blanc were mentioned. So is it possible to make a dry cocktail with Dolin Blanc?
Ultimately, the choice of vermouth is always a personal decision, based on one’s own sense of taste, and there can hardly be a right or wrong answer. But what do the historical sources actually say? So let’s go on a brief search for clues.
Vermouth styles
First of all, of course, there is the question of what a French vermouth actually is. Simon Difford writes that in the former Savoy, there were different production areas for vermouth, which also differed in their style of vermouth. [1] This certainly refers to Turin, where the “Vermouth di Torino” was produced, the epitome of Italian vermouth, and Chambéry, which gave its name to the “Vermouth de Chambéry“. But the vermouth of Noilly Prat, which is not Vermouth de Chambery, is also a French vermouth.
It is generally assumed, as Simon Difford reports, that Italian vermouth is sweet and French vermouth is dry. It follows, then, that in any recipe calling for a French vermouth, a dry vermouth should be used, and then, if an Italian vermouth is to be used, a sweet one. But such a distinction, he interjects, was never clear-cut, and besides, both dry and sweet versions are made in Italy and France today. [1]
These introductory words may suffice to clearly outline the question. What is a French vermouth? How is its use in historical recipes to be understood? Which vermouth could be used from a historical perspective?
The classification
We do not know what used to be understood by the terms “sweet” and “dry”. Today, however, it is simpler, because a vermouth is classified as follows on the basis of its sugar content due to an EU regulation: [2]
a) ‘extra-dry’: in the case of products with a sugar content of less than 30 grams per litre; [80 grams as stated is wrong, as can be seen from the german version]
b) ‘dry’: in the case of products with a sugar content of less than 50 grams per litre;
c) ‘semi-dry’: in the case of products with a sugar content of between 50 and 90 grams per litre;
d) ‘semi-sweet’: in the case of products with a sugar content of between 90 and 130 grams per litre;
e) ‘sweet’: in the case of products with a sugar content of more than 130 grams per litre.
Based on this definition, it is clear for some of today’s readers: if, for example, a book published in 1909 says that a “Martini Cocktail” should be prepared with Italian vermouth, but a “Martini Cocktail, Dry” with a French vermouth, it is quite clear that a dry French vermouth is required, with a sugar content of less than 50 g per litre – after all, this is the definition of a dry vermouth. A sweet French vermouth with a sugar content of, for example, 130 g/l should not be used under any circumstances. One then deduces that all recipes that call for French vermouth may also only be prepared with a dry vermouth. Unfortunately, there are weighty historical sources that contradict this conclusion.
Two examples of a French vermouth
In the following observations we will consider two different products, namely from the producers Dolin and Noilly Prat. What distinguishes these two?
Noilly Prat
The company was founded in 1813 by Joseph Noilly, and in that year he is said to have perfected a recipe for his vermouth. His son Louis took over the business, and when his brother-in-law became part of the company in 1843, it was renamed Noilly Prat. [4-55][9] The company was founded in Lyon in 1813, in 1843 they moved to Marseille, about 300 km to the south, and since 1853 they have been based in Marseillan, about 200 km to the west on the Mediterranean coast. [17] In 1876 Noilly Prat was not represented at the United States Centennial Commission International Exhibition, nor is there any indication that their vermouth could be obtained in New York of the time, as it was in San Francisco and New Orleans. [6-89] Noilly Prat Dry, said to have originated in 1813, contains 30 grams of sugar per litre, [3][9] and is said to have been the world’s first dry vermouth. [7-478] Noilly Prat Extra Dry, composed in 1955 for the American market, [31-498] even only 21 g/l. [18]
Dolin
The roots of the Dolin company in Chambéry go back to 1815. At that time, Joseph Chavasse began producing liqueurs in the commune of Les Échelles, located in Savoie. Six years later, in 1821, he invented a vermouth recipe called Vermouth de Chambéry. The company was later renamed Dolin when his son-in-law Ferdinand Dolin ran the business. The company is said to have been the first to launch a light vermouth, Vermouth Blanc. [4-56][4-171][7-218][8] 1876 Dolin was represented at the United States Centennial Commission International Exhibition. [6-89] His product received several awards and medals at competitions and fairs in Philadelphia, St. Louis, Paris and London in the 19th century. [7-218] Dolin Blanc contains 130 g of sugar per litre. [7-219][19] The company states that it was launched in 1881. [29] Dolin Dry, on the other hand, contains only 30 g per litre. [7-219]
Exports to the USA
It is of course difficult to determine which vermouth was exported to the USA, when and in what quantities. On the basis of the few surviving records, it can be said that Cora first exported its Vermouth di Torino in 1838, [4-58] Gancia in 1850, [4-60] Martini & Rossi in 1867 [20-135][22-297] or 1868. [6-89] The pre-eminence of Italian vermouth may be gauged from the fact that as early as 1877 Martini & Rossi supplied 3/4 of the vermouth imported into the USA. [21-297]
Noilly Prat, as a representative of French vermouth, first supplied New Orleans and New York in 1844. [4-55][6-89] We could not find any information for Dolin. However, the fact that they were represented at the important exhibitions and won medals and awards there suggests that they were well known and available. So we are not making a mistake in the historical context when we consider both Noilly Prat and Dolin vermouths for the following examples, because we must assume that both were available in the USA. To what extent and in what place we cannot say, because unfortunately Noilly Prat and Dolin are practically not mentioned in historical books or newspaper advertisements of the 19th century. A text published in 1899, which we will discuss in more detail later, can also give us information about this, even though it seems to be the case at first. This text states: “The manufacture of vermouth in France is confined almost exclusively to the city of Marseilles, where all the important manufactories exist. … The two principal points in Europe from which vermouth is exported are Marseilles and Turin.”[15-753] Noilly Prat, however, was located in Marseillan, 200 km away, and Dolin Blanc in Chambéry, about 370 km away, at the time of publication. So the author excluded both producers? Was he perhaps only talking about them being shipped via the port of Marseille? Moreover, the article first appeared in London. So is he perhaps not talking about the American market at all, but the English market? These are many questions that could be interesting to investigate – but in the context of the question posed here, it is irrelevant. Noilly Prat Dry and Dolin Blanc were in all probability known and available in the USA. So you can use them in the historical context without any problems. But do you still get a “Dry Cocktail” with a sweeter Dolin Blanc?” – that was the question.
The use of French vermouth in mixed drinks
O. H. Byron: The Modern Bartenders’ Guide. 1884, page 21. [10-21]
French vermouth was used in bars early on. For example, O. H. Byron published a recipe for a Manhattan cocktail with French vermouth as early as 1884. [10-21] So it is by no means the case that this vermouth only found its way into bars much later. However, French vermouth was not used more widely until around 1900.
One need only look at the three different editions of Harry Johnson’s “Bartender’s Manual” to see this by way of example. The first edition from 1882 only knows the Vermouth Cocktail, which he prepared simply with “Vermouth”. [12] The next edition from 1888 also knows the Martini Cocktail and the Manhattan Cocktail, both also with “Vermouth” [13] In the 1900 edition, the three aforementioned recipes are still included and prepared with “Vermouth”. However, other mixed drinks have been added, namely with vermouth: Morning Cocktail, Little Egypt, Bradford À La Martini; with Italian vermouth: Trilby Cocktail, Bijou Cocktail; with French vermouth: Thorn Cocktail, Montana Cocktail, Star Cocktail, Silver Cocktail, St. Joseph Cocktail, Marguerite Cocktail, Imperial Cocktail, Reform Cocktail, Klondyke Cocktail, Brazil Cocktail, Black Thorn, Tuxedo Cocktail, Turf Cocktail. [14]
Harry Johnson: The New and Improved Illustrated Bartenders‘ Manual . 1900. Page 162-163. [14-162][14-163]
Interestingly, Harry Johnson writes about the Manhattan Cocktail in 1900: “It is the bartender’s duty to ask the customer, whether he desires his drink dry or sweet.”[14-163] He had not yet given this hint in his 1888 edition; this may be an indication that by 1900 mixed drinks were also drier. He also tells us that the sugar content should be varied according to the customer’s wishes.
The Cocktail Book, published anonymously in 1901, is even more precise. The normal Manhattan Cocktail is made with sugar syrup and Italian vermouth. If it is to be dry, the sugar syrup is omitted, and if it is to be extra dry, French vermouth is used instead of Italian. [22-17] This is interesting because even the Dry Manhattan cocktail is to be made with Italian vermouth, so you don’t necessarily need a French one.
Anonymus: Anleitung zur Bereitung Amerikanischer Eis-Getränke. (1898), page 38. [11-38]
Nevertheless, a French vermouth is usually used for a dry cocktail because it seems to contain less sugar. That is why today one hears again and again that a Dry Martini must also be made with a French vermouth, but by no means with an Italian one. On closer inspection, however, this is a misconception. Of course, it seems that an Italian vermouth contains more sugar than a French one. But this does not necessarily distinguish the two types from each other. In the “Anleitung zur Bereitung amerikanischer Eis-Getränke”, which is said to have been published in 1898, a price list is printed. It says: “Vermouth wine. Vermouth de Turin, sweet … Vermouth de Turin, dry, tart” – “Vermouth Wein. Vermouth de Turin, süss … Vermouth de Turin, trocken, herb”. [11-38] This proves that an Italian vermouth was by no means always sweet. It was also available dry. So the classification as Italian does not really say anything about the sugar content, just as it does not say anything about the sugar content of a French vermouth. But if the two cannot necessarily be distinguished by the sugar content, then by what? The colouring? The aroma?
Journal of the Society of Arts. 11. August 1899, page 753. [15-753]
A text published in 1899 both in London and in St. Louis can give us information about this. The article describes that a French vermouth is made with dry white wine, and an Italian vermouth with a sweet white wine, [15-751][16-346] but this is not the main point, because a wine can be sweetened. What is more important is this statement: „The quality of the vermouth manufacture in France depends in a great measure upon the sort of wine used. The wines most employed are those of the valley of the Rhone, certain Spanish wines, and the wines of the extreme South of France. There is a difference between the French and Italian vermouth. A number of French manufacturers make Italian vermouth, however, not for the purpose of deceiving the customer as to its origin, but merely as a type of vermouth, distinct in flavour from the article known as French vermouth.“ [15-753][16-346]
What do we learn from this? The main distinction between French and Italian vermouth is not the sugar content, but the aroma! So if a French vermouth was specified in a recipe, it was not necessarily to reduce the sugar content, but because a particular aroma was intended. If one accepts this as the primary intention, many other things become clear. The aroma is the key. A Rob Roy is simply not an Affinity Cocktail. A Bamboo Cocktail is simply not an Adonis Cocktail. And even if the Martini Cocktail and the Martinez Cocktail were initially indistinguishable from each other, today we generally understand them to mean something different. The difference here is not the sweetness, the liqueur or the colouring. The difference is the aroma that results from the vermouth used. For us today, a Martini Cocktail is simply not a Martinez Cocktail.
Kansas City Journal, 7. Februar 1899, page 10. [30]
This judgement is confirmed as early as 1899. At that time, people wrote about the Manhattan Cocktail: “The true Manhattan cocktail is always made with Italian vermouth, but at half the places where they undertake to serve it French vermouth is substituted, and the fine flavor is altogether destroyed. French vermouth is a sort of wine, Italian vermouth is a cordial pure and simple. They are as different as milk and molasses. A cocktail made from the French brand is no more a Manhattan than it is a Spanish omelette.” [30]
Nevertheless, it is true that usually a mixed drink prepared with a French vermouth is less sweet than with an Italian vermouth. We have already written that Noilly Prat Dry contains 30 g/l of sugar and Dolin Blanc 130 g/l. Carpano Antica Formula, on the other hand, as a typical Italian vermouth, contains 190 g/l [23]. This leads us back to the postulate formulated at the beginning, that a Dry Martini must also be made with a Dry Vermouth, i.e. in our example with Noilly Prat Dry. What are we to make of this? Can a Dolin Blanc also be used as an alternative?
Example calculations
In order to be able to make a well-founded statement about which French vermouth one should work with in old recipes, a little mathematics helps, which we apply to recipes selected as examples. In doing so, we have selected recipes from those books that offer several different variants of the same mixed drink, because this is the only way to really compare the variants with each other. What is dry for one author may no longer be so for another. This is particularly important with recipes as variable as those for the Martini Cocktail, Martinez Cocktail or Manhattan Cocktail, which could be prepared with or without sugar syrup, with or without liqueur, with Italian or French vermouth.
We calculate on the following basis:
Noilly Prat Dry contains 30 g/l sugar.
Dolin Blanc contains 130 g/l sugar.
Carpano Antica Formula contains 190 g/l sugar.
Old Tom Gin contains 35 g/l sugar, as evidenced by the analysis of a bottle from 1854. [25]
Sugar syrup (2:1) contains 800 g/l sugar, as we found in our own experiment.
Liqueur contains at least 100 g/l sugar. It can also be considerably more. Crèmes must contain at least 250 g/l, Crème de Cassis even at least 400 g/l. [24] For the example calculation we use 100 g/l.
The melt water increases the volume of a mixed drink by 35%.
1 dash equals 5 ml; as we have found in trials, this is a suitable amount for liqueurs when preparing old recipes.
As an example, let us now look at the recipe for a Martini Cocktail published by John Applegreen in 1909. He makes it with 1 dash of sugar syrup, 1/2 jigger of Old Tom gin and 1/2 jigger of Italian vermouth. [26-2]
With this, the following calculation can be made: The volume is first calculated from the sum of 1 dash of syrup, 1/2 jigger of gin and 1/2 jigger of vermouth, which is 5 ml + 30 ml + 30 ml, i.e. a total of 65 ml. Melting water increases this amount by 35% to 87.75 ml. This contains 4 g of sugar from the syrup, 5.7 g from the vermouth and 1.05 g from the gin, making a total of 10.75 g of sugar. If you extrapolate this amount to one litre, you get a sugar content of 122.51 g/l. We now evaluate this result according to the EU regulation for vermouth; for us, this should also be valid for mixed drinks, and we classify the mixed drinks on the basis of the sugar contents given there. With 122.51 g/l sugar, this Martini Cocktail can be described as “semi-sweet”.
We now apply this method of calculation to other recipes, and if they state that a French vermouth should be used, we take Dolin Blanc and Noilly Prat Dry as alternative ingredients.
John Applegreen: Applegreen’s Bar Book. 1909, page 2. [26-2]
Let us first look at the recipes published by John Applegreen in 1909:
Martini Cocktail.[26-2]2 dashes orange bitters; 1 dash syrup; 1/2 jigger Tom gin; 1/2 jigger Italian vermouth; Piece lemon peel; Ice; Strain into cocktail glass. Calculation: 122.51 g/l sugar, semi-sweet.
Martini Cocktail, Dry (1). [26-2]Same as above except to omit the syrup. Calculation: 83.33 g/l sugar, semi-dry.
Martini Cocktail, Dry (2). [26-2]Use small mixing glass, into which put 2 dashes of orange bitters; Piece of lemon peel; 2 or 3 lumps of ice; 1/3 jigger of French vermouth; 2/3 jigger of dry gin; Stir well and strain into a cocktail glass. Calculation: With Dolin Blanc 32.10 g/l sugar, dry; with Noilly Prat Dry 7.41 g/l, extra dry.
For better comparability with the following recipes, in which we use 45 ml of spirits and vermouth each, as we always do with a Manhattan Cocktail or a Martinez Cocktail, you can also do the above calculation. Logically, the values only change for the first recipe, because in this one the sugar content is reduced from 122.51 g/l to 110.14 g/l as a result. However, this does not change the classification.
Jacob A. Didier: The Reminder. 1909, page 11-12. [27-11][27-12]
Let us take as further examples the recipes published by Jacob A. Didier in 1909:
Martini Cocktail.[27-11]Use a mixing glass, half fill with cracked ice. 2 dashes of orange bitters. 1 dash of curacoa. 1/2 drink of Italian vermouth. 1/2 drink of Tom gin. Stir, strain into a cocktail glass, twist lemon peel on top and serve. (Add cherry or olive if desired). Calculation: 82.85 g/l sugar, semi-dry.
Martini Cocktail (Dry).[27-12]Use a mixing glass, half fill with cracked ice. 2 dashes of orange bitters. 1/3 drink of French vermouth. 2/3 drink of dry gin. Stir, strain into cocktail glass and serve. (Add cherry or olive if desired). Calculation: With Dolin Blanc 31.10 g/l sugar, dry; with Noilly Prat Dry 7.41 g/l sugar, extra dry.
Let’s now look at examples of the Manhattan Cocktail, which were published in 1898 in the anonymous book “Cocktails”:
Manhattan Cocktail.[28-27]FILL mixing-glass half-full fine ice, add two dashes gum-syrup, two dashes Boker’s bitters, one-half jigger Italian vermouth, one-half jigger whiskey. Mix, strain into cocktail-glass. Add a piece of lemon peel. Calculation: 144.97 g/l sugar, sweet.
Manhattan Cocktail – Dry.[28-28]PREPARE same as Manhattan Cocktail, leaving out syrup. Calculation: 70.37 g/l sugar, semi-dry.
Manhattan Cocktail – Extra Dry.[28-28]MIX same as Manhattan Cocktail. Leave out syrup, and use French vermouth in place of Italian. Calculation: With Dolin Blanc 48.15 g/l sugar, dry; with Noilly Prat Dry 11.11 g/l sugar, extra dry.
Here, too, one might object that the proportion of vermouth and whiskey should be increased from 30 ml to 45 ml each. But here, too, only the first recipe reduces the sugar content from 144.97 g/l to 122.59, which means that the cocktail would have to be described as sweet instead of semi-sweet.
Because we will refer to this in the following, the last example is an old-fashioned cocktail. Stirred with 60 ml of whiskey and 5 ml of sugar syrup, it has a sugar content of 45.58 g/l, which means it can be described as dry.
What can we deduce from these observations? With an Italian vermouth, the recipes result in a sweet, semi-sweet or semi-dry mixed drink with sugar contents between 70 g/l and 145 g/l, on average 101 g/l. The “dry” variants with Dolin Blanc as the French vermouth range from 32 g/l to 48 g/l, with an average of 37 g/l, and can thus all be classified as dry. This is not much, because as shown, a normal Old-Fashioned cocktail, which also qualifies as dry, is much sweeter with around 46 g/l of sugar.
If you use Noilly Prat Dry, on the other hand, you get between 7 g/l and 11 g/l, on average 9 g/l, and these variants are to be classified as extra dry. One can of course discuss whether, for example, 1 dash always corresponds to 5 ml, whether one should only use a 1:1 syrup instead of a 2:1 sugar syrup, and how much sugar is really in the liqueur. But this discussion is superfluous, because it does not change the essential statement: with an Italian vermouth, the mixed drinks are always sweet on average. With a Dolin Blanc, on the other hand, which still contains 130 g/l of sugar, they become dry. With a dry vermouth like Noilly Prat, the mixed drink even becomes extra dry.
The historical recipes say nothing else: If you want to get a dry result, you should use a French vermouth. As a rule, there is no mention of a dry vermouth, only of a dry drink. Now, it is nevertheless the case that sometimes a dry vermouth or even explicitly a Noilly Prat is prescribed in order to obtain a dry mixed drink, such as the first recipe for a Dry Martini cocktail from 1904 with a total sugar content of around 11 g/l; [5-27] but this is not a contradiction, because at that time, of course, one did not adhere to the nomenclature of the EU regulation. They did not always distinguish between dry and extra dry in today’s sense of the EU regulation, and what was extra dry for one was dry for another.
It follows that one can of course use a dry vermouth for the preparation of dry mixed drinks, but not necessarily. If you use a sweeter French vermouth, you will also get a dry mixed drink.
https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A31991R1601 Council Regulation (EEC) No 1601/91 of 10 June 1991 laying down general rules on the definition, description and presentation of aromatized wines, aromatized wine- based drinks and aromatized wine-product cocktails.
Jared Brown & Anistatia Miller: The Mixellany Guide to Vermouth & Other Apéritifs. ISBN 978-1-907434-25-9. 2011.
Frank Newman: American-Bar. Boissins Anglaises & Américaines telles qu’on les prépare. 2. edition. Paris, Société francaise d’imprimerie et de librairie, 1904.
Adam Ford: Vermouth. ISBN 978-1-58157-296-4. 2015.
Thomas Majhen: Die Barfibel. ISBN 978-3-8442-5233-9. 2012.
O. H. Byron: The Modern Bartenders‘ Guide, or Fancy Drinks and How to Mix Them. Containing Clear and Practical Directions for Mixing All Kinds of Cocktails, Sours, Egg Nogg, Sherry Cobblers, Coolers, Absinthe, Crustas, Fizzes, Flips, Juleps, Fixes, Punches, Lemonades, And Pousse Cafes, Together With Complete Directions and Receipts for Making All Kinds of Domestic Brandies, Beers, Wines, Cordials, Extracts and Syrups. New York, 1884.
Anonymus: Anleitung zur Bereitung Amerikanischer Eis-Getränke etc. und Kochrecepte für die feine Küche unter Verwendung unserer Special-Bodega-Marken. Frankfurt, The Continental Bodega Company, without year.
Harry Johnson: New and Improved Bartender’s Manual or: How to Mix Drinks of the Present Style And Containing a Valuable List of Instructions and Hints of the Author in Reference to Attending Bar, and also a Large List of Mix-Drinks Together With a Complete List of Bar Utensils, Wines, Liquors, Ales, Mixtures, Etc. Ab Seite 77: Practisches, Neues und Verbessertes Handbuch für Barkeeper, Salon- und Hotelbesitzer, Küfer, Weinbauer, Hausfrauen etc. enthaltend practische Winke und Anweisungen für Barkeeper, vollkommen correcte Rezepte aller gemischten Getränke der gegenwärtigen Zeit, Listen sämmtlicher Bar-Artikeln und Utensilien, Regeln zur Behandlung von Liquors, Bier, Ale, Porter, Wein und Cider in Fässern sowohl als auch Flaschen, etc., etc., mit einem Anhang der Anleitung zur Erzeugung von Wein und Cider. New York, Samisch & Goldmann, 1882.
Harry Johnson: New and Improved Illustrated Bartender’s Manual or: How to Mix Drinks of the Present Style And Containing a Valuable List of Instructions and Hints By the Author in Reference to Attending Bar: Also a Large List of Mixed Drinks, Such as American, British, French, German, Italian, Russian, Spanish, Etc., Etc., With Illustrations, And a Complete List of Bar Utensils, Wines, Liquors, Ales, Mixtures, Etc., Etc. Ab Seite 103: Neues und Verbessertes Illustriertes Handbuch für Bartender, oder: Wie man Getränke mischt, enthaltend Practische Regeln, Winke und Anweisungen über sämmtliche Bedürfnisse, gründliche Belehrung über alle Einzelheiten des Geschäfts, vollkommene und correcte Rezepte aller gemischten Getränke der Jetztzeit die in Amerika, England, Deutschland, Frankreich, Italien, Russland, Spanien und anderen Ländern beliebt sind, sowie Listen sämmtlicher Bar-Utensilien, Anweisungen zur richtigen Behandlung von Liqueuren, Weine Bier, Ale und Porter in Fässern und Flaschen. New York, Eigenverlag, 1888.
Harry Johnson: The New and Improved Illustrated Bartenders‘ Manual or: How to Mix Drinks of the Present Style, Containing Valuable Instructions and Hints by the Author in Reference to the Management of a Bar, a Hotel and a Restaurant; also a Large List of Mixed Drinks, including American, British, French, German, Italian, Russian, Spanish, etc., with Illustrations and a Comprehensive Description of Bar Utensils, Wines, Liquors, Ales, Mixtures, etc., etc. Revised Edition. New York City, 1900.
Anistatia Miller & Jared Brown: Spirituous Journey. A History of Drink. Book Two: From Publicans to Master Mixologists. First Edition, Mixellany Limited, London, 2009. ISBN 0-9781907434-06-8. Page 229.
Gaz Regan: The Bartender’s Gin Compendium. ISBN 978-1-4415-4688-3. 2009.
Anonymus: The Cocktail Book: A Sideboard Manual for Gentlemen. Boston, L. C. Page & Company, 1901. According to copyright deviating from the title page 1900.
John Applegreen: Applegreen’s Bar Book or How to Mix Drinks. 3. überarbeitete Auflage. Chicago, The Hotel Monthly, 1909.
Jacob A. Didier: The Reminder. An Up-to-Date, Bartenders‘ Vest Pocket. How to Mix Drinks of the Present Time. Containing Clear and Practical Directions for Mixing the Most Popular Plain and Fancy Drinks, Such as Cocktails, Daisies, Fixes, Fizzes, Flips, Sours, Cobblers, Punches, Rickeys, High Balls, Frappes, Juleps, Hot Drinks, Etc. Etc. 3. edition. New York, The Outing Press, 1909.
Anonymus: Cocktails. How to Make Them. Providence, Livermore & Knight Co., 1898.
We were asked whether a dry French vermouth must always be used in historical recipes that call for French vermouth, or whether a sweeter variant could also be used. Specifically, the two alternatives Noilly Prat Dry and Dolin Blanc were mentioned. So is it possible to make a dry cocktail with Dolin Blanc?
Ultimately, the choice of vermouth is always a personal decision, based on one’s own sense of taste, and there can hardly be a right or wrong answer. But what do the historical sources actually say? So let’s go on a brief search for clues.
Vermouth styles
First of all, of course, there is the question of what a French vermouth actually is. Simon Difford writes that in the former Savoy, there were different production areas for vermouth, which also differed in their style of vermouth. [1] This certainly refers to Turin, where the “Vermouth di Torino” was produced, the epitome of Italian vermouth, and Chambéry, which gave its name to the “Vermouth de Chambéry“. But the vermouth of Noilly Prat, which is not Vermouth de Chambery, is also a French vermouth.
It is generally assumed, as Simon Difford reports, that Italian vermouth is sweet and French vermouth is dry. It follows, then, that in any recipe calling for a French vermouth, a dry vermouth should be used, and then, if an Italian vermouth is to be used, a sweet one. But such a distinction, he interjects, was never clear-cut, and besides, both dry and sweet versions are made in Italy and France today. [1]
These introductory words may suffice to clearly outline the question. What is a French vermouth? How is its use in historical recipes to be understood? Which vermouth could be used from a historical perspective?
The classification
We do not know what used to be understood by the terms “sweet” and “dry”. Today, however, it is simpler, because a vermouth is classified as follows on the basis of its sugar content due to an EU regulation: [2]
Based on this definition, it is clear for some of today’s readers: if, for example, a book published in 1909 says that a “Martini Cocktail” should be prepared with Italian vermouth, but a “Martini Cocktail, Dry” with a French vermouth, it is quite clear that a dry French vermouth is required, with a sugar content of less than 50 g per litre – after all, this is the definition of a dry vermouth. A sweet French vermouth with a sugar content of, for example, 130 g/l should not be used under any circumstances. One then deduces that all recipes that call for French vermouth may also only be prepared with a dry vermouth. Unfortunately, there are weighty historical sources that contradict this conclusion.
Two examples of a French vermouth
In the following observations we will consider two different products, namely from the producers Dolin and Noilly Prat. What distinguishes these two?
Noilly Prat
The company was founded in 1813 by Joseph Noilly, and in that year he is said to have perfected a recipe for his vermouth. His son Louis took over the business, and when his brother-in-law became part of the company in 1843, it was renamed Noilly Prat. [4-55] [9] The company was founded in Lyon in 1813, in 1843 they moved to Marseille, about 300 km to the south, and since 1853 they have been based in Marseillan, about 200 km to the west on the Mediterranean coast. [17] In 1876 Noilly Prat was not represented at the United States Centennial Commission International Exhibition, nor is there any indication that their vermouth could be obtained in New York of the time, as it was in San Francisco and New Orleans. [6-89] Noilly Prat Dry, said to have originated in 1813, contains 30 grams of sugar per litre, [3] [9] and is said to have been the world’s first dry vermouth. [7-478] Noilly Prat Extra Dry, composed in 1955 for the American market, [31-498] even only 21 g/l. [18]
Dolin
The roots of the Dolin company in Chambéry go back to 1815. At that time, Joseph Chavasse began producing liqueurs in the commune of Les Échelles, located in Savoie. Six years later, in 1821, he invented a vermouth recipe called Vermouth de Chambéry. The company was later renamed Dolin when his son-in-law Ferdinand Dolin ran the business. The company is said to have been the first to launch a light vermouth, Vermouth Blanc. [4-56] [4-171] [7-218] [8] 1876 Dolin was represented at the United States Centennial Commission International Exhibition. [6-89] His product received several awards and medals at competitions and fairs in Philadelphia, St. Louis, Paris and London in the 19th century. [7-218] Dolin Blanc contains 130 g of sugar per litre. [7-219] [19] The company states that it was launched in 1881. [29] Dolin Dry, on the other hand, contains only 30 g per litre. [7-219]
Exports to the USA
It is of course difficult to determine which vermouth was exported to the USA, when and in what quantities. On the basis of the few surviving records, it can be said that Cora first exported its Vermouth di Torino in 1838, [4-58] Gancia in 1850, [4-60] Martini & Rossi in 1867 [20-135] [22-297] or 1868. [6-89] The pre-eminence of Italian vermouth may be gauged from the fact that as early as 1877 Martini & Rossi supplied 3/4 of the vermouth imported into the USA. [21-297]
Noilly Prat, as a representative of French vermouth, first supplied New Orleans and New York in 1844. [4-55] [6-89] We could not find any information for Dolin. However, the fact that they were represented at the important exhibitions and won medals and awards there suggests that they were well known and available. So we are not making a mistake in the historical context when we consider both Noilly Prat and Dolin vermouths for the following examples, because we must assume that both were available in the USA. To what extent and in what place we cannot say, because unfortunately Noilly Prat and Dolin are practically not mentioned in historical books or newspaper advertisements of the 19th century. A text published in 1899, which we will discuss in more detail later, can also give us information about this, even though it seems to be the case at first. This text states: “The manufacture of vermouth in France is confined almost exclusively to the city of Marseilles, where all the important manufactories exist. … The two principal points in Europe from which vermouth is exported are Marseilles and Turin.” [15-753] Noilly Prat, however, was located in Marseillan, 200 km away, and Dolin Blanc in Chambéry, about 370 km away, at the time of publication. So the author excluded both producers? Was he perhaps only talking about them being shipped via the port of Marseille? Moreover, the article first appeared in London. So is he perhaps not talking about the American market at all, but the English market? These are many questions that could be interesting to investigate – but in the context of the question posed here, it is irrelevant. Noilly Prat Dry and Dolin Blanc were in all probability known and available in the USA. So you can use them in the historical context without any problems. But do you still get a “Dry Cocktail” with a sweeter Dolin Blanc?” – that was the question.
The use of French vermouth in mixed drinks
French vermouth was used in bars early on. For example, O. H. Byron published a recipe for a Manhattan cocktail with French vermouth as early as 1884. [10-21] So it is by no means the case that this vermouth only found its way into bars much later. However, French vermouth was not used more widely until around 1900.
One need only look at the three different editions of Harry Johnson’s “Bartender’s Manual” to see this by way of example. The first edition from 1882 only knows the Vermouth Cocktail, which he prepared simply with “Vermouth”. [12] The next edition from 1888 also knows the Martini Cocktail and the Manhattan Cocktail, both also with “Vermouth” [13] In the 1900 edition, the three aforementioned recipes are still included and prepared with “Vermouth”. However, other mixed drinks have been added, namely with vermouth: Morning Cocktail, Little Egypt, Bradford À La Martini; with Italian vermouth: Trilby Cocktail, Bijou Cocktail; with French vermouth: Thorn Cocktail, Montana Cocktail, Star Cocktail, Silver Cocktail, St. Joseph Cocktail, Marguerite Cocktail, Imperial Cocktail, Reform Cocktail, Klondyke Cocktail, Brazil Cocktail, Black Thorn, Tuxedo Cocktail, Turf Cocktail. [14]
Interestingly, Harry Johnson writes about the Manhattan Cocktail in 1900: “It is the bartender’s duty to ask the customer, whether he desires his drink dry or sweet.” [14-163] He had not yet given this hint in his 1888 edition; this may be an indication that by 1900 mixed drinks were also drier. He also tells us that the sugar content should be varied according to the customer’s wishes.
The Cocktail Book, published anonymously in 1901, is even more precise. The normal Manhattan Cocktail is made with sugar syrup and Italian vermouth. If it is to be dry, the sugar syrup is omitted, and if it is to be extra dry, French vermouth is used instead of Italian. [22-17] This is interesting because even the Dry Manhattan cocktail is to be made with Italian vermouth, so you don’t necessarily need a French one.
Nevertheless, a French vermouth is usually used for a dry cocktail because it seems to contain less sugar. That is why today one hears again and again that a Dry Martini must also be made with a French vermouth, but by no means with an Italian one. On closer inspection, however, this is a misconception. Of course, it seems that an Italian vermouth contains more sugar than a French one. But this does not necessarily distinguish the two types from each other. In the “Anleitung zur Bereitung amerikanischer Eis-Getränke”, which is said to have been published in 1898, a price list is printed. It says: “Vermouth wine. Vermouth de Turin, sweet … Vermouth de Turin, dry, tart” – “Vermouth Wein. Vermouth de Turin, süss … Vermouth de Turin, trocken, herb”. [11-38] This proves that an Italian vermouth was by no means always sweet. It was also available dry. So the classification as Italian does not really say anything about the sugar content, just as it does not say anything about the sugar content of a French vermouth. But if the two cannot necessarily be distinguished by the sugar content, then by what? The colouring? The aroma?
A text published in 1899 both in London and in St. Louis can give us information about this. The article describes that a French vermouth is made with dry white wine, and an Italian vermouth with a sweet white wine, [15-751] [16-346] but this is not the main point, because a wine can be sweetened. What is more important is this statement: „The quality of the vermouth manufacture in France depends in a great measure upon the sort of wine used. The wines most employed are those of the valley of the Rhone, certain Spanish wines, and the wines of the extreme South of France. There is a difference between the French and Italian vermouth. A number of French manufacturers make Italian vermouth, however, not for the purpose of deceiving the customer as to its origin, but merely as a type of vermouth, distinct in flavour from the article known as French vermouth.“ [15-753] [16-346]
What do we learn from this? The main distinction between French and Italian vermouth is not the sugar content, but the aroma! So if a French vermouth was specified in a recipe, it was not necessarily to reduce the sugar content, but because a particular aroma was intended. If one accepts this as the primary intention, many other things become clear. The aroma is the key. A Rob Roy is simply not an Affinity Cocktail. A Bamboo Cocktail is simply not an Adonis Cocktail. And even if the Martini Cocktail and the Martinez Cocktail were initially indistinguishable from each other, today we generally understand them to mean something different. The difference here is not the sweetness, the liqueur or the colouring. The difference is the aroma that results from the vermouth used. For us today, a Martini Cocktail is simply not a Martinez Cocktail.
This judgement is confirmed as early as 1899. At that time, people wrote about the Manhattan Cocktail: “The true Manhattan cocktail is always made with Italian vermouth, but at half the places where they undertake to serve it French vermouth is substituted, and the fine flavor is altogether destroyed. French vermouth is a sort of wine, Italian vermouth is a cordial pure and simple. They are as different as milk and molasses. A cocktail made from the French brand is no more a Manhattan than it is a Spanish omelette.” [30]
Nevertheless, it is true that usually a mixed drink prepared with a French vermouth is less sweet than with an Italian vermouth. We have already written that Noilly Prat Dry contains 30 g/l of sugar and Dolin Blanc 130 g/l. Carpano Antica Formula, on the other hand, as a typical Italian vermouth, contains 190 g/l [23]. This leads us back to the postulate formulated at the beginning, that a Dry Martini must also be made with a Dry Vermouth, i.e. in our example with Noilly Prat Dry. What are we to make of this? Can a Dolin Blanc also be used as an alternative?
Example calculations
In order to be able to make a well-founded statement about which French vermouth one should work with in old recipes, a little mathematics helps, which we apply to recipes selected as examples. In doing so, we have selected recipes from those books that offer several different variants of the same mixed drink, because this is the only way to really compare the variants with each other. What is dry for one author may no longer be so for another. This is particularly important with recipes as variable as those for the Martini Cocktail, Martinez Cocktail or Manhattan Cocktail, which could be prepared with or without sugar syrup, with or without liqueur, with Italian or French vermouth.
We calculate on the following basis:
As an example, let us now look at the recipe for a Martini Cocktail published by John Applegreen in 1909. He makes it with 1 dash of sugar syrup, 1/2 jigger of Old Tom gin and 1/2 jigger of Italian vermouth. [26-2]
With this, the following calculation can be made: The volume is first calculated from the sum of 1 dash of syrup, 1/2 jigger of gin and 1/2 jigger of vermouth, which is 5 ml + 30 ml + 30 ml, i.e. a total of 65 ml. Melting water increases this amount by 35% to 87.75 ml. This contains 4 g of sugar from the syrup, 5.7 g from the vermouth and 1.05 g from the gin, making a total of 10.75 g of sugar. If you extrapolate this amount to one litre, you get a sugar content of 122.51 g/l. We now evaluate this result according to the EU regulation for vermouth; for us, this should also be valid for mixed drinks, and we classify the mixed drinks on the basis of the sugar contents given there. With 122.51 g/l sugar, this Martini Cocktail can be described as “semi-sweet”.
We now apply this method of calculation to other recipes, and if they state that a French vermouth should be used, we take Dolin Blanc and Noilly Prat Dry as alternative ingredients.
Let us first look at the recipes published by John Applegreen in 1909:
For better comparability with the following recipes, in which we use 45 ml of spirits and vermouth each, as we always do with a Manhattan Cocktail or a Martinez Cocktail, you can also do the above calculation. Logically, the values only change for the first recipe, because in this one the sugar content is reduced from 122.51 g/l to 110.14 g/l as a result. However, this does not change the classification.
Let us take as further examples the recipes published by Jacob A. Didier in 1909:
Let’s now look at examples of the Manhattan Cocktail, which were published in 1898 in the anonymous book “Cocktails”:
Here, too, one might object that the proportion of vermouth and whiskey should be increased from 30 ml to 45 ml each. But here, too, only the first recipe reduces the sugar content from 144.97 g/l to 122.59, which means that the cocktail would have to be described as sweet instead of semi-sweet.
Because we will refer to this in the following, the last example is an old-fashioned cocktail. Stirred with 60 ml of whiskey and 5 ml of sugar syrup, it has a sugar content of 45.58 g/l, which means it can be described as dry.
What can we deduce from these observations? With an Italian vermouth, the recipes result in a sweet, semi-sweet or semi-dry mixed drink with sugar contents between 70 g/l and 145 g/l, on average 101 g/l. The “dry” variants with Dolin Blanc as the French vermouth range from 32 g/l to 48 g/l, with an average of 37 g/l, and can thus all be classified as dry. This is not much, because as shown, a normal Old-Fashioned cocktail, which also qualifies as dry, is much sweeter with around 46 g/l of sugar.
If you use Noilly Prat Dry, on the other hand, you get between 7 g/l and 11 g/l, on average 9 g/l, and these variants are to be classified as extra dry. One can of course discuss whether, for example, 1 dash always corresponds to 5 ml, whether one should only use a 1:1 syrup instead of a 2:1 sugar syrup, and how much sugar is really in the liqueur. But this discussion is superfluous, because it does not change the essential statement: with an Italian vermouth, the mixed drinks are always sweet on average. With a Dolin Blanc, on the other hand, which still contains 130 g/l of sugar, they become dry. With a dry vermouth like Noilly Prat, the mixed drink even becomes extra dry.
The historical recipes say nothing else: If you want to get a dry result, you should use a French vermouth. As a rule, there is no mention of a dry vermouth, only of a dry drink. Now, it is nevertheless the case that sometimes a dry vermouth or even explicitly a Noilly Prat is prescribed in order to obtain a dry mixed drink, such as the first recipe for a Dry Martini cocktail from 1904 with a total sugar content of around 11 g/l; [5-27] but this is not a contradiction, because at that time, of course, one did not adhere to the nomenclature of the EU regulation. They did not always distinguish between dry and extra dry in today’s sense of the EU regulation, and what was extra dry for one was dry for another.
It follows that one can of course use a dry vermouth for the preparation of dry mixed drinks, but not necessarily. If you use a sweeter French vermouth, you will also get a dry mixed drink.
Sources
explicit capitulum
*