Shaken or Stirred: Which Makes the Best Martini?
“A dry martini,” he said. “One. In a deep champagne goblet.”
“Oui, monsieur.”
“Just a moment. Three measures of Gordon’s, one of vodka, half a measure of Kina Lillet. Shake it very well until it’s ice-cold, then add a large thin slice of lemon-peel. Got it?”
-Ian Fleming, Casino Royale
Like a manhattan or an old fashioned, a martini is on its surface a simple drink to make: dry vermouth, gin (traditionally) or vodka (modern), and an olive as garnish. But a martini is something special; it’s lodged in the popular imagination, through no small fault of the man quoted above, Ian Fleming’s super spy James Bond. Whether through Fleming’s novels or film adaptations featuring Sean Connery, Daniel Craig, or countless others, if there’s one thing the average person knows about James Bond, it’s his preference of martini: shaken, not stirred.
The question is: Why? Does shaking vs. stirring change the taste of the martini? And if so, which is better? Naturally for this question we decided to hold a tasting to see how a martini fares when shaken or when stirred. For this tasting, we went with two gin martinis, made with Bombay, one shaken and one stirred.
Stirred Martini
Nose: The nose of a martini is a lovely thing, subtle and herbal and bitter. The stirred martini had notes of pine, bitter orange peel, and juniper — not too overpowering. The aroma of the vermouth was almost indistinct, and served mostly to highlight the aromas of the gin.
Palate: The initial taste of the stirred martini was briny, lightly acidic sea salt from the vermouth. Then came the gin, with the promise of the nose being borne out by juniper, bitter citrus peel, and a light Christmas-tree pine. Gin can be a tough thing for an alcohol novice to wrap their heads around, but a martini is a good, aromatic, interesting way to try something new.
Shaken Martini
Nose: The nose of the shaken martini was similar to the stirred martini, if perhaps a bit more piney. The decision of shaking or stirring didn’t seem to factor much into the nose.
Palate: Here’s where things get radically different. To start, the shaken martini was much colder, as a result of the gin being shaken up with the ice. (Many shaken martinis will even have ice chips in the drink, which some drinkers consider offensive.) The chill of the drink translated over to the taste, which was light and very, very subtle, almost to the point of not tasting like much of anything at all. There were slight notes of juniper and peel and pine, but they were buried beneath a watery simplicity. As the martini warmed up, the flavor became a bit stronger, but it was still more jumbled and indistinct than the stirred martini was.
Conclusions
So why did the drinks turn out this way? A lot of it has to do with the cold: Like a glass of white wine, it’s easy to over-chill a martini by shaking it, and the primary result of a too-cold martini is that it becomes much more thin and tasteless. This is compounded by the fact that shaking introduces more water into the drink via melted ice; a stirred martini will be a bit stronger, and thus more flavorful. As well, gin is a sensitive spirit and vigorous shaking has the result of muddling its taste. (There’s much talk of “bruising the vermouth” if you shake a martini, but it’s the gin that has the bigger problem.) All in all: A stirred martini is indeed more interesting and flavorful than a shaken one.
If there’s not much to recommend a shaken martini over a stirred one, then why does James Bond order them? The answer is twofold: first of all, Bond is the ultimate bad boy, and that extends to his choice in drinks. He doesn’t follow our rules, and from his first appearance in Casino Royale back in 1953, he was a man that blazed his own path. If society tells us to stir our martinis, of course Bond is going to be the type of guy who drinks them shaken. The other reason is more mundane. Look at his recipe again. In addition to the gin and vermouth, Bond requests a measure of vodka, making it a drink that he named The Vesper, after that book’s femme fatale. Vodka is a much heartier spirit than gin is, and if you’re drinking a vodka martini, shaking might actually be good for it, since vodka is best when it’s ice cold. Of course, given that Bond is drinking a martini with both gin and vodka in it, perhaps he just prefers a weaker drink with some water in the mix.
So that’s another taste test done, and another curious corner of the history of spirits explored. If you feel like trying this experiment yourself, let us know in the comments which style you prefer, and why!
Love this guys take on all things potable. How do I rate this A+?
Why is vodka a heartier spirit? I thought typically they are both neutral grain spirits
There’s an even more authoritative pop culture drinker who shakes his martinis: no less than Nick Charles, in The Thin Man, who not only specifies that a martini should be shaken, but that it should be done to the rhythm of a waltz.
In the books, Bond has about as many stirred martinis as shaken, and his preferred drink is bourbon (Fleming himself switched from gin to bourbon because it was “healthier”). Fleming had also never tried a Vesper when he wrote the recipe, and years after the fact finally had a bartender make him one, which he found to be “unpalatable.”
Nice comparison, by the way. It’s good when someone pairs them in a taste test rather than just repeating “common knowledge.” For my money, they both taste awful, because I cannot abide Martinis.
Thanks for the comment, Keith. I think I read somewhere that throughout all of Fleming’s books, his Bond averages something like a drink every 7 pages.
I love a well shaken vodka dirty martini so it has ice crystals!
After doing some reading on this topic, elsewhere, and side-tracking into some reading about Kina Lillet (discontinued in 1986), I think it’s worth considering the following when thinking about the dilution issue with shaking a martini vs. stirring it: 1) Gordon’s gin is what he specifies, which was 47% alcohol at the time (lowered in 1992 to around 37%), so unless you’re using another gin of comparable proof now (or a stronger vodka), you’ll end up with a more diluted flavor; 2) Kina Lillet had quinine in it, which gave it a more pronounced bitter character than modern Lillet blanc has, so again, the shaking would likely not have watered down the flavor as much because the drink itself would have had a stronger bitter character; 3) this recipe combines two high-proof liquors with a fortified wine that has a bitter character to it, and that’s the entirety of the ingredients, so again, the dilution of the shaking would probably have balanced the drink a lot more in 1953 than a version made with identical ingredients today. It’s worth noting that some clever people have modified the recipe to adapt it to currently available ingredients (Cocchi Americano and using a 100-proof vodka) in an attempt to offset the change to Gordon’s alcohol content, if Gordon’s is the choice of gin. But I think what’s ultimately the most interesting thing to note is the comment in this article about how Fleming finally had one of these years later and found it “unpalatable”, so it sounds like the Vesper Martini “shaken not stirred” sounded great on paper (or in a film) but in reality isn’t something anyone would probably want to drink.
Ugh, wish I could edit the comment. I reversed the logic of my argument in point one, which should have made the case that the Gordon’s gin of 1953 would have been less affected by dilution than a modern version due to the stronger alcohol taste; the way point 2 is written makes more sense that way. As for the “entirety” of the ingredients, technically the lemon peel and water are also ingredients, not just the three liquors. Finally, the Ian Fleming comment referenced was from a different article, not this one, but apparently he hadn’t actually had a “Vesper” when the book was published, only having one made for him years afterward, and it sounded like he wasn’t impressed with his invention.
Great article! I do have a question, though. Everything I read about this points to dilution and temperature, but isn’t that dependent on technique? If you stir a drink for long enough, wouldn’t it get just as cold and watered down as a shaken one? Granted, a shaken drink may stay cold a little longer because of the ice chips, but that’s the only real difference I can see. (Aside from the aeration/oxidation stuff that seems negligible.)
For me, the real difference comes down to control. Stirring minimizes the number of variables involved. Ice cubes are always the same size, i’m using the same number of them, and then stir for a set amount of time (with a timer.) If the balance isn’t right, I can adjust one thing for the next time. This is usually the stirring time, but is sometimes the amount of ice. Most bars will shake a cocktail because it is quick. I would never request a stirred drink if that’s not part of the bartenders workflow, but at home I almost always stir.