Categorizing mixed drinks

Cocktails and mixed drinks
21

To me, one of the main weaknesses of the timeline I drafted is that it fails to address the wellsprings and emergence of today’s mainstream contemporary mixology.

To be fair, it’s a squirrelly one.

  • heavily influenced by culinary fashions, “gourmet thinking”, and high concept
  • overwrought, often under-refined and under-delicious (meh/ pointless or whacky-for-the-sake-of-concept)
  • generally pretty expensive
  • cocktails as proprietary and principal product line alongside the appetizers, main courses, and desserts
  • Instagram-ready (look at this whacky drink!)
  • too many ingredients and/or too many weird ingredients, and/or too many infused/modified/home-made ingredients
  • ephemeral; difficult-to-impossible to replicate (although in most cases, nobody would even think to try)
  • ubiquitous

By its definition, there are no famous representative examples to point at.

As far as I know, we haven’t even gotten to naming this stuff. The closest nominee I’ve heard yet was from Jeff Berry: “stunt drink”.

2 Likes
22

The draft timeline, above, is now an exhibit on the Cocktail Kingdom Library web site. While it’s undoubtedly rife with imperfections, I think it’s pretty effective, and I can refine it in the future.

3 Likes
23

I love when I drink and eat at the same time

24

Hello @martin ,

First of all, I would like to tell you that the work you did with the comparison table and the timeline is amazing. Thank you for dedicating your time to this topic that we are so passionate about and for sharing that work with us.

Now my question is related to Grog. I see it frequently in cocktail books in its cold version, its hot version and even in tiki versions so I would like to know what is your opinion about it.

Can it be considered a separate cocktail category, and if not, which of the above categories would it fall into?

Thank you very much and greetings to the whole community.

1 Like
25

I think this would be my take on grog, with some help from Wayne Curtis’ bit in the Oxford Companion:

Originally, grog was just watered-down rum. The word emerged in a particular circumstance: within the British Navy. Outside that context, it developed a life of its own as another word for alcoholic drink, along the lines of hooch and booze. As Wayne put it, Donn Beach appropriated it in the 1930s and 1940s for various punches on the menu at his Don the Beachcomber’s.

So, my point of view would be that no, a “grog” is not a category, it’s just a word for an alcoholic drink that carries some convenient naval/seafaring associations.

1 Like
26

My Total Mixology project has presented an interesting opportunity to adapt the idea of drink categorization for database use using rules. One critical aspect is that categories can overlap so that drinks can be members of more than one category, rather than being rigidly this or that. My approach is inspired by history, but not bound by it. The main goal is to be useful. So far, I’m up to 29 categories, the most recent being spritz.

To use spritz as an example, the rules—as of this writing—are:

  1. wine or liqueur base (i.e., largest quantity alcoholic ingredient)
  2. lengthened by 5% or more (i.e., some sort of added water-based lengthening agent)
  3. effervescence must be present (typically, this will come from the lengthening agent, but it’s a little loosey-goosey right now)

This strategy has netted around 150 drinks (out of ~3500) including recipes that are explicitly Spritzes and many that are not, but are at least in the general ballpark of a spritz. Ultimately, I may need to add an additional criterion of some sort or massage the existing ones to tighten the net a little, but I think it’s interesting how well this approach can work.

1 Like
27

I am still riding Martin’s Index for all it’s worth and haven’t upgraded to Total Mixology. But to be argumentative, why wouldn’t water-diluted spirit—grog, fine a l’eau, scotch and water—stand as its own category? And could it be lumped with “on-ice”?

28

It just might be its own category, although I wonder if it is interesting enough? Is scotch and water even a recipe?

29

[EDIT - That got really long, sorry! Hope it makes sense]

It’s such a hard puzzle to solve and it really comes down to where your line is on what qualifies as different. If Grog isn’t a category, then I would argue that neither is a toddy, cobbler, swizzle or sangaree (all just watered down booze with occasional sugar added, so let’s put them neatly under Sling).

I’m trying to solve this quite actively at the moment as part of a training piece going on top of the Barchive Project so am reading into a lot of lateral references for simplifying large unwieldy systems. (Carl Linnaeus and biological taxonomy, Darwin’s theories of evolution, James Murray’s approach to the Oxford English Dictionary, Dmitry Mendeleev and the Periodic Table, etc.)

The problem with drinks is similar to the Boundary Paradox of biological taxonomy where Binominal Nomenclature (Homo Sapiens, Tyrannosaurus Rex, etc.) breaks down due to it’s linearly hierarchical nature. Once you take into account multiple dimensions (DNA compared to the original approach of just physical appearance) it quickly gets out of control.
Similarly in mixed drinks, as you pointed out when referencing @Splificator 's work, there was the era of ice changing things up and later the move to “fancy” as the posh wine drinkers started moving out of the private parties to the public drinking rooms which brought in another glassware revolution. We now have ingredients, ice and glassware on multiple dimensions and this is before considered that the ingredients again have their own system where a decision must be made on which dimensions to use (ABV, raw material, sugar, titratable acid, bittering, flavouring, etc.) and when it A is different from B.

This moves nicely into the concept of clades (now used in biology for more specific DNA approaches beyond the Linnaean taxonomy) where an entity can have multiple, previously considered different, ancestors. The problem is that this further complicates things because then you have almost infinite variations. This new approach is what’s caused the (technically true) conclusion that there is no such thing as a “fish”. I’ll get back to that in a second…

This contrast between high level concepts and very fine tuned specifics is also called the “problem of universals” on how to define objects. Plato argued that truth only exists on the plane of abstraction (Ie, tidy high level conceptual boxes such as sours, cocktails, highballs) whereas his pupil Aristotle was on the far opposite side stating that only by studying the finer details do we understand truth. (Ie, a Dry Martini with Tanqueray has “nothing” to do with a Dry Martini with Beefeater because they are different ingredients via different ABV, botanicals, etc. so should be considered differently)

Ultimately it comes down to what you’re trying to achieve with the groupings. If it’s for a “search” problem then I’d love to chat nerdy about it but I feel this is the wrong forum for that. (Very happy to discuss this separately though. I think something like word2vec for semantic word understanding would be a “break through” if done purposefully for food science…)
If, on the contrary, the goal is teaching and education then I am firmly in the camp of abstractions with fewer dimensions.

Back to the “Fish” not being real.
The modern age of DNA has concluded there is no such thing as a fish due to multiple species having separately evolved to survive under water as it’s hugely beneficial from a survival perspective. Does this mean the word “Fish” is now meaningless? Of course not! It’s a super helpful concept for teaching kids (and discussing among adults) about aquatic animals.
Similarly, “Cocktail” is traditionally well defined but is nowadays basically a synonym for the term “Mixed Drinks”. Same goes for “Martini” although that’s glassware dependant.
Does this make the terms “Cocktail” and “Martini” useless? Absolutely not (But it does complicate discussing drinks categories with guests…)

What Gary Regan did with the “Birds of a Feather” chapter in Joy of Mixology followed by the Cocktail Codex by the Death & Co guys has been a brilliant foundation, but it’s still something that has big room for further development.

For learning purposes it’s useful to go down the route of actual taste (the sense, not flavour perception) as a foundation. Almost all drinks have sugar added in some form so a foundation of balance with the Sour (Punch) and Bitter (Cocktail) category works as have been pointed out by many others before. Another balancing element to sweet is fat. Cocktail Codex referred to these as “Flips” to cover all egg/dairy drinks and UKBG had a “Zooms” category which was spirit, honey and cream. Gary Regan didn’t have a category for covering this use of dairy.

Peche Barchive - Zooms (Spirit, Sugar, Cream)
Peche Barchive - Zooms (Spirit, Sugar, Cream)976×344 284 KB

It sounds silly, but those 3 (Sour, Bitter, Dairy) cover the highest level of conceptual balance understanding. Then there’s two primary ways to go from here;

  1. We can look at changing up the sweet element which gives us Sidecars, Margaritas, etc.
  2. We can look at defining the boundary between the 2 key ones (Sours, Cocktails) which bring us to drinks such as Crusta, Casino, Journalist.
    (When defining this middle category we are also conceptually moving onto split based Cocktails such as the American Trilogy. Unless “Spirit” is a satisfying category, in which case “Sweet” or “Bitter” should also be and now we’ve broken everything)

I’m assuming most of us on this forum (And a lot of senior bartenders) would love to dissect in detail the specific nuances between using 5ml more of less of various citrus juices with different ice in different glasses and declaring them different drinks with varying names such as Gin Fix, Gin Sour, Gin Crusta, Gin Thing, Fitzgerald, West Indian, Bennett, Fresh Gimlet etc. but in the world of usefulness this is terrible.
It’s fun, yes! But terribly unhelpful…

And as much as I love spending hours digging into the internet archive or newspapers.com, I have to stand with the simplistic foundations of Carl Linnaeus when he wrote in the introduction to his first edition of Systema Naturae:
“The first step in wisdom is to know the things themselves; this notion consists in having a true idea of the objects; objects are distinguished and known by classifying them methodically and giving them appropriate names. Therefore, classification and name-giving will be the foundation of our science.”

Mental models, abstractions and neatly labelled categories just always wins when it comes to usefulness for learning.

3 Likes
30

One of the really interesting things the Madrusans have done with their Cocktail Companion is to formally deploy a modern drink typology that reuses, extends, and augments historical drink categories that were mostly forgotten in the 20th century to make them more relevant, today, particularly in their style of bars.

Collins: spirit + lemon juice + sugar + soda
Subsumes the Coolers. A popular and expanded drink form in the Petraske diaspora. (Notably, in their hands, these drinks tend to not be large scale thirst quenchers, rather a kind of iced fizz, sans egg white, with high flavor concentration.)

Rickey: spirit + lime juice + sugar + soda
What was one drink is now a home for Collinses and Coolers made with lime juice.

Buck & Mule: spirit + citrus + ginger syrup + soda
The buck didn’t even really qualify as a category, historically, and the mule was a lone drink, but given the fondness for ginger syrup in the Petraske diaspora, they now have a large repertoire to document.

Highball & Mixed Drink: spirit/wine + soda/carbonated mixer/juice
I feel their pain—these drinks gotta go somewhere. Stuff like the Bloody Mary winds up here.

Cup & Punch: spirit(s) and/or wine, liqueur(s) + citrus juice + sugar + spices + (optional water)
Reflects that contemporary punch is beloved by the cognoscenti, but far removed from its original megacategory status. Even Further diminished, as many punches are carved out of this to their own, separate “Tropical” category (below)

Fizz: spirit + lemon + sugar + egg white + soda
Refocused category that, again, reflects the Petraske school’s undying devotion to the egg white drink (for which I am most grateful).

Sour: spirit + citrus juice (predominantly lemon) + sugar + (optional) egg white
Fairly obvious, although the Daiquiri and its variations have been spun off into their own category (below).

Crusta, Daisy & Sidecar: spirit + citrus juice + liqueur(s) and/or syrup(s)
I’m still not totally sold on “crusta” as a catchall for any daisy-ish drink with a sugared rim, nor that the Sidecar is a clear drink category (they uncomfortably treat it as halfway between a category and break-out like Daiquiri or Negroni). Sort of begs the question: if the only difference, today, between a sour and a daisy is whether there’s a liqueur or flavored syrup in it, why does the distinction even matter?

Gimlet & Shake: spirit + lime juice + sugar
I apparently don’t run in the right circles to be familiar with the “Shake”, but the Gimlets seem to all be gin drinks, and the same rough idea with other spirits is a “Shake”. The only commonality seems to be sweetened lime juice and/or lime cordial, so this is sort of like the expansion of the Rickey (above). The drinks could otherwise be Sours or Daisies, so it’s a little awkward. But still kind of interesting.

Daiquiri: rum + lime + sugar
Drink to category. They have a lot of drinks to discuss, here, but they’re mostly a mix of sours and daisies, so I’m not sure how I feel about it.

Batida: spirit + lime chunks + sugar
Actually a rough template for an umbrella of “rustic” drinks that typically involve muddled or shaken fruit pieces and such. This needed to happen. It’s nothing new “south of the border”. It’s one of the smallest recipe sections of the book, and it could probably be much longer.

Fix & Swizzle: spirit + citrus juice + sugar and/or liqueur and/or syrup
The key detail omitted in the formula is that these are all crushed ice drinks. This all makes total sense for these recipes and how they are typically deployed. Again, Petraske school obsessions are showing through, here, and I generally love it.

Cobbler: wine, fortified wine and/or spirit + sugar and/or liqueur + (usually) citrus juice
Another crushed ice drink category that pairs with the above. Nobody has done more for the modern cobbler than the Petraske diaspora. :heart:

Julep: spirit + sugar + mint
Smash: spirit + sugar + mint + muddled fruit
Stinger: spirit + creme de menthe
The Madrusans lump these together, despite offering different templates, because they are minty. At least the Juleps and Smashes have a lot of structural commonality, including crushed/cracked ice. I’m not sold on the Stinger and its friends being tacked on, but if not here, where?

Negroni: gin + sweet vermouth + Campari
Drink to category. There are now a lot of them, they have to go somewhere, and it is commonplace to describe them all as “negronis”.

Martini: gin/vodka/genever/pisco/some agave/sherry/sake + vermouth + bitters
Long a de facto category of Martini-related drinks that briefly (1980-90s) expanded to encompass every drink before contracting again. As a practical matter, I am okay with it, but the genever drinks would belong in the Manhattan section (below)

Manhattan: whisky/brandy/some agave + vermouth + bitters
Separating these from the Martinis can get a little arbitrary, but there are a lot of drinks to deal with, and most are intuitive.

Old Fashioned & Sazerac: spirit + sugar + bitters
The closest we get to a “cocktail” category, and it is deep. Again, nobody has done more for the Old Fashioned than the Petraske diaspora. :heart: One of the Petraske peculiarities, however, is that there’s no place for, say, a Whiskey Cocktail served up—it’s always built in a rocks glass and iced.

Flip: spirit + whole egg or egg yolk + sugar
An easy one. Maybe the easiest?

Toddy: spirit + sugar (predominantly honey) + boiling water
A small section of obvious cases.

This winds up the Madrusan’s primary “branches” as they call them. There is no “Cocktail” group. No slings. No sangarees. No pousse-cafés (okay by me).

They then proceed with some bonus sections that avoid lumping a ton of recipes under “Mixed Drinks” but are probably only satisfying if you approach them from just the right angle:

Aperitivo & Spritz
Sparkling
Beer
Tropical
Coffee
Cream
Non-Alcoholic

31

To pull completely in the opposite direction of Madrusan, I don’t believe adding a lot of named drinks to a list does anything helpful to the concept of categorising drinks. (I love the book though and I think it’s a tomb many bartenders could learn from!)

The problem is of dimensionality which ends up being confusing to young bartenders.

Yes, a Negroni is different to a Martini which is different to an Old Fashioned which is different to a “Fancy” cocktail.

But if the differences between these are either individually named ingredients (Campari, Vermouth) or type of glass (Fancy Cocktail vs Old Fashioned), then they become non-comparable and non categorisable.

The Highball is the single biggest bullshit category we have for this exact reason. It’s defining a serving style, or glass, instead of anything related to the mix itself.
An Americano is an entirely different category of drink than a Screwdriver and pretending it isn’t is just lying.

Regan did a stellar job with adding sub-categories for both Sours and Highballs but still fell short with the rest of them because we us humans love to make connections where none exist. There’s a reason he ended up with a 27th category called Orphans for all the rest…

I’ve just started digging into Cocktail Theory by @Nosestradamus and I love that he tested out hundreds of ratios across the basic families to conclude that the balance of a drink is fully “taste” based and entirely separate from aroma and flavour.

It simplifies the concept of grouping drinks and gives us a neat 1st dimension to start on, from which we can (re)build further.

I finished this write up earlier today on the topic:

To build further upon the linked article, I think we have a clean 1st layer categorisation of:

Sour - Bitter - Creamy (See also my previous comment on the Zoom).

That gives us just three foundational cocktails to build everything else on…

Daiquiri - Old Fashioned - Zoom (Or Flip, if you prefer)

Two of these of course have major sub-variations within them, but I think treating a Daisy as something different than a Sour is just confusing.

If we are comfortable saying the Daisy is a Sour with a sweet liqueur, why aren’t we also comfortable with the equivalent on the Old Fashioned side for the sweet riffs like the Monte Carlo or Fancy Free?
I’d even argue the Martini came about as a twist on the Old Fashioned with Vermouth standing in for both sweet and bitter. The new fancy glassware just added an unexpected dimension we tried to define it by, despite a glass not fitting our taste receptors very well…

As for the Highball as a category, I think the above then makes it fairly simple.

There are two kinds, just like we have the sour camp vs the classic (bitter) cocktail camp.
We have an Americano which is a long Cocktail and we have a Collins which is a long Sour. Both are balanced with the same tastes as the Old Fashioned and the Daiquiri.

The Bloody Mary is a wild one, but ultimately it’s a combination of fruit, citrus and spices, which in this system makes it a Collins. (Or Punch, depending on which term makes everyone happiest on this proposed category of long sour)

(The Pousse Cafe is difficult for the same reason as the Highball is a terrible category… It’s a glassware/stylistic group rather than having anything to do with it’s composition)

4 Likes
32

Here’s a new one:

2 Likes
33

Thanks @martin! You’ve been faster than me, I just wanted to add it :smiley:

34

I started to go down a comparable route many years ago, but I lacked the vision and clarity of thought to make it actually work. Like you, the ingredients led the way, but I had too many “dimensions”, no clear or workable methodology for assigning magnitude, and I abandoned it. However, having once made the halting attempt, I can immediately recognize what you’re done here.
With your system, I don’t think the cubic visualization has much practical future beyond supporting the documentation, but from an analytic standpoint (categorical memberships), you’ve laid out a straightforward, simplified model I can probably work with, and most critically, it all boils down to terse, clear verbiage (‘low bitter, high sweet, low umami’)

35

This is incredibly interesting, @armin! I’ve been very interested about systems like this for a long time (though I haven’t thought about it as deeply as you or @martin have)

I’m curious about the way you’ve chosen to handle alcohol and/or dilution. The experiences of drinking an Americano and a Negroni feel fundamentally different to me, despite their non-overlapping ingredients having no flavor dimension in your system. You help clarify the distinction between a Gin & Tonic and a 50/50 Martini with the added “bitter wine” dimension at the order level, but that obviously doesn’t differentiate the Americano vs. the Negroni.

I realize a seventh top-level dimension would make visualizations much more complicated. And it doesn’t necessarily line up with the scientific basis of flavor receptors that your Taste Space is built on.

So I’d be interested to know your thoughts on how the perceived booziness of a drink should play into its categorization.

36

Armin can probably expand on this, but in the “Distinguishing between the globular clusters” section, he explicitly excludes dilution from the matrix. I suppose dilution (a complex topic on its own) could be seen as yet another dimension—not entirely unlike color—that is simply out of scope.

I don’t think there is or will ever be “one system to rule them all” or one that can cover every perspective and variable. You can have multiple systems and even employ them in combination to obtain higher quality results. For example, you could use Armin’s parameters to isolate all the sweet, bitter drinks, and then afterwards filter them by abv range, which is would be one easy approach to factoring in dilution.

1 Like
37

For what it’s worth, as a considerably simpler and (for me) useful categorization scheme, early on in my hobby I adopted/adapted the categories I found on the menu of the Anvil Bar in Houston:

  • Sour & short
  • Boozy & alluring
  • Herbal & spirituous
  • Light & long
  • Tropical & tiki
  • Bitter & bold
  • Odds & ends

Early on I appended “light and bubbly” (mostly for champagne-added cocktails that felt different than “long” highball-type drinks) and “cool and refreshing” (for lighter summery rocks drinks that weren’t really tropical). That structure has worked well for me for years. For one thing it’s easy to apply (I nearly always have an unambiguous sense of which box any new drink should be put). For another, these broad-brush categories feel to me like useful entry-points for narrowing down what drink I might make for myself or a guest.

Of course, the optimal shape of any categorization scheme is entirely dependent on the work it is intended to do. My grouping of recipes fits my needs but others might well need more complexity.

38

Akin to the recipe cube above, the Kindred Cocktails website set up a database of cocktail ingredients scored by how much they offer these flavor categories:

  • sweet
  • sour
  • salt
  • bitter
  • pepper
  • savory
  • herbal
  • funk
  • pie
  • spice
  • floral
  • nut
  • mint
  • smoke
  • citrus
  • tropical
  • berry
  • fruit
  • creamy
  • fizz
  • alcohol
  • whiskey
  • gin
  • wine
  • beer

Then the recipes get a score profile based on what each ingredient brings, and what share of the volume it represents.

This way, every drink entry on the site can show a list links to “similar cocktails” ranked by how closely their score profile matches that of the entry. It’s a system that has some promise but I confess it’s not one of the aspects of that excellent site that I find myself using that much.