“Complexity” in Sour Beer – Objective Reality or Meaningless Rationalization?

If you spend any time around sour beer brewers you will inevitably run into brewers (and may even be one yourself) who does not like kettle sours. Many brewers of traditional styles are quite sour about these quick turn-around beers, with brewers Gose being particularly salty about the very concept of quick souring. And, as is common when someone has a strongly held opinion, many of these individuals look for “objective facts” to rationalize their belief. One of the more common “facts” – and the one I find the most irritating – is the concept of “complexity”, and how traditional sours supposedly have it, and kettle sours supposedly lack it.

When these discussions come up I often find myself pushing back. Which is ironic since I brew about a 6:1 ratio of traditional sours:kettle sours, and generally prefer the former. There are, of course, a lot of other claims made about why kettle sours are supposedly inferior. But most of these (e.g. claims that every kettle sour out there is full of diacetyl and butyrate) are so absurd that they don’t stand up to even the most limited scrutiny.

So onto “complexity”

Complexity – Does It Even Matter?

I’m going to address the issue of defining complexity in the next section, but before going down that road, I think its worth asking a question that is almost never asked in these discussions.

Is complexity always a good or desirable thing?

Clearly, those using “complexity” as a knock against kettle sours think so, but that doesn’t make it true. A well brewed Bohemian Pilsner is the epitome of simplicity – a single malt note, an intense but narrow hop flavour, and minimal yeast character. And yet, this is a beloved style where any additional flavours or aromas – e.g. more complexity – is considered a fatal flaw.

Many of the kettle sours I brew (and the ones I prefer to buy) take a similar approach to flavour and aroma. The goal isn’t a complex mix of flavours. Rather, the goal is a dominant “feature note” which is supported and amplified by the underlying beer. Fresh fruit flavours, intense dry hopping, and intense spicing are common in kettle sours, with many beers highlighting a single ingredient. The same “feature note” approach is not common in traditional sours, and there is a good reason for this. Adding intense flavours to traditional sours almost always leads to clashes with the fermentation character. Complexity, in this case, leads to a muddled mess.

So in many cases, measuring a kettle sour by its “complexity”, is to use the wrong yardstick. You’re measuring it by something that it is not intended to be – akin to trying to measure the distance from London to New York in degrees Fahrenheit.

Of course, there are also kettle sours that do aim for complexity – e.g. combining fruits and spices, multiple hops, and even adding additional organisms to the fermentation. For these beers, where complexity is a goal, than the yardstick set by traditional sours may be apt…

…but what is “complexity”?

Defining Complexity

Discussions of a beers complexity often fall apart because no one can agree on what “complex” means. And that’s because our perceptions of “complexity” are largely subjective. It a classic case of “I’ll know it when I see it” – an attempt to categorize something, even though the category is subjective or lacks defined (or definable) parameters. In simpler terms, in these discussions each person has their own internal concept of “complexity”, rather than there being an agreed upon standard.

So can we define “complexity” in an objective manner that lets all of us measure complexity with the same yard stick? I’d argue “probably not”, but that doesn’t mean that people haven’t tried. In the food sciences (including beer, wine, whisky and other fermented/distilled beverages), two approaches tend to be taken:

Chemical Complexity

With modern technologies like GC/MS it is possible to take a sample of a beer (or other food) and detect the presence and concentration of nearly every chemical present in that beer – in some cases, identifying substances present in the parts-per-billion range. On the surface this would appear to be a completely objective measure of “complexity”, as the beer with more chemicals in it would be the one which is more complex.

Or would it?

The issues with this approach are two-fold. Firstly, just because a chemical is there doesn’t mean we can taste it. Some chemicals are flavour/aroma-less – i.e. we can neither taste nor smell them. Some chemicals are antagonistic in our ability to taste/smell them. For example, ferulic acid – commonly found in beer – will reduce your ability to perceive bitter-tasting compounds. And some of those compounds may be present at levels below our ability to detect.

The second issue is that widely different chemicals can taste more-or-less the same. Take these four compounds, all of which can be found in beer:

These are very different chemicals (ester, thiol, terpenic acid, and cyclic monoterpene, respectively), but all produce a citrus-like flavour and aroma. So you could very well have a beer that was chemically complex, and yet in perception, is one-dimensional.

Flavour Wheel

That chemical complexity doesn’t equal flavour complexity has long been appreciated by food scientists. Which is why flavour and aroma wheels are often used by food scientists (and enthusiasts) to formalize the perception of a food. The interpretation of the aroma and flavour is still subjective – your palate will perceive flavours differently than mine – but these wheels do allow people to “normalize” their palates to a defined and consistent standard.

We could also, arguably, use this to measure complexity – a beer that is scored as having more of the flavours/aromas from the wheel would be more complex.

But its not quite as simple as this. Look closely at the flavour wheel to the left – in it, flavours are grouped by similarity. E.G. the various flavours linked by the orange arc are all fruity/floral. Would a beer with five identifiable flavours, all in the fruity/aromatic (orange) section, be more or less complex than a beer with “only” four flavours, but where each individual flavour is from a different “flavour group”?

So…Complexity?

If you made it this far than I think you appreciate the issue of defining complexity. Truly objective measures like chemical complexity are unlikely to reflect the complexity perceived by the drinker. Methods to measure the perception of complexity will have subjective elements.

Complexity – however you define it – is a product of ingredients and methods. Traditional sours tend to get most of their complexity from the latter, while kettle sours tend to draw more from the former. Because of this, you can design and brew a kettle sour every bit as complex as a traditional sour (measured either chemically, on a flavour wheel, or elsewise) via ingredients. Take, for example, my Gose of Christmas Past, a Gose-inspired quick sour featuring spices, fruits, and an expressive yeast. The layering of flavours in this beer, and the way the beer evolved as it warmed in the glass, and aged in bottles, was every bit of complex as any traditional or mixed-ferment sour that I’ve made. You can also brew fairly one-dimensional traditional beers as well. In fact, in my my experience most traditional sours tend to be one dimensional out of the fermenter, with the “complexity” these beers ultimately have arising more from the blending of these beers as it does from the fermentation itself.

Those Goalposts Have Legs!

The final point I would make about this controversy is that of the ever-moving goalposts you encounter when you challenge a traditional brewers view of “complexity”. Point out that you can use ingredients to build complexity in a kettle sour, and you’ll immediately be told that you have to use the same base beer for the comparison to be valid.

And, of course, the base beer that you must use for the comparison is that of a traditional sour.

Or it’ll be claimed that those other ingredients are somehow irrelevant.

Or it’ll be claimed that using “adjuncts” is cheating.

Or it’ll be claimed that only fermentation and/or aging character counts when measuring complexity.

Or you’ll get the stubborn “no matter what, it’s less complex”.

So What Was The Point of All This?

Does this post have a point? Or was I just an angry old man yelling at passing clouds?

If there is a point, its this. Its okay to not like a beer style or brewing approach. It’s okay for the reasons of your dislike to be matters of subjective taste or preference. But don’t go around pretending your completely subjective preferences stem from some sort of objective fact. They are opinions, and like all opinions, they are neither universally shared, nor do they stem from a universal and objective truth.

Remember – opinions are like assholes. We all have them, but its what you do with them that determines if yours stink.