Tea flavor / aroma wheels reconsidered (with a Spanish version)

Early on in my blog history I considered the use of tea flavor or aroma wheels in tea tasting.  My main critique of those related to noticing that the flavor or aroma examples were all from temperate climate or "Western" examples, when that doesn't necessarily reflect a complete range of experienced flavors.

That was back at the end of 2013; what did I know of tasting tea then?  Or now, really.  Some tea reviewers tend to draw on Asian flavor references quite a bit instead (eg. jujube or Chinese date, tropical fruits, herbs more common to Asia or tropical regions), but those are still an exception.

a cool wheel version from a vendor who is no longer active

All the same it's more a starting point reference, a way to break flavor / aroma experience into categories (fruity, floral, earthy, mineral intensive, vegetal, etc.), with examples of what one might experience.  Both of the wheels I cited as an example then contained limited tropical fruit references, to be clear; that one issue wasn't completely one-sided.  It's odd that I'm not seeing bergamot or muscatel on those, now that I think of it.  "Citrus" makes it; maybe it just has to be broken out from that, in the case of the first.

I just ran across a reference of a wheel translated into Spanish by a tea enthusiast and blogger in Mexico, in the "Life Sip by Sip" blog (by Dennis Diaz, a translated version of the Tea Master's original, but as that earlier and unrelated Temple Mountain version shows the forms tend to repeat).  I talked to that blog author three years ago, when working on a post about tea culture in Mexico, but don't remember noticing that.  Or maybe I did; that was three years ago.  I'll add her impression here (automatically translated from a blog post):

Knowing the origin of the tea, infusing it correctly and without haste, as well as drinking it, enjoying the sensations, emotions and feelings that it causes us, leads us to conscious consumption.

When we drink a cup of tea, our senses receive information that is transmitted to the brain in order to be interpreted and translated into unique images about what we are tasting and feelings that, most of the time, are very pleasant.

Regardless of whether you are an expert tea consumer or someone who is just taking their first steps in the exquisite and wonderful world of Camellia Sinensis, the identification of the aromatic notes and the interpretation of these images and feelings generated will enrich the experience.

Maybe it is read [interpreted as?] complicated, but the secret is to be able to look within ourselves for what we feel and remember when having tea...

Limitations and reservations about this approach

I can't settle for myself if it makes a lot of difference, or is even better, to describe the aspects in a tea or not.  I do that in writing reviews, or course, but I enjoy just experiencing a tea without adding any analysis when I'm not reviewing.  Also I tend to appreciate the overall balance of aspects in any given tea, without necessarily focusing on the flavor as a singular main input.  I don't "get" cha qi or body feel aspects to the same extent many describe but a tea that is thin in feel or lacks aftertaste experience range also loses some appeal.

A recurring theme in this blog has been how different people would naturally interpret the same flavors in different ways.  It's hard to pin that down to one clear, representative example.  Another theme I've been noticing (not really related) is how some tea reviewers (bloggers, of course) tend to mention transitions between rounds in teas that sound as if they are really describing a dozen different versions of tea, while others assert that the main flavors remain consistent, with more shift in relative balance than new flavors joining in, with other range continually dropping out.

What to make of all that?  I'm really not sure.  I can't conclude that "it's all subjective," or guess that some people can accurately identify an objectively correct interpretation better.  Maybe it's somewhere in the middle; the patterns really are there, but interpretation plays a significant role.  In one of my favorite posts I asked a few blogger contacts and a tea maker friend for input about how that works out, in this post: "Imagination and subjectivity in tea tasting, with blogger and expert input."

There's probably no need to repeat ideas from there or try to capture patterns described; input on all that varied.  My wine maker friend leaned towards the take that variations in descriptions is not a problem, running counter to what I take to be the more standard general consensus in that field.  But then I took up and quit wine interest a long time ago (and never had got far) maybe that mainstream stance has shifted since.

Input on subjective interpretation from a favorite chocolate blogger in that post was interesting (I've only ever talked to one chocolate blogger; just being weird in that part):


Lisabeth's answer:  Certainly chocolate tasting is subjective to some extent. Everyone tastes food differently, and so will relate the flavours they taste in chocolate to different foods. In the case of Madagascar-origin chocolate, for instance, the chocolate is so clearly fruity that most people will identify that as a flavour in chocolate made from cocoa beans grown in Madagascar. But whether they taste raspberry versus grape or lemon, is a whole other thing. 


That definitely works.  I suppose grape could come across as similar to raspberry in the context of chocolate flavors, and since there really probably are a lot of complex aromatic compounds in the chocolate (as in tea versions) maybe the range of what is being interpreted really is broad.  This subject reminds me of considering what it would take for a machine to be able to "taste" tea in a post.  It does drift into how testing can identify compounds that map to known aromas (or flavors, in one sense of that term), with half that post was about mapping function projects relying on a human taster instead.

a Wikipedia aromatic compounds table; some degree of objective aroma interpretation is justified

Tasting practice and indirect reference of aroma wheels can help.  Scent training kits seem like the next natural step in that progression, but really one might naturally consider what the goals are prior to taking any of those experiential steps.  Or trying out a new approach with no clear goal could still be valid.

the Australian Tea Master's version, from here

Guidance on use of flavor determination in reviews or analysis

In the first draft version of this post it stopped there, with more about why it might not make sense to break out flavors (/ aromas) than about how it could be useful.  If someone wanted to review teas or just write down their own impressions it could make plenty of sense.  I'll add some thoughts on how that might go based on a half dozen years of doing it, broken up by range of what is typically experienced, those standard sections on those wheel.


Floral character:  a lot of tea is floral in flavor, it's just not simple to break that down to distinct flower types, at least for me.  I tend to specify if a floral aspect is light and sweet, like a wildflower, or slightly richer but still light in tone and sweet, like an orchid, or a bit heavier and richer, like lavendar or rose.  Jasmine I'm familiar with since it's around a lot here; that's distinctive.  Beyond that I'm more or less just guessing, or keeping that range of interpretation general.

Fruit:  this can be trickier than it would seem, since this is a really common food range.  Citrus comes up a lot in varying forms, and berry, raisin, and grape can be distinct, but lots of flavors tend to stay a bit vague.  Dried fruit tones are just as common as fresh fruit; dried pear and fresh pear are two different things, for example.  I think I do notice a lot more tropical fruits for eating a lot of them; seems normal enough.  Sun-dried tomato and related savory range can be very pleasant in teas (and I'll get into basic tastes versus aromas in the vegetal part following, since that's really about umami).

Spice:  cinnamon isn't so uncommon, but beyond that these can be hard to identify.  Aromatic spices reminding me of incense tones come up a good bit, or mild, earthy, subtle spices closer to root spices (like ginseng or sassafras).

Vegetal range:  this is probably the easiest range to identify.  Wood tones can vary a lot (cedar versus redwood or fresh hardwood, dark tropical hardwood versus mild coffee or leather tones), but those are all typically placed under "earthy" category instead.  Seaweed can apply almost as a default for Japanese green teas, but really umami is a taste picked up by your tongue, and not a scent based aroma.  It's interesting how the "super-taster" theme would seem to apply more to that scope, since it is usually used in relation to sensor count and sensitivity to tastes versus aromas, and having problems with tastes like bitterness, but the two themes get combined.

Mineral:  this often acts as a base for other flavor range, as I experience it.  It can be very hard to describe because it relates to how different rocks might taste or smell--not necessarily  familiar ground.  I did some desert hiking and rock climbing in earlier life phases and I do end up referring back to that, the smell of different rocks.  Artesian wells (naturally flowing versions) tend to express a specific range that can come up, a sweeter mineral taste.  I sometimes wonder if this isn't more typically a taste versus an aroma, even though it's not exactly within the sodium chloride (table salt) experience of salt range.

Earthy:  it's interesting tasting range outside of food scope, and trying out just about any shu pu'er leads straight there.  Tar or petroleum can actually be pleasant, although fresh asphalt not so much.  Peat can be a more neutral flavor range than it sounds.  Geosmin (like beet or dirt) comes up a lot related to one type of storage conditions environment for aged sheng pu'er.  Forest floor is a useful and common description, but to me it helps to narrow that to a wetter Spring or dryer Fall forest scent range.  That scent would also vary by the type of trees dropping leaves, and descriptions can be more or less detailed and poetic.  I'm not seeing coffee on either of those wheels, and that can come up, either as a light-roasted flavor aspect or closer to the char found in French Roast.

Char / fire:  closer to that last point, some variation of this is primary in many roasted teas, and some that aren't roasted.  Liu Bao tends to taste a little like Chinese barbecue pork, which is in between charcoal, slate, and a savory range flavor.  Tobacco isn't so uncommon, but I tend to think of that as earthy range, not so far from leather.

Sweet:  malt is under "Sweet" in that Temple Mountain wheel (and in the Tea Master's version; that's a good functional approach for that range).  Really malt seems to relate to a soft, sweet flavor like ovaltine or else a more mineral intensive related version in Assam (Indian black tea).


All that is just a starting point; people's experiences and interpretations would vary.   That last category of sweetness, and the umami and mineral cases, bring up the idea that part of flavor experience is taste picked up by the tongue.  It's easy to not notice that distinction in tasting experience, but it can be identified quite clearly by just plugging your nose while tasting.  Trying to review teas with a cold brings up the same theme; all of aroma drops out.

To me sweetness stands out as critical for helping some teas balance properly.  Other flavor range may or may not "work" if a compensating, balanced mix of experience isn't present, and in general more sweetness works better for supporting the rest.  Or maybe that just relates to my own sugar addiction.

Bitterness is an odd special case as tastes go, most common in young (new) sheng pu'er.  Per my own experience almost any presence of that aspect is negative, until you acclimate to it, and then after that it can be perceived as positive.  Experience would definitely vary tied to personal subjective preference related to that, even more so than for any other aspect range, even though it's really all a bit subjective.

One last point:  if I keep saying that taste / flavor / overall effect preferences are subjective why does it continually come up that reviews or descriptions are presented as objective?  I think there are multiple parts to that.

Common convention could lead to a form of objectivity; shared subjective agreement.  I just went through a long discussion of how Dan Cong versions are most typically lightly oxidized (relatively speaking, in relation to other oolong range), and low to medium level roasted, less so than Wuyi Yancha versions tend to be, because that's widely regarded as an optimum for the tea type potential.  I'll skip the rest of that discussion and leave off here as if that's actually made the point it barely brings up: a 'best" range can be established by convention.

Trueness-to-type is really related but slightly different; matching a style for a specific tea version and positive aspects for that type isn't exactly about personal likes, or those shared by a group.  As an example, high mountain Taiwanese oolong really should be intense in flavor, with pronounced floral range and underlying mineral (a specific form of that), rich in feel, with an extended aftertaste experience.  Someone might love or not like creaminess comparable to milk or butter in Jin Xuan (cultivar) based versions, but it's typical for that tea plant type.

Neither of these scopes are going to be clear based on just trying a single tea; identifying them would require exposure to group norms (in the first case) or other tea-type examples regarded as ideal versions.  Individual preferences can still vary in any number of ways but both these contexts set up groundwork for more objective forms of interpretation.

Postscript addition, related to coffee:

Peter Jones, a manager at the Boulder Trident Bookesellers and Cafe (coffee and tea shop), and a tea and coffee enthusiast, mentioned how this relates to coffee tasting and terminology development in a Facebook group post comment.  He only scratched the surface there, but as one might expect coffee is a bit further ahead related to use of standard terminology, tasting approach, tasting process documentation, etc.  Wine tasting surely is further along too; tea enthusiasts and professionals get it that some parts of the tea tradition are very old and developed and others, the more Western-facing aspects, are younger, still being worked out.

I won't run through a long tangent about coffee tasting here, or how that related background goes.  I'll mention an interesting article on that (A Guide to the Coffee Taster's Flavor Wheel), and include an image of a more recent version of one, what I take to be the current standard version:

According to a Specialty Coffee Association background reference use of images of that wheel is open but also restricted:

The Coffee Taster's Flavor Wheel by SCA and WCR (©2016) is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. 


Anyway, that file version was from here, with a little more background in that article.  The previous wheel version, discussed in that guide to wheel use article cited earlier, included a separate wheel that added a second complete version covering common flaws in coffee.  Just great! 

There's no short take on how these references are more developed and functional than the tea versions; it seems a terribly long and involved subject.  This World Coffee Research link (the other organization that developed that new version) leads to more background, if it is of interest.  Or this SCA article says just a little about use, which might be general enough that it's still of interest to people tasting other things, like wine, or tea.

Is there one best oolong brewing temperature?

This is about there being an objective best case versus subjective preference variation related to water temperature used to brew oolong.  That's not catchy, is it?  Let's try the popular media version:

Do you like your oolongs steeped in boiling point water, or use water that's a bit cooler instead?  Some people claim there is one right answer.


Of course personal preference relates to some degree to every choice and sub-theme related to tea.  Or does it?  Are tea bags generally inferior to mid-range quality loose tea?  I think so.  I guess you could still put those teas in a bag and they wouldn't turn out so differently, but that's complicating the real issue, the actual generality.

Would most people agree that green tea should be brewed using water at less than boiling point temperature?  Sure.  Even an issue like that one, selected to have one clear answer, might not be as objectively clear as it first seems. 

As a bit of background, different references say different things about what temperature to use for oolong.  Check out this version from the Reddit tea subforum page:

There it is then; brew oolongs at 85 C / 185 F, we're finished.  Or maybe not.  That guide may apply better to Western style brewing than Gongfu style, even though one doesn't typically encounter the idea that temperature choices would vary uniformly between the two approaches.  They can, though; along with using a higher proportion of tea to water using much shorter infusion times can serve some of the same purpose as moderating temperature, related to offsetting astringency or other aspect range that could be experienced as negative. 

Since this subject came up related to a tea group discussion let's start from there instead.

Brewing oolong at boiling point temperature

Someone asked about staff in Taiwanese tea shops universally brewing oolong using boiling point temperature water in the Gong Fu Cha FB tea group.  This triggered an interesting discussion of whether or not water at that temperature really is best for preparing oolongs or not.  There are lots of references and tables out there related to standard brewing parameter suggestions (I wrote about that awhile back) and those vary for oolongs.  To summarize, several different opinions emerged:


-brew oolongs using boiling point water temperature, obviously (about half the responses, or maybe just over that).


-only flawed oolongs will give better results at lower than boiling point temperatures, related to offsetting flaws; use boiling point temperature water for every high quality oolong type.


-only brew the lightest versions at lower temperatures (the least oxidized, in this context, although the styles discussed also tended to be less roasted too).


-brew oolongs at whatever temperature you prefer the outcome related to; subjective preference for specific aspects is the ultimate guide.


The argument for the "only use boiling point" option is that tea components extract differently at different temperatures, and for better quality teas--combining the first two answers, sort of--that's really the optimum.

I was arguing for the last, more or less, but probably as much related to defending people's option to prefer whatever parameter they want.  I do tend to brew most oolongs using boiling point water, although I also shift parameters quite a bit, to see what happens related to that.  Per my understanding it does just vary the aspect range a little if you use slightly lower temperature (90-95 C, 195-205 F).  When experimenting with different oolongs--they come in quite a range of types--some do seem better prepared at slightly cooler temperature. 

Of course it is hard to exclude quality level and flaws in specific teas as a factor related to that, and it's impossible for me to taste a tea using someone else's aspect preferences (their sense of taste, in the one sense).  I guess it's conceivable that I'm just not drinking good enough tea, or that my own flawed preferences are a problem.  That sounds like I'd be joking (and it can be hard for even me to tell when I am or not), but preferences do tend to shift over time, and I'm definitely not claiming to have arrived at a God's eye perspective on tea, just somewhere in between only getting started and having sorted some things out.

all different oolongs, but only part of the range

For Dan Cong it depends on the character of the version; some tend to be softer, and boiling point water is fine, and if that characteristic astringency is a bit more intense using slightly cooler water moderates that (versus using flash infusions; both will work, but the results aren't identical).  I'm not so sure that would be a case of a tea considered as flawed, since those teas vary by lots of factors, but of course teas come in a range of quality levels, and it does seem like less of an issue for better versions, so maybe it really is that simple.  I've tried versions sold more as commercial, mass produced Dan Cong that were different in quality and character, but they vary by plant type / aspect range in better versions too.  Ya Shi (duck shit) tends to be softer and more subtle than Mi Lan Xiang, for example (or also aroma based versus flavor intensive, per a different way of using some terms, but that also introduces a second set of concerns).

One comment by Thomas Smith attempted to explain an actual difference in using slightly cooler water, about changes that go along with variation:


[Brewing at boiling point] Definitely pushes more taste to the forefront, bringing it more in line with the intensity the aroma can provide. The cost seems to be a reduction in the ease of distinguishing taste characteristics, a skew of sourness over sweetness, a move to flavors closer to darker leafy greens and citrus in the lighter oxidized teas (as compared to romaine/endive crispness and a fruitiness closer to pomes and drupes), and a slight comparative dampening of aroma in secondary and tertiary infusions. Can provide more heft, though, which can be pleasant.


Interesting!  He added more to that, within the context of additional discussion:


Hotter water can relinquish more solutes more readily, but can also more rapidly force volatiles out of solution, leaving the aroma potentially less complex by the time it is consumed. Certainly not a huge issue for many wulongs, but a factor in some. And, as indicated earlier, oversaturation can diminish one’s ability to actually perceive the full complexity that can be achieved. The compounds that are soluble at appreciable levels with 100 degree water when steeped at typical short brewing times are also soluble at 85 degrees, given slightly more time. There is merit to experimentation.


A couple of those are subtle points that could easily be taken the wrong way, extended to mean something they seem not to intend, or to reach specific conclusions that really aren't being argued.   As I took it he's not really making a case for or against brewing at any temperature in those comments, instead talking around related factors.  Anyone interested in running through all the points made on either side--or the various sides--can take a look at the discussion, since individual comments and points would be in response to the others.

Related to green tea brewing in Vietnam

The grounding of any optimums in tea preference is problematic.  Even for that same point related to green tea, for which almost no one is ever going to advocate use of full boiling point temperature, it's my understanding that it is completely conventional to brew green tea at that temperature in the Vietnamese tea tradition.  The idea is that some people there love the astringency and bitterness that results from that practice (or so I take it).

Of course you can't generalize a whole country like that, and it's likely that not only are some people influenced by lots of factors (foreign traditions, personal experience, and whatever else), there are also regional variations in both teas and brewing conventions.  I just talked to Huyen about that (that friend who sent some Vietnamese tea samples), and she confirmed that there are really two schools of thought about green tea brewing temperature there.  What she actually did say:

green tea brewing at boiling point in Hanoi

With green tea: there are 2 ways brewing. First, if you want to focus about the flavor, you can use boiling water at 95-98 degree.  But if you want to focus about the taste, you can brew at 80-85 degrees.  The time to steep tea is 20-25 seconds.  That’s my way to brew tea.



Someone might justifiably claim that flavor and taste are synonyms, so opposing them doesn't work (or the same would apply slightly differently for the Chinese description opposing flavor and aroma, really, since taste is tongue-based but most of flavor is really aromatic component based, that is, aroma). 


I think what's going on in both cases are references to real, valid distinctions, just framed in these other cultural and language based contexts, which aren't translating well.  Once you try two versions of the same tea, one described as flavor-intensive and the other aroma-intensive (not related to Huyen's comment, still on the Chinese descriptions framing), then you know exactly what that means, and it makes sense.  My take is that there's probably more of a translation issue going on, that these are extended from use of Vietnamese terms.


If the distinctions in what we taste with our tongue versus what's picked up through aromatic compound sensation--and the related terminology--isn't familiar I'd recommend Barb Stuckey's video reference (the author of Taste What You're Missing; that would also work for further reading).


Related to it being one of two common conventions to use boiling point water for green tea in Vietnam (per this account), then could we claim that a broad part of the entire local tradition and those individual shared preferences are somehow wrong?  If the benchmark is inter-subjective agreement among Western tea enthusiasts then maybe we could; it might be that they're doing it wrong.

Or maybe not.  I ran across a reference source claiming that green tea should be brewed using boiling point water not so long ago (not expressed quite that simply, in a Tea Master's blog post on green tea).  This could be just be my impression but there seems to be a broad, general divide in tea preferences between people claiming that any tea should be brewed using boiling water, and very short infusions, with others holding more to the ranges expressed on charts and such.  It may be that the generality correlates quite a bit with regions or branches of tea traditions.

really nice Assam oolong (reviewed here); but as well to not complicate this

McDonald's hamburgers as an analogy for rejecting subjective preference

An influential and experienced member of the tea community provided an example of a broad inter-subjective agreement that we could define as wrong:  if a group of people were to choose the objectively best burger alternative between McDonald's versions and a high-quality fresh-ground version (Angus beef or the like), and some chose McDonald's versions as their favorite (which would actually happen), couldn't we say their preference is wrong?  Phrasing for that gets tricky, doesn't it?  It's hard to express "best" in such a way that it's clear that you don't mean "according to personal preference," since that is one thing that would tend to mean.

Why is McDonald's burger version clearly inferior?  Let's start with the beef itself:

Thankfully, McDonald’s and several other chains recently stopped using the “pink slime” in their beef. But the vast majority of fast food beef comes from CAFO (concentrated agricultural feeding operation) cows. Not only is this horrible for the animals and the environment, but eating meat from sick animals will only make you sick.  Eat a McDonald’s hamburger and you might be getting a mouth full of antibiotics, hormones, and dangerous bacteria.


Beyond all that (some of which you may not actually be able to taste) there are over 50 questionable chemicals and additives in a single burger (including the other ingredients), per that article, and it's not even possible to order those cooked to specific temperatures.  They're just not great.

As I interpret this we're not comparing McDonald's food to the local steakhouse burger version, per the tea brewing temperature analogy; it maps closer to the "best burger in Bangkok" debate that never completely stops in a local FB foodie group.  Maybe somehow one of those burgers really is best, but it seems more likely that most could agree some of the outlier opinions are based on an odd preference range, or perhaps limited exposure to the others.  Maybe it would be possible to come up with a short list, to identify some answers and preferences as more correct.  No one even jokes about McDonald's being best in that group; if they were having a bad day the admins might ban you.


I'm not really finished with mapping out all of this oolong brewing temperature issue, but as I see it there is no end to it.  It seems likely that as with exposure to a broad range of foods--corresponding to McDonald's foods lovers trying a range of fresh food alternatives--people would gravitate more towards consensus opinions.  I can't say if everyone would only use boiling point temperature water for oolongs, with enough brewing experience, and if using the highest quality tea versions.


A friend brought up a related point when talking about tea-houses in Taiwan a couple months ago; to him they tend to brew tea too strong, seemingly appreciating the feel and aftertaste aspects in the tea more than the flavor, and ruin the effect of the taste by over-brewing, per his preferences.  That was seconded by a relatively identical comment in that discussion.  That doesn't map directly onto the hot water versus slightly cooler debate for oolongs, but some of the same concerns could just be playing out a different way. 

Another friend added a thought that muddies the waters just a bit further:  a cigarette smoker would be more inclined to favor stronger flavors, perhaps even flavors in a certain range, and this could push them towards distinct preferences (eg. oolongs brewed at boiling point, versus liking some types of oolong brewed cooler).  It's sometimes mentioned that a lot of exposure to tea leads to preferences skewing in certain directions, not just related to being more right, to fine-tuning awareness, but also to natural progressions occurring.  I've discussed this in relation to de-emphasizing flavor as a main factor in the past, but I've seen it mentioned related to flavor range too.

another interesting non-standard oolong, from Toba Wangi in Indonesia 

Conclusion

To me there is an interesting tension between two competing ideas: 

1)  personal preference ("taste") is individual and varied, and differing opinions can be valid

2)  with enough exposure and progress along a natural preference curve there are consensus right answers related to patterns of agreement (objectivity in the sense of inter-subjective agreement). 


I'm not really arguing for or against either, I just don't see one simple answer as identifying the facts of the matter in most cases related to tea preference.  Sometimes it works better to say "beginners tend to like that aspect range, more experienced tea drinkers this other," but for any one outcome it's rarely that simple.

Since lots of people gave a lot of great input it's also worth reading that group discussion, in light of these sorts of concerns, of course also based on one's own experiences in brewing oolongs.