Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Of course you can taste temperature differences and see color differences. The question is whether you can taste wine differences without temperature differences and color diffefrences.


This is a bit like having a taste test of frozen vegetables. But, yes, you can taste differences, and often those differences correlate to a certain "kind" of wine.

If you give me a wine to taste, I may tell you "this tastes like a zinfandel". And maybe it's not a zin. Maybe it's a pinot that was grown and processed in a way that ended up tasting like zin. Of course I can tell when intense blackberry/blueberry fruit flavors are flooding my mouth, and all the dissolved solids fill the crevices of my mouth and weigh down my palette, just like you can tell the difference between strawberry jam and raspberry jam.

The question is not can you taste a difference. The question is, can you guess what it was supposed to be? Hell no. Grapes with red skin can be used to make white wine. Of course I may be wrong about what grape it was, or what percentages of which varietals, or from what region, or year, etc. But I know when something tastes strongly of blackberries and when it doesn't.


> But, yes, you can taste differences, and often those differences correlate to a certain "kind" of wine.

Is there any evidence for this? Whenever I've read looked into studies on wine tasting, they invariably find that people can't reliably distinguish what they think they can.


Almost all the studies I saw were about price or quality, and I'm not saying I can discern either. If the wines are similar, I won't be able to tell them apart. But if one tastes like blackberry, one like honey, one like coffee, one is sweet, and one is sour or acidic, those are clearly different qualities and easy to discern.


Sure, but that amounts to saying "if a difference can be discerned then there's a discernible difference". I'm saying if there's anything really to wine tasting, there's got to be something that can be reliably shown in experiments - e.g. that sommeliers were found to reliably determine a wine's X, for some value of X. I've poked around for evidence like that, and there doesn't seem to be any.


I don't know what you're searching for, but with conditions like that I'm pretty sure you won't find it. It sounds a lot like you're trying to look objectively at a subjective art form.


> a subjective art form

Lots of people claim that it's not just subjective, and that's what I'm asking about. You said as much in the post I quoted and replied to.

I mean: if someone says Bordeaus pair well with croutons, obviously that's an opinion. But if they say a wine's taste discernibly correlates to its region or vintage, I don't think it's unreasonable to ask if there's any evidence.


Those people saying it's not subjective are wrong, because wine is far too complicated for a human to decipher.

The fruit/veg is the output of a plant's function, whose input (simplified) is light, water, temperature, and soil. Two identical seeds will grow two slightly different tasting fruit if they have different access to those inputs. Most gardeners can attest to how many variables are needed just to grow a vegetable at all, much less get it to grow as well as it can.

On top of that, wine grape vines are often selected for the particular fruit they develop, but are often grown in regions they aren't suited to. So they need all kinds of complicated grafting techniques and treatments to survive the local climate, pests, bacteria and fungus.

Once you can actually get the grapes to grow, you have to harvest them when they are just the right ripeness for wine-making. Since a natural environment is variable, you never really know when this is, but good winemakers establish a good feel for when this is. Much like the above gardening processes it's still mostly unscientific, though it is informed by data and experience.

Then you get to process the grapes and ferment them. This unique grape that has gone through so much to be born is now subject to one of several processes to try to convert sugars into alcohols. The fermentation process used (typically yeast fermenting in barrels) depends on several dozen compounds and a hundred factors all being right for the yeast to thrive and convert the grape's sugars into alcohol and carbon dioxide.

Just imagine: the land, air, water, temperature, pests, gardener's care, grape varietal, fermentation... all of these involve complex processes, and a single variable changed can throw everything off. All of this adds up to one unique bottle of wine.

If a person claimed that they could detect all of this from a single sip of wine, no matter how standardized the process of tasting, they're lying. The only thing they can hope to do is guess where this wine was from, what type it was, how it was grown, how it was fermented, etc. They do this by studying the above entire chain of events in every wine growing region in the world, and based on that information, try to deduce the origin of the wine, and judge its quality.

Considering how many wines there are and how many variables go into each wine, it's kind of insane to think you could decipher all of this with a human palette. So wine tasting is a weird guessing game/hobby. A lot of people do make a living off of accurately determining all these things, but there's no way you could call this a science. It's an art.


> If a person claimed that they could detect all of this from a single sip of wine...

Nobody's talking about anything remotely that grandiose. If someone simply says they can tell a red from a white more often than not, that's still an objective claim that can be tested. Just because nobody can know everything every time doesn't mean it's all subjective.

More to the point, if someone says "you can taste differences, and often those differences correlate to a certain "kind" of wine", that's also a testable claim - i.e. something I'd believe if there's evidence for it. Hence I asked if you knew of any. (At this point I'll take the answer as read. ;)


> The question is whether you can taste wine differences without temperature differences

Temperature differences greatly affect the taste, so what's the purpose of that test?


Obviously, to determine whether there's an inherent taste difference between red and white wine when other variables are controlled for.


Which is meaningless, because nobody is trained to taste red wine when it's chilled to white wine temperatures. This is like saying that a classifier trained to distinguish between dogs and cats where the pictures are taken during the day should generalize to images of dogs and cats taken at night. If the classifier isn't able to distinguish between dog and cat images taken at night then it must be that there is no difference between dogs and cats.


The analogy is not a good one because the human sense/perception of taste is not a narrow classifier trained on limited data. I can taste something I've never tasted before and still a) describe the flavor, b) determine that it is different from other flavors I've tasted before, and c) recognize the new flavor again in the future.

If people with trained palates for wine cannot tell a chilled red apart from a white, or a room-temperature white apart from a red, it may not prove that the conventional wisdom around wine tasting is bullshit, but it sure suggests it strongly.


A chilled red wine tastes and smells completely different from a non-chilled red wine. Go ahead and try it yourself.

Also, wine tasting is absolutely a narrow classifier trained on limited data (in this case, tasters are not trained to classify red wine that is chilled). It's not surprising that wine tasters have overfit their classifier to the types and conditions of wine that they have been trained and tested on.


> A chilled red wine tastes and smells completely different from a non-chilled red wine. Go ahead and try it yourself

I believe this, but it's not what I said. I also assume a room temperature white smells and tastes completely different from a chilled one.

If there is nobody in the world who can, with significant accuracy, sort a set of reds and whites held at the same temperature into a set of reds and a set of whites, then it is reasonable to say that a lot of the conventional wisdom around wine tasting is probably wrong.

If wine tasters are too far up their own butts, use regular people instead, or people with more generally refined palates (e.g. chefs). Train somebody specifically for this task, even. At this point it's a hypothetical exercise.

Claiming e.g. "only trained wine tasters are qualified to make this distinction, and serving wines at 'incorrect' temperatures makes it impossible for trained wine tasters to tell the difference because they've 'overfit their classifiers'" is absolute hogwash.


Oh come on. I'm no wine enthusiast but the difference between a white and red wine is like apples to oranges.


I have done the same experiment. Only 2 out of 10 wine enthusiasts could with statistical certainty tell the difference between a rather big collection of white and red wines served at the same temperature.


And this is a meaningless experiment because all it demonstrates is that temperature (and the aromas and tastes produced at a given temperature) is an important feature for classifying wine, not that the wines are indistinguishable.

If these wine enthusiasts trained to drink all wine at the same temperature they would eventually learn to pick up the differences between red and white wines. But they did not train, so their classification error is high.


I dont know about that. Then they should have at least been able to correctly identify the wines served at optimal temperature (the red ones), but I can't say they did.

I didn't serve any spritzig white wines though. Then it would have been very obvious.


I'm not sure what you are saying. Your wine enthusiast friends were not able to identify the red wines served at the correct temperature in a blind setting?

That's not that surprising though! I enjoy drinking wine but I am really terrible at blind tasting. However, I have done blind tastings with people who _are_ trained sommeliers and it's very humbling to watch them taste and then methodically eliminate incorrect results until they eventually come up with the correct answer. Now, they are not always perfect, but for the most part they always at least get within the ballpark, whereas I am all over the map.

So, from my experience at least, I do believe that identifying wine in a blind tasting is a trainable skill, and it's not just hocus pocus.


Try it blindfolded some time, with both wines at the same temperature. Unless one of them is a super-sweet wine (moscato, gewurztraminer etc), you might be surprised at the result.


As my sibling said, try it out (with different whines)! I remember some red wines which tasted like whites and vice versa. Blind tastings teach humility.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: