The snobbery charge is fair for about a third of what gets said about coffee cups, and the honest thing to do is name which third before defending the rest. Claims that a particular clay imparts character to a glazed drink, that a hand-thrown cup performs differently from a machine-made one of the same weight, that a specific glaze colour transforms the coffee, and that a cup can add anything at all to what is in it, are either unverified or straightforwardly wrong. The rest of it is arithmetic and geometry, and it is easier to defend than most of what people spend money on.
A cup changes an espresso through temperature and shape, not through anything mystical, and the effect is real because the drink is tiny. Twenty five millilitres of liquid carries very little heat and very little aroma, so the vessel it lands in decides the temperature of your first sip, how much of the smell reaches your nose, how the liquid arrives at your tongue, and which drink you have actually made. None of it outranks the beans or the grind. All of it sits on top of them.
The reason this argument holds for espresso and collapses for a mug of filter coffee is worth getting straight first.
Why the drink's size decides how much the cup matters
Racineux and Tran put espresso at roughly ten times the concentration of filter coffee, and it arrives in a fraction of the volume. A double is around 50 ml. A single is around 25 ml, which weighs about 25 grams.
A thick stoneware demitasse can weigh 150 grams or more.
Pour a 90°C liquid into a room-temperature object with six times its mass and a broadly comparable capacity to soak up heat, and a large share of that drink's heat goes into the ceramic before it reaches your mouth. It does not cool gradually. It cools on contact, then keeps cooling from a lower start.
That is the whole reason nearly every espresso recipe in Anette Moldvaer's Coffee Obsession opens by warming the cup, and it is the strongest available evidence that vessel behaviour is a taste variable rather than a decorative one. A book that specifies a vessel for around sixty drinks and starts almost every one of them with the same instruction is telling you something. The practical version of that instruction, and the cheapest improvement available to anyone who owns an espresso machine, is covered separately in why four seconds of hot water is worth more than most upgrades.
Do the same arithmetic with 300 ml of filter coffee in a thin porcelain mug and the effect nearly vanishes. Lots of liquid, very little mass, no problem to solve.
Temperature decides what you are able to detect
The cup decides where on the temperature curve your first sip lands, and that curve moves the flavour.
The clearest illustration is how professional tasters work. The Specialty Coffee Association's cupping protocol has evaluation begin at about 71°C (160°F) and continue as the same cups fall to 38°C (100°F), stopping at 21°C (70°F). Jessica Easto, in How to Taste Coffee, reports Steen's advice to wait until around 54°C (130°F) before tasting seriously, on the grounds that very hot liquid dulls detection generally. Nobody would build a protocol around a descending temperature ladder if temperature were not doing something.
The underlying physiology is not fully settled, and the honest way to cite it is as a review rather than as a finding. Karel Talavera and colleagues surveyed the evidence on how temperature influences taste perception in Cellular and Molecular Life Sciences (2006), covering work on temperature-sensitive ion channels involved in taste transduction. A good deal of that evidence comes from rodent work, which is a real limit on how confidently anyone should apply it to your kitchen. What is not in dispute is the everyday observation that the same coffee reads differently at 70°C and at 45°C, which anyone can confirm in one sitting.
An unheated heavy cup takes that decision out of your hands. It drops the first sip several degrees below wherever you intended it to land, and it does so before you have tasted anything.
Shape moves the aroma around
Most of what you call flavour is not happening on your tongue.
Easto's account is the useful one here. Five signals reach you through the tongue: sweet, salty, sour, bitter and umami. Anything more specific than those five is smell, and smell arrives by two separate paths. One is the obvious path, sniffing the cup before you drink. The other runs backwards out of your mouth. Warmth releases aroma compounds from the liquid on your tongue, and your next breath out carries them into the nasal cavity from behind. That second path does most of the work, which is why holding your nose while you drink flattens coffee into hot bitter water in about two seconds.
Both paths depend on volatile compounds getting into the air. That makes the space above the liquid part of the apparatus.
An inward-tapering cup, narrower at the rim than at its widest point, keeps that vapour concentrated over the surface where your nose is going to meet it. A wide flaring cup vents it into the room, cools the drink faster through the larger surface, and gives you a quieter first sip. Neither is better in the abstract. A tapered cup suits a small drink you are meant to smell before you drink. A wide one suits something you will finish briskly and something you want to look at, which is why latte art cups flare and espresso cups do not.
Interior curvature does a second job that is specific to espresso. Moldvaer specifies a curved base for espresso cups rather than a hard internal corner, on the reasoning that the shot meets a continuous surface instead of an edge and the crema layer is disturbed less on the way in. That is a claim about how the crema arrives, not about how the drink tastes. We would rate it as plausible and not independently verified, and it matters less than most people think, since crema tells you far less about a shot than its reputation suggests.
The rim decides how the liquid arrives
The rim is the last surface the drink touches before your tongue, and it controls the delivery.
A thin rim releases the liquid as a narrow sheet onto a small area of the tongue, which tends to read as brighter and cleaner. A thick or rolled rim spreads it wider and slower, which tends to read as rounder and softer, and mutes perceived acidity a little.
This is a perceptual claim, not a chemical one. Nothing about the rim changes what is dissolved in the coffee. We have not run it past a panel, and it is one of the few claims in this category you can test yourself in ten minutes with two cups and one brew, which is worth more to you than our opinion.
Volume decides which drink you made
The least discussed and most concrete effect of all. For a fixed espresso dose, cup volume alone determines what you have built.
Moldvaer's glossary puts a demitasse at roughly 90 ml, and 60 to 90 ml is the range that makes an espresso, a ristretto or a macchiato. A cappuccino or flat white lives at 150 to 180 ml. If you own one 220 ml cup, you cannot make a flat white in it, because 50 ml of espresso plus enough milk to fill 220 ml is a latte by ratio no matter what you write on the menu. This is the mechanism behind drinks getting worse as the cups got bigger, and it is not a matter of taste. It is division.
Where the snobbery charge actually lands
Colour. The claim that a white cup makes coffee taste substantially more intense than a black one circulates constantly in specialty coffee, and we have not verified it. Our position on what a cup's colour can and cannot be said to do is that the perception research is plausible, second-hand to us, and not something we will repeat as fact while it happens to suit a shop selling white ceramic.
Material as flavour. A properly vitrified glaze contributes nothing to the drink. The exception is a porous or crazed body, which can hold residue and odour, and that is a hygiene fact rather than a flavour feature.
Provenance. A hand-thrown cup and a machine-made cup of the same mass, thickness and geometry behave identically in the hand and in the mouth. Buy the hand-thrown one because you want to look at it every morning for a decade. That is a real reason and it is not a performance reason.
What a good cup will not do
It will not improve your coffee. This is the framing error underneath most drinkware marketing. A cup has no way to add anything to what came out of the machine. The best it can do is take less away, and the difference between a good cup and a bad one is the size of the loss.
It will not compensate for anything upstream. Stale beans, an uneven grind, a channelled shot: none of these care what they are served in.
And a heavy cup will actively make things worse if you do not warm it, which means the thick stoneware demitasse that looks most serious on the shelf is the one that punishes carelessness hardest. That is a fact about the object we would rather you had before buying than after.
If you own one cup
Warm it. That single habit is worth more than replacing it.
If you are buying, the specification that gives the least away is a 60 to 90 ml cup with real wall thickness, a white interior so you can see what you made, and an inward taper at the rim. Warmed before it is filled, that object holds a small drink in a decent temperature range for the four or five minutes an espresso is meant to last, keeps the aroma over the liquid rather than over the counter, and shows you honestly what came out of the machine.
Everything else about it is a matter of what you want to look at, and there is nothing wrong with that as long as nobody tells you it is a flavour argument.