日本語版

Bitter crossing — 2026-09-06

Astringent-bitter

しぶ苦い (shibunigai) — our coined term

Astringent-bitter refers to a taste state in which the puckering sensation of astringency and bitterness both remain at once, without one erasing the other. It is the lab's coined term.

What this page establishes

• The name "astringent-bitter" is a term the lab uses for the state in which astringency and bitterness remain together; it has not been found as a headword in dictionaries.
• Multiple reviews have confirmed that astringency is not a basic taste but is instead classified as an oral tactile sensation that includes dryness, roughness, and a puckering feeling in the mouth.
• In experiments using phenolic substances added to white wine, it has been confirmed that which of astringency or bitterness dominates reverses depending on how far polymerization has progressed.
• It has been reported, in a comparative experiment on Riesling and Gewürztraminer, that the judgment of which of astringency or bitterness dominates flips simply by changing the measurement method.

How it arises

Astringency and bitterness are not the same sensation. Astringency is a tactile sensation that includes dryness, roughness, and a puckering feeling in the mouth, while bitterness is a taste mediated through the mechanism that receives bitterness. Even so, the two often arise together in foods and drinks, and how strongly one is felt can affect how the other is felt. No claim in the lab's claim data yet directly shows a mechanism, at the level of receptors (cells), by which the two move each other.

In solution

Experiments using tannin-like compounds dissolved in water show that astringency and bitterness, though arising from the same group of substances, shift their balance depending on how far polymerization has progressed. In synthetic flavan-3-ol solutions made from catechin and epicatechin, as the degree of polymerization rose from monomer to dimer to trimer, the maximum intensity and duration of bitterness fell, while the maximum intensity of astringency rose.

An experiment using a model wine solution containing oligomeric tannins derived from grape seed also shows that the two sensations do not respond identically to the same conditions. Moving the pH changed only astringency, while moving the ethanol concentration moved both astringency and bitterness — an asymmetric result showing that astringency and bitterness cannot be explained by a single mechanism alone.

In practice

A review summarizing how foods and drinks are perceived when taken into the mouth states that astringency and bitterness are felt together across multiple studies, and that because how one is perceived affects the other, there are cases where assessors confuse the two.

In a sensory experiment that added several phenolic substances to actual white wine, the polymers tannic acid and grape-seed tannin scored higher on astringency than on bitterness, while the monomers gallic acid and catechin scored higher on bitterness than on astringency — the reverse pairing. The effect of the degree of polymerization seen in the solution experiments appears in the same direction in the practical setting of actual wine.

Exceptions and reservations

Astringent-bitter is not simply a matter of "what is astringent is bitter." An experiment adding tannic acid to a quinine solution shows the reservation at the center of this crossing. Raising tannic acid concentration from 0.005 to 0.05 (w/v)% suppressed the bitterness score of 0.1 mM quinine to about 61% of the control. But raising the concentration further to 0.15 (w/v)% produced no further suppression — instead, the resulting bitterness score was higher than the control: a relationship whose sign reverses. The authors themselves attribute this reversal not to a change in receptor binding but to the astringency of the high-concentration tannic acid solution itself, and note that bitterness and astringency were rated on the same scale in this experiment. However, the authors themselves also state explicitly that this attribution to astringency is unproven, and the paper does not present measured astringency intensity values.

The response when another taste is layered onto an astringent compound is also not uniform. In an electronic-tongue measurement in which umami amino acids were made to co-occur with an ester-type catechin, bitterness was suppressed cleanly as concentration rose, while the effect on astringency is reported only as "complicated," without a single direction. Conditions that move bitterness do not necessarily move astringency by the same amount.

Even the conclusion of which sensation dominates can flip depending on how it is measured. In an experiment comparing skin-contact Riesling and Gewürztraminer using three methods — descriptive analysis, time-intensity, and temporal dominance of sensations — the same sample that descriptive analysis and time-intensity judged to be the most bitter wine was judged, under temporal dominance of sensations, to be dominated instead by the astringent mouthfeel. It is worth remembering that the very naming of astringent-bitter is a description made after a measurement method has already been chosen.

Examples from around the world

The tannins that produce astringency vary how far bitterness overlaps with that astringency depending on the degree of polymerization and the extraction conditions. Strong infusions ofgreen tea infusion andKudingcha, andred wine, are often cited as leading examples where bitterness is perceived layered onto astringency.

Persimmon (kaki) and itspersimmon tannin (kaki-tannin), on the other hand, are placed here as an example that marks the boundary of astringent-bitter. Persimmon's astringency is a strong tannin-derived puckering sensation, but the lab records persimmon's astringency as distinct from bitterness and does not count persimmon as an example of astringent-bitter. The boundary set down here is that being astringent and being bitter do not always hold true at the same time.

Try it at home

Narrow it to a single variable. Steep the same tea leaf (green tea) while changing only the water temperature across three steps: lukewarm, medium, and hot.

1. Put the same amount of the same tea leaf into three teapots, and steep for the same time while changing only the water temperature across 3 conditions: lukewarm, medium, and hot.
2. Keep the amount of tea leaf, the amount of water, and the steeping time the same across all conditions.
3. Take the infusion into your mouth and record the intensity of bitterness and the intensity of astringency (the puckering feeling in the mouth) in separate columns.
4. Line up the three conditions and compare whether bitterness and astringency moved in the same direction, or whether only one of them moved.

Status: candidate (to be published once a tasting-comparison record is attached).

What is not known

• Whether the dryness, roughness, and puckering feeling underlying astringency are mutually independent sensations, or only appear to be one sensation split apart, has not yet been measured.
• The solution experiments showing the relationship between degree of polymerization and astringency/bitterness used model aqueous solutions, and whether the same reversal occurs in actual wine has not been directly confirmed.
• Within the sources gathered for this page, no comparative source has been found that confirms the astringent-bitter state in beverages or foods other than green tea, Kuding tea, red wine, and persimmon.

Between the dictionaries

Whether the terms しぶ苦い and 渋苦い — the same word written two ways — stand as headwords in Japanese dictionaries is left unconfirmed. What could be confirmed is not a headword but a juxtaposition in a thesaurus column: the Digital Daijisen's thesaurus entry for渋い lists 苦い and ほろ苦い as synonyms.

This juxtaposition shows that astringency and bitterness have also been placed close together in language, but it is not evidence that the dictionary fuses the two into a single word. The real contrast ishoronigai (gently bitter) (mildly bitter), whose dictionary definition sticks entirely to language that softens bitterness itself — "a little bitter," "somehow bitter" — and makes no mention of astringency.

Mildly bitter is a word for a state that weakens intensity along the single axis of bitterness, while astringent-bitter is a word for a state in which a separate sensation, astringency, sits beside bitterness and does not disappear. Weakening versus standing side by side — the axis is different.

Frequently asked

Why do both astringency and bitterness intensify when tea is brewed strong?

Because both astringency and bitterness can arise from tannin-derived compound groups, both ratings tend to move as extraction progresses. However, how much and in which direction they move depends on the conditions, and the pH and ethanol results from the solution experiments show that sometimes only one of the two moves.

Does stronger astringency in wine mean stronger bitterness?

It cannot be said outright. In an experiment adding substances to actual white wine, the opposite pairing was observed: the more polymerized the tannin, the more astringency exceeded bitterness, while for less polymerized phenolic substances, bitterness exceeded astringency. Which one dominates can even flip depending only on which measurement method is used.

Can persimmon be called astringent-bitter?

No — that is the lab's position. Persimmon's astringency is strong, but no academic source has been found that describes its astringency as a form of bitterness, and a separate research note,Persimmon Is Astringent, Not Bitter, treats astringency and bitterness separately.

Related: Does any seasoning remove bitterness? — Sugar and salt /Persimmon Is Astringent, Not Bitter /Huigan — Tea's Time of Sweetness After Bitterness /Bitterness, Astringency, and Egumi /Bitter crossings