日本語版

Bitter crossing — 2026-09-06

Roast-bitter

こげ苦い (kogenigai) — our coined term

Roast-bitter is NIGAMI LAB's coined term for a taste state in which the fragrant aroma produced by roasting or charring is felt at the same time as bitterness, without cancelling it out.

What this page establishes

• The name "roast-bitter" is one NIGAMI LAB has given to a state in which roasted color or a charred aroma persists alongside bitterness at the same time; it has not been confirmed as a headword in any Japanese-language dictionary.
• Both a case in which lengthening the roasting time leaves the bitterness score unchanged, and a case in which raising the brewing temperature increases the bitterness score, have been reported from experiments that actually measured food and drink.
• A case in which Maillard reaction products lowered the bitterness score of a quinine solution has been confirmed by electronic-tongue measurement.
• Cases in which a darkly charred layer of rice is described not as bitter but as fragrant or delicious can be confirmed in sources on paella and guoba.

How it arises

Roasting and charring are processes that remake an ingredient with heat, and the compounds that result do not head toward bitterness alone. No claim directly demonstrating this combination at the receptor (cell) layer exists so far. We lay it out across two layers: solution and cooking.

In solution

The Maillard reaction, one of the chemical foundations of roasting and charring, is not a reaction that only creates bitterness. In an experiment that added a glycated beef protein hydrolysate to a quinine solution, the bitterness score measured by an electronic tongue dropped significantly, by up to 38%. It is a case in which a charring product worked to cancel out bitterness that was already present.

One review states that which way it turns depends on the heating temperature range. It describes umami-type free amino acids as more likely to form at 80-100°C when cysteine is present, and bitter-type free amino acids as more likely to form at 100-120°C — showing that even the same Maillard reaction can head toward bitterness or toward another taste depending on temperature.

In cooking

Experiments that actually roasted or brewed food and drink and rated bitterness turn up both cases where strengthening the roast leaves bitterness unmoved, and cases where a different variable moves bitterness instead. In a study that roasted Tieguanyin oolong tea at 105°C for 1, 3, or 5 hours, with an additional roast of 130°C for 1 hour applied to the 5-hour batch, the aroma compounds tied to roasted fragrance and the color of the tea leaves changed clearly, while the bitterness score given by a trained panel stayed stable even as roasting time was extended.

In coffee, by contrast, it was brewing temperature rather than roasting itself that moved bitterness. In a sensory experiment that brewed dark-roast beans by full immersion under three conditions — 4°C, 22°C, and 92°C — the bitterness score rose significantly as the water temperature increased, and sourness also rose at the same 92°C condition. Bitterness and sourness did not cancel each other out; they moved in the same direction.

For okoge itself, claims from academic sensory panels are still thin. One blog explaining word meanings attributes the fragrant aroma of okoge to the Maillard reaction and the caramelization reaction, while stating that "if it is over-charred, bitterness appears and becomes unpleasant," whereas "a moderate char is a treasure trove of umami and aroma." It is a view in which the degree of charring decides whether a dish stays on the fragrant side or steps over into bitterness.

Exceptions and reservations

Naming this roast-bitter does not mean that roasting or charring always brings bitterness along with it. An experiment in which a trained panel used a 10 cm scale to evaluate the process by which socarrat forms in paella reports that as the charred layer in the pan advances, the intensity of "roasted" rises while "burnt" is kept down — but bitterness itself is not included in this evaluation scale. Even within the same experiment, roasted aroma and charred aroma standing out, and bitterness standing out, are measured separately.

The word "bitter" itself also wavers between taste and aroma. In a study evaluating roasted hojicha samples, the aroma description for the medium roast (150°C, 12 minutes) used the term "Bitter aroma," while the light roast (130°C, 10 minutes) was described only as a pleasant roasted aroma. Here, "bitter" is a description of an aroma the evaluators smelled, not a measurement of bitterness as a taste taken into the mouth. The number of evaluators and the scale are not stated in the text either, so it would be premature to read the aroma description directly as a claim about bitterness.

Descriptions on the guoba and socarrat side, if anything, point to a boundary of "charred but not bitter." An article by an independent Chinese food writer (a statement with no academic verification) describes the char on guoba as "a golden, fragrant char — aroma, not bitterness," distinguishing it from the failed form charred to black carbon.

The normative dictionary of the Valencian language likewise defines the rice of socarrat as "partially charred but at times very tasty," evaluating it as tasty rather than bitter. Charred color and bitterness are treated as separate axes even within dictionary and food-culture descriptions.

The practice of adding salt or sugar partway through roasting to round off the bitter aftertaste is also recorded in the roasting histories of various regions. But salt and sugar are buffers that cancel bitterness without leaving their own flavor behind; the counterpart there is not fragrance itself. That process is addressed in a separate research note.

Examples from around the world

• Okoge — the charred layer of rice that forms at the bottom of the pot when cooking rice over a fire
• Nurungji — the charred layer of rice at the bottom of the pot in Korea, eaten as a snack or steeped in hot water as sungnyung
• Guoba (rice crust) — the charred layer of rice that forms at the bottom of the pot in China; also eaten on its own
• Socarrat — the layer that forms late in cooking Valencian-style paella, where the rice at the bottom of the pan sticks fast to it
• Pasta all'assassina — a dish in which raw pasta is seared directly against the bottom of the pan, deliberately charring it
• Basque burnt cheesecake — a cheesecake finished by charring its surface black at high heat for a short time
• hojicha roasting — the green-tea roasting process, divided into light, medium, and dark roasts by exhaust temperature and time
• coffee brew — a drink whose bitterness is reshuffled by roasting products and brewing conditions

Try it at home

Narrow it down to one variable. Compare the okoge of rice cooked from the same rice, changing only the heat.

1. Cook the same amount of rice in the same pot, and after turning off the heat, vary only the heat used for the final simmer across three conditions: low, medium, and high.
2. Keep the amount of rice, the water ratio, and the length of the final simmer the same across all conditions.
3. Record the color of the resulting okoge, from a light charred brown to a color close to black, and rate the intensity of fragrance and the intensity of bitterness in separate columns.
4. Line up the three conditions and compare where the boundary falls between the range in which fragrance increases and the range in which bitterness clearly emerges.

A second track: using the same dark-roast beans and the same grind, vary only the temperature of the brewing water across three conditions — cold, room temperature, and hot. Keep the ratio of grounds to water and the brewing time the same, and rate bitterness and sourness separately.

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

What is not known

• The experiment in which Maillard reaction products lowered the bitterness of quinine is limited to electronic-tongue measurement; a study confirming the same relationship with a human sensory panel has not yet been measured.
• Within the range of sources gathered for this page, we have not been able to locate the experimental conditions of the primary research behind the statement about which free amino acids — bitter-type or umami-type — form more readily at which heating temperature.
• For dishes such as okoge and guoba, described as looking charred yet not easily tasting bitter, we have not been able to locate, within the range of sources gathered for this page, academic sensory material that rated bitterness numerically.

Between the dictionaries

Whether the same-shaped word 焦げ苦い (kogenigai) stands as a headword in a Japanese dictionary is left unconfirmed.

The difference from horonigai (gently bitter) (ほろ苦い) can be explained as a difference of axis. Horonigai is a word for a state along a single axis of bitterness intensity, in which bitterness itself has been weakened; it applies when a counterpart taste or process mellows only the bitterness without leaving its own flavor behind. Roast-bitter is not that. It points to a state in which a separate sensation — the fragrant aroma produced by roasting or charring — does not weaken bitterness but settles in right next to it and remains without disappearing. Weakening versus standing side by side — the axis is different.

Frequently asked

Does roasting more deeply always make something more bitter?

Not universally. For coffee, raising the brewing water temperature increased the bitterness score, but for Tieguanyin oolong tea, the bitterness score stayed stable even as roasting time was extended. The compounds tied to roasted aroma and the color changed in both examples, so the compounds that move and the bitterness that moves are not necessarily the same.

Are okoge and guoba charred yet not bitter?

Many descriptions say it depends on degree. Descriptions of guoba and socarrat evaluate a golden, fragrantly charred state as aroma or deliciousness rather than bitterness, and separate it from the failed form charred to black carbon. For okoge too, a view has been stated in which a moderate char stays on the fragrant side, while over-charring brings out bitterness. Within the range of NIGAMI LAB's claim data, we have not located an academic sensory experiment that measured exactly where that boundary falls.

Is the Maillard reaction only a reaction that strengthens bitterness?

No. In an experiment that added Maillard reaction products to a quinine solution, the bitterness score actually fell. There is also a statement that whether bitter-type or umami-type free amino acids form more readily depends on the heating temperature range, so the same reaction works both to create bitterness and to hold it back.

Related: Does any seasoning remove bitterness? — Sugar and salt /Roasting Makes Bitterness, and Dark Roast Changes It /Butter Roasting and Southeast Asian Kopi — Roasting That Wraps Up Bitterness /Can Cooking Create Bitterness? /Bitter crossings