Japanese Sentence Intonation: Falls, Rises, ね, よ, よね

Japanese sentence intonation is the melodic shape of a whole utterance. It includes the gentle fall of a statement, the lift on か, the rise that turns a casual plain-form verb into a question, and the small tail tunes that ride on the sentence-final particles ね, よ, and よね.¹² These contours are independent of word-level pitch accent. In everyday Japanese, they do much of the speech-act work: the same written string can be a statement, a question, or an acknowledgement depending only on how the right edge moves.³⁴

Overview

What sentence intonation is, and what it is not

Sentence intonation is a property of phrases and utterances, not of words. Vance puts the distinction at its sharpest: when a pitch pattern belongs to a phrase, it is called intonation; when it belongs to a word, it is called pitch accent.¹

In the Pierrehumbert and Beckman model that underlies the X-JToBI transcription system, Tokyo Japanese prosody has two independent layers. Lexical pitch accent is a per-word H*+L drop. Phrase- and utterance-level intonation is encoded as boundary tones (H%, L%, LH%, HL%) at the right edge of accentual and intonational phrases.⁵²

The two layers do not overwrite each other

A boundary rise at the end of a question does not erase the word-internal HL drops that came before it. The lexical accent pattern of each word stays intact through the body of the sentence; the boundary tone only shapes the last mora or two at the right edge.¹³

The X-JToBI inventory used in the Corpus of Spontaneous Japanese labels right-edge tunes with a small set of symbols: L% for the unmarked low boundary of a declarative, LH% for the question rise, HL% for an explanatory rise-fall typical of casual speech, and H% for a high tone that adds prominence.⁶²

田中たなかさんは学生がくせいです。¹
"Tanaka is a student."

田中たなかさんは学生がくせいですか。⁷
"Is Tanaka a student?"

今日きょうは寒さむいです。¹
"It's cold today."

Why this matters for the JLPT N4+ learner

JLPT N5 listening tracks use studio-clean examples, and their intonation is often flatter than spontaneous speech. The Corpus of Spontaneous Japanese contains rise-fall and HL% boundary tones that textbook audio often does not represent.⁶

In practice, the same string can be a statement, a question, or an acknowledgement depending only on the right-edge tune. そうですか with a rise is a real question; the same string with a fall is "I see."⁴ Intonation, not the individual sounds, selects the speech act, so a learner who hears only the segments will miss the move.³

そうですか。⁴
"Is that so?" / "Really?" (rising か)

そうですか。⁴
"I see." / "Oh, okay." (falling か)

How Japanese intonation relates to pitch accent

Pitch accent is per-word, intonation is per-utterance

Pitch accent in Tokyo Japanese is lexical: a word is specified for at most one HL drop, and the drop's location distinguishes lexical items (the canonical 端 / 箸 / 橋 contrast).⁸ Tokyo Japanese organises word-level patterns into four classes: heiban (no drop), atamadaka (drop after mora 1), nakadaka (drop inside the word), and odaka (drop after the final mora, audible only on the following particle).⁸⁹

Intonation, by contrast, is phrasal. Boundary tones operate at the right edge of the accentual phrase and the intonational phrase, where they affect the last mora or two of a phrase and leave the lexical drops upstream untouched.⁵²

雨あめが降ふる。⁸
"It rains."

飴あめが好すきです。⁸
"I like candy."

Intonation rides on top: the declination contour

Tokyo Japanese shows a gradual downward drift in F0 (fundamental frequency, heard as pitch) over the course of an utterance. In the Pierrehumbert and Beckman framework, the intonation phrase is the domain in which downstep cumulatively lowers each subsequent accentual phrase. A separate final-lowering effect operates in roughly the last 250 ms of the utterance.⁵

After a lexically accented phrase, each following accentual phrase has a lower pitch range than the one before it. The staircase resets only at intonational-phrase boundaries.²⁵ The utterance as a whole is signalled at both edges: an enlarged initial rise on the left edge and a phrasal H tone or final lowering on the right edge.⁵

The shape sketched here is the unmarked declarative: the pitch peaks of successive accentual phrases sit progressively lower, then the final mora drops to the L% boundary.

田中たなかさんは今日きょう学校がっこうに行いきました。¹
"Tanaka went to school today."

What a final rise or fall changes, and what it leaves alone

The boundary tone is local. In X-JToBI the boundary pitch movement is labelled on the final mora or final mora-pair of the accentual phrase; the lexical HL drops upstream survive unchanged.⁶²

The TUFS Japanese Pronunciation module states the same point for learners: Japanese expresses a question "while maintaining the accent patterns of the words in the sentence, by sharply raising only the final mora at the end of the sentence."¹⁰

The default declarative tune: a gentle fall

The shape: gradual decline, final-mora drop

The unmarked declarative carries the declination contour described above: an initial accentual-phrase rise, cumulative downstep across later accentual phrases, and final lowering at the right edge.⁵² In X-JToBI, the default declarative boundary is labelled L%. It marks a low right edge with no rising pitch movement, which is the gentle fall learners hear.²⁶

田中たなかさんは学生がくせいです。¹
"Tanaka is a student."

今日きょうは天気てんきがいいです。¹
"The weather is nice today."

Polite-form sentences: です/ます endings carry the fall

In a polite-form sentence, the right-edge L% boundary lands on the です or ます ending. The same string with an LH% boundary on the last mora becomes a yes/no question, treated in the next section.²⁷

Ishihara's experimental data on Tokyo Japanese show that yes/no questions exhibit an F0 prominence on the verb together with a sentence-final rise; the corresponding declarative lacks both cues.⁷

学生がくせいです。⁷
"I'm a student."

学生がくせいですか。⁷
"Are you a student?"

Plain-form declaratives: same shape, faster

The declination contour is independent of style. A plain-form declarative (だ, the bare ru-form of a verb, or an i-adjective in its dictionary form) carries the same L% boundary on its final mora and the same cumulative downstep over the utterance.⁵²

The plain form is shorter than its polite counterpart, so the same fall is compressed into less time and the drop on the final mora feels sharper to the ear.

学生がくせいだ。¹
"I'm a student." (plain)

雨あめが降ふる。⁸
"It rains."

Question intonation 1: the か-final rise

Rising か: the true yes/no question

The X-JToBI label for the rising-question boundary is LH%, which transcribes a scooped, concave rise across the final mora of the accentual phrase.²⁶ Ishihara's production study of Tokyo Japanese shows that a yes/no question ends with this sentence-final rise and also has focal F0 prominence on the verb. The matching declarative carries neither cue.⁷

か without the rise is not yet a question

TUFS makes the pedagogical point directly: even with か present, "if there is no sharp rise of pitch on the か, the utterance does not become an interrogative."¹⁰ The か particle marks question-form morphologically; the rise on か is what selects the question speech act.

学生がくせいですか。⁷
"Are you a student?"

田中たなかさんですか。¹⁰
"Are you Tanaka-san?"

もう行いきましたか。¹
"Have you already gone?"

Falling か: not a question, an acknowledgement

The same string can carry a falling か (L% boundary). When it does, it no longer asks a question and instead acknowledges newly received information.¹¹ The pragmatic reading is "I see" or "Oh, okay." Nagano-Madsen treats the so-called か-final question with a fall as a typological baseline rather than as an exception.¹¹

Pragmatically, the falling そうですか is the conventionalised "I see" use, distinct from the rising そうですか that is a real question.⁴

そうですか。⁴
"Is that so?" / "Really?" (LH% on か)

そうですか。⁴
"I see." / "Oh, okay." (L% on か)

Reading a falling か as a question

A learner who hears そうですか with a low, falling tail and reads it as a yes/no question will reply as though they were asked something. The Japanese speaker has just acknowledged what was said; the next turn is theirs, not the learner's.⁴²

Polite-context note

Boundary pitch movement in Japanese is gradient, and Igarashi notes that it can be suppressed without losing the rest of the prosodic structure.³ In formal speech, a か-question is often produced with a reduced or absent final rise. In that case, か itself does the interrogative work.

Ishihara reports that the sentence-final rise is reliable in conversational yes/no questions, but it is one cue among several. Another cue is focal prominence on the verb. This is consistent with a low-rise polite-context reading.⁷

何時なんじにお越こしになりますか。¹
"At what time will you arrive?"

Question intonation 2: the rise without か

The mechanism: drop か, raise the final mora

Japanese permits a rising declarative: a sentence that is grammatically declarative (no か, no overt question marking) but is realised with a rising boundary tone and interpreted in context as a polar or biased question.¹²

The rise targets the final mora. Ueki's discourse-corpus study labels the same boundary tone HRT, the high-rising terminal. Ueki notes that it is "an upward intonation like that which occurs at the end of an interrogative phrase, yet an utterance with HRT is not an interrogative," so the syntactic form remains declarative while the prosody opens a response slot.¹³ Hirayama's account analyses the rise as marking that the speaker is not making a direct commitment to the propositional content. The rising intonation alone invites the hearer to update the common ground rather than asserting outright.¹²

行いく?¹²
"(You) going?"

もう食たべた?¹²
"Already ate?"

学生がくせい?¹²
"(You're) a student?"

Register: friends, family, low-stakes work talk only

The rise-without-か pattern is documented as a feature of casual discourse. Ueki's corpus draws its examples from informal conversation and explicitly pushes back on the stereotype that the rise is gendered. Ueki finds that "some people use HRT regardless of age."¹³

Hirayama notes that the rise without か is structurally available across declarative, imperative, and interrogative clause types. Its non-commitment effect, however, makes it pragmatically marked in formal registers. In a polite or formal context, the same content takes か.¹²

Rising declarative in polite speech sounds wrong

Used toward a stranger, a customer, or a superior, the bare-rise question reads as flippant or incredulous rather than genuinely inquisitive. In those settings, the content needs か (with an LH% rise) or か (with a low, reduced final), not the か-less rise.¹²³

今日きょう来くる?¹³
"Coming today?"

寒さむい?¹³
"(Are you) cold?"

What this article does and does not cover

This article catalogues the rise-without-か pattern as one tune inside the wider sentence-intonation system. It does not unpack every pragmatic detail: when a rising declarative biases toward an expected answer, when it presents new information that the speaker is reluctant to commit to, or how it interacts with the ね and よね tunes. Those questions deserve a dedicated treatment of their own.

Sentence-final particle tunes

ね with a downward-then-up tail

The ね particle takes either a falling or a final-rising contour, and the two are pragmatically distinct. Falling ね marks shared-feeling synchronisation ("we agree on this"); rising ね is a confirmation-seeking move ("you agree, don't you?").³¹²

A common conversational realisation pairs a fall on the predicate with a small rise on ね itself, the downward-then-up tail. Igarashi groups this contour with the LH% family of boundary pitch movements.³

いい天気てんきですね。¹
"Nice weather, isn't it?"

寒さむいですね。¹
"It's cold, isn't it?"

そうですね。⁴
"That's right." / "Indeed."

よ flat or falling

In its informing use, the よ particle typically carries a flat or falling boundary tone, with an L% boundary on よ itself. This is the prosodic correlate of the asserting or informing function that distinguishes よ from ね.¹²³

Hirayama's analysis separates yo-falling (the speaker presents new content the hearer should now have) from yo-rising (the speaker presents content without committing to it directly). The two contours convey distinct discourse moves with the same particle.¹²

これですよ。¹²
"It's this one."

違ちがいますよ。¹²
"That's not right."

よね: two particles, one fused tune

The よね particle packages the assertive force of よ with the confirmation-seeking move of ね. Its typical realisation carries the ね-style rise on the final particle. The speaker asserts and invites agreement at once.³

Ueki's discourse-corpus data document よね with a high-rising terminal as a confirmation-seeking move whose function is closer to "we both already know this, right?" than to a true information question.¹³

あの店みせ、美味おいしいですよね。¹³
"That restaurant is good, isn't it?"

田中たなかさん、明日あした来きますよね。¹³
"Tanaka's coming tomorrow, right?"

な and the masculine/internal-thought register

な takes a falling tail and is used predominantly in soliloquy or internal-thought register. Ueki's spontaneous-discourse data find it patterning with monologic, low-volume self-address.¹³ It is flagged here rather than drilled, since pointing it at an interlocutor in polite settings sounds confrontational or unguardedly internal.

いい天気てんきだな。¹³
"Nice weather, huh." (to oneself)

The contrastive は: a falling tune in mid-sentence

Plain は vs. contrastive は

は has two uses that can be distinguished by prosody. Topic-marking は is melodically unmarked; contrastive は carries a small pitch peak on the marked constituent and a radical lowering of the pitch on the phrases that follow it.¹⁴

Heycock's chapter restates the long-standing observation (attributed to Kuno's work on -wa / -ga) and supports it with experimental and corpus evidence. Heycock characterises contrastive topics as "the presence of a prominent high-pitch accent on some part of themselves and a radical lowering of the pitch accent of the phrases following them; non-contrastive topics are conversely characterized by the lack of both features."¹⁴

私わたしは学生がくせいです。¹⁴
"I'm a student."

Worked pair

The same written string supports two contours. Read with the topical は, the line is melodically flat across 私は. Read with the contrastive は, a pitch peak lands on 私は and the pitch lowers on everything that follows.¹⁴

私わたしは学生がくせいです。¹⁴
"I'm a student." (topic は, flat across 私は)

私わたしは学生がくせいです。¹⁴
"As for me, I'm a student (whatever the others are)." (contrastive は, peak on 私は, lowering on 学生です)

Why textbooks rarely teach this

The contrast lives entirely in the prosody. Print text cannot represent it without explicit intonation marks, which Japanese reference grammars typically reserve for phonetics chapters.¹⁴¹

Heycock observes that the topical / contrastive split has been a methodological problem for that reason: the same surface string supports both readings, and only prosody, together with context, consistently disambiguates them.¹⁴

Contrast with English yes/no question intonation

English rises gradually; Japanese lifts only the final mora

TUFS frames the Tokyo Japanese question rise as targeting only the final mora, while the rest of the sentence preserves its lexical pitch accents: "while maintaining the accent patterns of the words in the sentence, by sharply raising only the final mora at the end of the sentence."¹⁰ The X-JToBI LH% label is a boundary-local tone applied at the right edge of the accentual phrase; it is not a global ramp across the utterance.²⁶

In English, by contrast, the canonical yes/no-question intonation is a gradual H% rise spanning the nuclear pitch accent through the right edge of the utterance.⁵ Pierrehumbert and Beckman explicitly contrast the Japanese system with the English system they had previously analysed.

The English line rises across the whole utterance; the Japanese line keeps its accent pattern flat through the body and lifts only the last mora.

Why English-accented Japanese questions sound wrong

Applying an English-style gradual ramp to a Japanese question flattens the internal lexical HL drops, because the rise spreads across the word-level pitch contour. This is the prosodic origin of the perception that L1-English speakers sound uncertain rather than inquisitive when they ask a question in Japanese.⁵²

Igarashi notes that boundary pitch movements in Japanese are tonal events at a phonologically defined right edge, not gradient F0 raises. Misplacing the rise outside this edge produces a non-native percept.³

The fix: hold the declination flat, lift only the last beat

The classroom rule is straightforward: preserve the lexical accent patterns through the body of the utterance, retain the declination contour, and place the rise only on the final mora.¹⁰²

In practice, the rise lands on か, on the verb's final syllable when か has been dropped, or on whichever segment carries the boundary. Everything earlier in the utterance stays at its lexical pattern.

Good to know

The two-か trap: rising vs. falling か flip the speech act

The か particle marks question form morphologically, but the boundary tone selects the speech act. そうですか with a rising final mora is "Really?" or "Is that so?"; the same string with a falling final mora is "I see" or "Oh, okay." X-JToBI labels the two contrastive boundaries LH% and L%, respectively.²⁴

A learner who hears そうですか with a low tail and treats it as a question will reply as though they were asked something. The Japanese speaker has just acknowledged what was said, so the conversation needs to move on, not circle back.

そうですか。⁴
"I see." / "Oh, okay." (falling final, not a question)

English-style gradual rise on a Japanese question

A common N4-level mistake is to ramp pitch upward across 田中さんは学生ですか and end high. The Japanese realisation keeps the lexical accent pattern of each word and raises only the final mora (か).

Japanese boundary tones are local to the right edge, not ramps that span the utterance; the rise on か alone is the cue.¹⁰² The correct form keeps the body of the sentence at its declination contour and places the lift on the final mora only.

田中たなかさんは学生がくせいですか。⁷
"Is Tanaka a student?"

Rising declarative (rise without か) in formal speech

The rise-without-か pattern is restricted to casual discourse. In polite or formal contexts, the same content is realised with か and an LH% rise, or with か and a reduced final in very formal speech.¹²³

The rise alone reads as non-committal: the speaker is opening a response slot rather than committing to a proposition. That effect makes the bare rise inappropriate in registers where the speaker is expected to take ownership of the question.

な outside soliloquy / internal-thought register

な-final utterances pattern with monologic, self-addressed speech in Ueki's corpus.¹³ Pointing な at an interlocutor in a polite setting reads either as confrontational or as too internal for the situation. In polite speech, ね is safer.

Reading ね as English "right?"

The English-translation gloss "ね = right?" is convenient but coarse. ね with a small rise seeks confirmation of shared knowledge ("we agree on this, don't we?"). ね with a fall marks agreement that the speaker is registering for themselves, not interrogation.³¹²

The tune is what selects the move, so two utterances that look identical in print can be the speaker checking in (rise) or quietly synchronising with the listener (fall).

Four tunes, four labels

The system this article catalogues reduces to four right-edge tunes. Memorising the four shapes covers the high-frequency conversational space.²³

Tune	Boundary tone	What it does
Declarative fall	L%	Gentle final-mora drop on です/ます or plain form
Question rise	LH%	Sharp final-mora rise on か or on the last segment
ね tune	LH%	Small dip on the predicate, rise on ね / よね
よ tune	L%	Flat or falling on よ

Why the rise-without-か pattern exists at all

Japanese marks questions primarily with a sentence-final particle (か) plus a boundary tone. When the particle is dropped, the boundary tone alone can still carry the interrogative force, because the two layers are independent in the X-JToBI grammar.²¹²

That independence is what makes the bare rise viable in casual speech and what makes か with a falling tone a legitimate acknowledgement rather than a defective question.

See also

• Stress vs. Pitch: Does Japanese Have Stress?
• Japanese Pitch-Accent Notation: How to Read 0, 1, 2, 3 and the Overline Diagrams
• Pitch Accent for Japanese Verbs and Adjectives: The Binary Class Rule and Conjugation Shifts
• Sentence-Final Particles in Japanese (終助詞): Overview
• Japanese Filler Words and Hesitation Prosody: あの, えーと, まあ, and the Long-Vowel Stall
• Japanese Speech Rate: How Fast Do Native Speakers Actually Talk?

References

Footnotes

1. Vance, Timothy J. The Sounds of Japanese. Cambridge University Press, 2008. https://www.cambridge.org/9780521617543 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴

2. Venditti, Jennifer J. "The JToBI Model of Japanese Intonation." In Sun-Ah Jun (ed.), _Prosodic Typology: The Phonology of Intonation and Phrasing, Oxford University Press, 2005, ch. 7. Pre-print online: http://www.cs.columbia.edu/~jjv/pubs/jtobi-webversion.doc ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹

3. Igarashi, Yosuke. "Intonation." In Haruo Kubozono (ed.), Handbook of Japanese Phonetics and Phonology (Handbooks of Japanese Language and Linguistics 2), De Gruyter Mouton, 2015. https://www.degruyter.com/document/doi/10.1515/9781614511984/html ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³

4. Coto Academy. "How Do You Use the Japanese Phrase そうです (Sou Desu)?" Learner-publisher reference page. https://cotoacademy.com/how-do-you-use-the-japanese-phrase-%E3%81%9D%E3%81%86%E3%81%A7%E3%81%99sou-desu/ (limitation) ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹

5. Pierrehumbert, Janet B., and Mary E. Beckman. Japanese Tone Structure (Linguistic Inquiry Monograph 15). MIT Press, 1988. https://mitpress.mit.edu/9780262660631/japanese-tone-structure/ ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

6. Maekawa, Kikuo, Hideaki Kikuchi, Yosuke Igarashi, and Jennifer Venditti. "X-JToBI: An Extended J-ToBI for Spontaneous Speech." In Proceedings of the 7th International Conference on Spoken Language Processing (ICSLP / Interspeech), Denver, 2002. https://www.isca-archive.org/icslp_2002/ ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

7. Ishihara, Shinichiro. "The Intonation of Wh- and Yes/No-Questions in Tokyo Japanese." In Chungmin Lee, Ferenc Kiefer, and Manfred Krifka (eds.), Contrastiveness in Information Structure, Alternatives and Scalar Implicatures, Springer, 2017, pp. 339–415. https://doi.org/10.1007/978-3-319-10106-4_19 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

8. Kawahara, Shigeto. "The Phonology of Japanese Accent." In Haruo Kubozono (ed.), Handbook of Japanese Phonetics and Phonology, De Gruyter Mouton, 2015, ch. 11. Author pre-print: https://user.keio.ac.jp/~kawahara/pdf/HandbookAccentPublished.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵

9. NHK放送文化研究所 (NHK Broadcasting Culture Research Institute), ed. NHK日本語発音アクセント新辞典. 日本放送出版協会, 2016. https://www.monokakido.jp/ja/dictionaries/nhkaccent2/index.html ↩

10. Tokyo University of Foreign Studies (東京外国語大学), Language Modules, Japanese Pronunciation, Practical Edition, §1.9.1 イントネーション基礎. https://www.coelang.tufs.ac.jp/mt/ja/pmod/practical/01-09-01.php ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

11. Nagano-Madsen, Yasuko. "Mood Suffix and Question Intonation in Ryukyuan." In Proceedings of the 17th International Congress of Phonetic Sciences (ICPhS XVII), Hong Kong, 17–21 August 2011. https://www.internationalphoneticassociation.org/icphs-proceedings/ICPhS2011/OnlineProceedings/RegularSession/Nagano-Madsen/Nagano-Madsen.pdf ↩ ↩²

12. Hirayama, Hitomi. "Rising Declaratives in Japanese." In Japanese/Korean Linguistics 29, CSLI Publications, Stanford University. https://web.stanford.edu/group/cslipublications/cslipublications/site/JKONLINE/29/CH12.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵

13. Ueki, Kaori. "High Rising Intonation in Japanese Discourse." University of Hawai'i at Mānoa Working Papers in Linguistics, 36(7), 2005. https://scholarspace.manoa.hawaii.edu/items/bee75e57-821a-4338-9584-7ccd81e6deb9 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

14. Heycock, Caroline. "Japanese -wa, -ga, and Information Structure." In Shigeru Miyagawa and Mamoru Saito (eds.), The Oxford Handbook of Japanese Linguistics, Oxford University Press, 2008. Author pre-print: https://www.lel.ed.ac.uk/~heycock/papers/topic-draft.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸