～でしょう / ～だろう: Conjecture and Confirmation in Japanese

でしょう and だろう express the speaker's conjecture ("probably / I'd say so"). With rising intonation, they seek the listener's confirmation ("...right?").¹ They are the same auxiliary in two registers: でしょう is polite, and だろう is plain, paired the way です pairs with だ.¹

Overview

でしょう / だろう is an auxiliary that marks what the speaker takes to be true without asserting it as fact. A guess with this form is "not based on any particular information or evidence," and its plain gloss is "probably."¹

The same form does two pragmatic jobs, and intonation alone tells them apart. A falling tone reads as conjecture ("probably"). A rising tone seeks agreement ("...right?").¹ This article covers both uses, then places the form on the wider certainty scale from かもしれない up to にちがいない.²

The two jobs in one form

The conjecture reading is the default. The speaker offers an estimation rather than reporting something observed.¹ The hedge is built in: the auxiliary signals reasoning, not direct or indirect evidence for the claim.³

The confirmation reading appears when the same form is spoken with rising intonation and no か. Here the speaker already believes the statement and invites the listener to agree, as in 君も行くだろう？ ("You will go too, right?").¹

The か question is a third, softer use

Adding か turns the form into a softer, less direct question. 大丈夫ですか asks "Is it all right?" 大丈夫でしょうか reads more like "I wonder if it's all right."¹ This article focuses on the falling-tone conjecture and the rising-tone confirmation, the two jobs of the bare form.

でしょう and だろう as one paired form

だろう began as the informal conjecture form of the copula だ and was repurposed as a conjecture auxiliary; でしょう is its formal counterpart.¹ The pair maps onto the だ / です split: だろう is plain, and でしょう is polite.¹⁴⁵

The form lives in the epistemic-modality layer of the grammar, which marks how sure the speaker is. A Dictionary of Basic Japanese Grammar groups its relatives as かもしれない, にちがいない, ようだ, らしい, and そうだ, the conjecture and inference cluster.¹

For JLPT level, treat the placement as approximate. でしょう is the beginner form, tagged N5 by recognized reference lists. The plain だろう, along with its register and certainty nuance, belongs to the N4 layer.⁴⁵⁶⁷ No official Japan Foundation grammar-by-level list exists after the 2010 test redesign, so every "N5 / N4" placement is a publisher reconstruction rather than an official ranking.⁴⁵

Form: how to attach でしょう / だろう

でしょう / だろう attaches to the plain (informal) form of the predicate: Sentence(informal) + だろう / でしょう.¹ The auxiliary itself does not conjugate. Tense and polarity are carried by the predicate it follows.¹

Attachment to verbs, い-adjectives, な-adjectives, and nouns

Verbs and い-adjectives attach directly, because no だ is involved. After a な-adjective stem and after a noun, the だ drops before だろう / でしょう.¹ So 学生でしょう is correct, and 学生だでしょう is not.

Word class	Plain form	+ でしょう (polite)	+ だろう (plain)
Verb	行く	行くでしょう	行くだろう
い-adjective	高い	高いでしょう	高いだろう
な-adjective (だ drops)	静か(だ)	静かでしょう	静かだろう
Noun (だ drops)	学生(だ)	学生でしょう	学生だろう

The attachment pattern and the だ-drop come from the formation chart in A Dictionary of Basic Japanese Grammar.¹

あのアパートは高たかいでしょう。¹
"That apartment is probably expensive."

ロジャーはスキーが上手じょうずだろう。¹
"Roger is probably good at skiing."

あの人ひとは中国人ちゅうごくじんだろう。¹
"That man is probably Chinese."

Do not keep だ before でしょう / だろう

The most common attachment error is leaving だ in place after a noun or な-adjective. Write 学生でしょう, never 学生だでしょう. The だ that would normally close the predicate is absorbed by the conjecture form.¹

Past and negative bases

Because the auxiliary attaches to a plain-form predicate, the predicate supplies tense and negation. でしょう / だろう does not change shape for past or negative. It sits on 話した just as readily as on 話す, and on 来ない just as readily as on 来る.¹

彼かれはもう話はなしたでしょう。¹
"He has probably already spoken."

彼かれは来こないでしょう。¹
"He probably won't come."

でしょう and the volitional form: a shape worth keeping apart

The polite volitional ～ましょう and the plain volitional ～よう / ～おう are separate morphemes that express invitation or intention ("let's …" / "I shall …"), not conjecture.¹ でしょう / だろう, by contrast, attaches to a complete plain-form predicate and expresses a guess.

The point is functional rather than formal. 明日晴れるでしょう ("it will probably clear up") is a conjecture. 行きましょう ("let's go") is an invitation. They do different pragmatic work even though both end in a ～しょう / ～よう shape.¹

Nuance and usage contexts

Conjecture: "probably / I'd say" (falling intonation)

The core reading is the speaker's estimation, not asserted fact. The auxiliary marks a conjecture "not based on any particular information or evidence."¹ Hara formalizes this as an epistemic bias derived from reasoning. She notes that だろう is infelicitous when the speaker has direct or specific observed evidence for the claim.³

明日あしたは雨あめでしょう。¹⁸
"It will probably rain tomorrow."

関東地方かんとうちほう、明日あしたは小雨こさめが一日中いちにちじゅう降ふるでしょう。²
"Tomorrow in the Kanto region it will probably drizzle all day."

Confirmation: "...right?" (rising intonation)

With rising intonation and no か, だろう / でしょう asks for the listener's agreement.¹ The speaker already holds the belief and uses the form to invite confirmation.

君きみも行いくだろう？¹
"You'll go too, right?"

これ、きれいでしょう？¹
"This is pretty, right?"

面白おもしろいでしょう？¹
"It's interesting, right?"

A natural question is how this differs from the agreement-seeking particle ね. Both ask the listener to agree, but A Dictionary of Basic Japanese Grammar puts rising だろう / でしょう on the softer, less direct side of that pair.¹ The minimal contrast is 君も行くだろう？ ("You will go too, right?") against 君も行くね ("You will go too, won't you?"). The ね version comes across as the more direct request for agreement.¹

Rising でしょう？ is the softer option, not the pushier one

Some learner explanations frame でしょう？ as a presumptive "you must agree" move. The primary source points the other way: rising だろう / でしょう is softer and less direct than ね when seeking agreement.¹ Reach for it when you want to invite agreement gently rather than press for it.

Where でしょう / だろう sits on the certainty scale

This is the form's anchor position in the modality system. A Dictionary of Basic Japanese Grammar gives the ladder explicitly: a かもしれない sentence predicts lower accuracy than a だろう sentence, and much lower accuracy than a にちがいない sentence.² The diagram runs from low probability to high as かもしれない < だろう < にちがいない.²

The table below restates the same ordering with the rough force of each marker.

Marker	Force	Gloss
かもしれない	possibility, below 50% on the diagram's low end	"might / may"
だろう / でしょう	bias above 50% (Masuoka: roughly 50–80%)	"probably / I'd say"
にちがいない	high certainty	"must be / no doubt"

For a numeric anchor, Masuoka characterizes だろう as expressing roughly 50–80% probability. Hara reinterprets this as a bias of more than 50% (p more likely than not-p). On her account, the 80%-and-non-100% intuition arises as a cancelable pragmatic implicature rather than a fixed lexical value.³⁹ A separate marker, はず (logical expectation), sits nearby in the same modality cluster.

Epistemic adverbs that pair with でしょう / だろう

Probability adverbs such as たぶん, おそらく, and きっと can come at the front of the clause and fit naturally with the conjecture reading. The guess "sounds more certain with たぶん or おそらく, and even more certain with きっと," so きっと sits at the high-confidence end of the three.¹

アンダソンさんはたぶん日本にほんへ行いくだろう。¹
"Ms. Anderson will most probably go to Japan."

アンダソンさんはきっと日本にほんへ行いくだろう。¹
"I'm almost certain Ms. Anderson will go to Japan."

おそらく彼かれは来こないでしょう。¹
"He likely won't come."

The pairing is selective. だろう co-occurs with たぶん and きっと but is incompatible with もしかすると ("maybe"). That adverb signals low probability, below 50%, and so clashes with だろう's above-50% bias.³¹⁰ Among the compatible adverbs, おそらく is the more formal, written-register member of the たぶん–おそらく pair.¹

Register: the weather-forecast でしょう

でしょう is the standard register of Japanese forecasts. A Dictionary of Basic Japanese Grammar states it plainly: weather presenters on radio or TV use でしょう, the formal version of だろう, in forecasts. It cites 関東地方、明日は小雨が一日中降るでしょう。²

明日あしたは晴はれるでしょう。²⁸
"It will clear up tomorrow."

気温きおんは上あがるでしょう。²⁸
"Temperatures will rise."

The institutional reason is policy. The Japan Meteorological Agency's terminology guidance expresses a forecast's inherent uncertainty by fixed rules rather than over-specifying it. That makes a measured-prediction form like でしょう the house style.⁸ It reports a likely outcome without claiming certainty, exactly the "conjecture not based on particular evidence" sense applied institutionally.¹²

Good to know

Bare だろう reads as strongly male in casual speech

だろう is the plain conjecture form, the だ-side of the だ / です split, while でしょう is the polite form usable by anyone.¹⁵ In casual speech, plain だろう carries a strongly masculine register, especially as a confirmation だろう？. It sits within the danseigo (men's-speech) set of assertive plain endings, where women more often soften to でしょ, a rising の, or other だ-series softeners.¹¹

Treat this as a tendency, strongest in the bare casual confirmation use, not as an absolute rule. The polite でしょう is the gender-neutral, safe choice for every speaker.¹¹¹

The contracted だろ is colloquial and out of scope here

The clipped だろ (だろ？) is a casual variant of だろう covered elsewhere; this article keeps to the long forms だろう / でしょう.

でしょう is conjecture, not flat certainty

Learners often over-translate でしょう as a plain "will," which drops the hedge the form always carries.¹³⁹ Treating 明日晴れるでしょう as a guarantee of sunshine misstates it, because the form predicts rather than asserts. For an asserted fact, the plain present tense does the job.

明日あした晴はれます。¹
"It will be sunny tomorrow."

Set against 明日晴れるでしょう ("it will probably be sunny tomorrow"), the contrast is assertion versus prediction. でしょう keeps only a roughly 50–80% bias, never the full certainty of a flat statement.¹³⁹

Same form, intonation decides the job

A falling でしょう。 reads as "probably": the speaker is telling. A rising でしょう？ reads as "right?": the speaker is asking for agreement.¹ The two readings share one written form, so intonation separates them in speech.

だろう is the copula だ wearing its conjecture hat

だろう was originally the informal conjecture form of the copula だ, then repurposed as a conjecture auxiliary, with でしょう as its formal counterpart.¹ Reading だろう / でしょう as "the maybe-version of だ / です" makes both the attachment rule and the だ-drop intuitive: the conjecture form simply takes the slot the plain copula would have filled.

See also

• ～かもしれない vs ～にちがいない: Possibility and Certainty in Japanese
• ～はず: How to Express Logical Expectation in Japanese
• Inferential Suffixes in Japanese: ～そう, ～よう, ～らしい, ～みたい Compared
• Tense, Aspect, and Mood in Japanese: A Map
• The Japanese Copula: です, だ, である Explained
• Polite vs. Plain Japanese: です/ます vs. だ (丁寧体・普通体)

References

Footnotes

1. Makino, Seiichi, and Michio Tsutsui. A Dictionary of Basic Japanese Grammar. The Japan Times, 1986. Entry "darō / deshō," pp. 100–102. (Romanization in this source: "darō / deshō," with long-vowel macron; here matched as "darō / deshō" on first cite, then "だろう / でしょう".) ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹ ↩²² ↩²³ ↩²⁴ ↩²⁵ ↩²⁶ ↩²⁷ ↩²⁸ ↩²⁹ ↩³⁰ ↩³¹ ↩³² ↩³³ ↩³⁴ ↩³⁵ ↩³⁶ ↩³⁷ ↩³⁸ ↩³⁹ ↩⁴⁰ ↩⁴¹ ↩⁴² ↩⁴³ ↩⁴⁴ ↩⁴⁵

2. Makino, Seiichi, and Michio Tsutsui. A Dictionary of Basic Japanese Grammar. The Japan Times, 1986. Entry "kamoshirenai," pp. 174–176, including Related Expressions I (the kamoshirenai < darō < ni chigainai probability diagram and the weather-forecast note). ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

3. Hara, Yurie. "Non-Propositional Modal Meaning." Manuscript, University of Delaware / University of Massachusetts, Amherst, January 31, 2006. https://www.semanticsarchive.net/Archive/WUxZjFiM/darou_hara.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

4. JLPT Sensei. "でしょう (deshou)" grammar entry, tagged JLPT N5. https://jlptsensei.com/learn-japanese-grammar/%E3%81%A7%E3%81%97%E3%82%87%E3%81%86-deshou-meaning/ (limitation: third-party JLPT-tagging site; no official post-2010 JLPT grammar list exists.) ↩ ↩² ↩³

5. JLPT Sensei. "だろう (darou)" grammar entry. https://jlptsensei.com/learn-japanese-grammar/%E3%81%A0%E3%82%8D%E3%81%86-darou-meaning/ (limitation: third-party JLPT-tagging site.) ↩ ↩² ↩³ ↩⁴

6. Bunpro. Grammar point "でしょう," tagged JLPT N5. https://bunpro.jp/grammar_points/9 (limitation: third-party JLPT-tagging site.) ↩

7. Bunpro. Grammar point "だろう." https://bunpro.jp/grammar_points/%E3%81%A0%E3%82%8D%E3%81%86 (limitation: third-party JLPT-tagging site.) ↩

8. 気象庁 (Japan Meteorological Agency). 予報用語「天気予報等で用いる用語」(解説). https://www.jma.go.jp/jma/kishou/know/yougo_hp/kaisetsu.html ↩ ↩² ↩³ ↩⁴

9. Masuoka, Takashi (益岡隆志). 『モダリティの文法』. くろしお出版, 1991. (Cited via Hara 2006³ for the 50–80% probability characterization of だろう.) ↩ ↩² ↩³

10. Sugimura, as cited in Hara 2006³ (Sugimura 2004): the co-occurrence of だろう with たぶん / きっと but not with もしかすると ("maybe," low probability). ↩

11. SturtzSreetharan, Cindi L. "Ore and omae: Japanese men's uses of first- and second-person pronouns." Pragmatics 19:2 (2009), pp. 253–278. https://benjamins.com/catalog/prag.19.2.06stu (used for the danseigo / men's-speech framing within which the masculine plain-form association sits). ↩ ↩²