I'm Brazilian, and I need to know which British literature says 'to' is indispensable after the word 'ought'.

For example: Your skin color ought not to dictate your future.

Could you give me examples of grammar books?

  • 1
    yes, for exemple: Your skin color ought not to dictate your future. Feb 14, 2017 at 16:21
  • 2
    Ought to is a unit. In older English texts, or modern texts whose authors are trying to sound old, one occasionally comes across ought separated from its to (?_Ought I to awaken him?_) but nobody would ever speak English that way except on a stage. In Modern English, oughtta is sufficiently fused to have developed its own "eye spelling", like wanna, gotta, shoulda, wouldna, and hafta. Ought is a past participle and not a lexical modal, so it needs a to_to head a complement infinitive, and negatives can go before any infinitive with _to. Feb 14, 2017 at 16:22
  • 1
    So, in the phrase "Your skin color ought not to dictate your future." is to indispensable? Feb 14, 2017 at 16:26
  • 3
    @JohnLawler For me, the negative and interrogative versions don't take the to, so he ought not eat works just like he need not eat and he dare not eat for me. This all sounds very formal to me, and does not form part of my normal, non-oratorical speaking register. :)
    – tchrist
    Feb 14, 2017 at 16:37
  • 2
    I'm voting to close this question as off-topic because it is a request for resources. Feb 14, 2017 at 17:02

3 Answers 3


SUMMARY: While Americans don’t always have to use the to after ought in negative contexts, the English apparently must do so everywhere, even negatively.

The word ought can today act as a true modal auxiliary only in negative contexts, which includes interrogative ones albeit in a super-formal register.

By true modal, I mean it acts exactly like must does in that:

  1. It does not inflect for tense. There is no distinction between past and non-past.
  2. It does not inflect for third-personal singular present, so there's never an -s.
  3. It takes a bare infinitive, not a to-infinitive.
  4. Its contracted negative undergoes subject–verb inversion dragging the negative particle along with it: Oughtn’t he ...

Negative examples taking a bare infinitive

  • One ought never add a to particle to this sentence.
  • Never ought one add a to particle to this sentence.
  • He oughtn’t bother her today.
  • You told me he oughtn’t bother her yesterday.
  • Oughtn’t she call him?
  • Ought she call him?

Positive examples taking a to-infinitive

However, in non-negative/interrogative contexts, we must relax modal trait #3 above and supply said particle.

  • One ought to add a to particle to this sentence.
  • He ought to bother her today.
  • You told me yesterday that he ought to bother her.
  • She ought to call him.

The positives all take a to particle, but the other modal properties still apply.

See also the other two semi-modals, need and dare, for the circumstances under which they, too, behave like full modals.


In your example from your comment:

Your skin color ought not *to dictate your future.

That is not something I can say in my own private idiolect, because for me the not needs to cancel the to for me:

Your skin color ought not dictate your future.

The other way sounds super old-fashioned to me. I understand that some green and pleasant pockets of England may still use the to-infinitive version in positives, though.

Corporal research has suggested that skipping the to for negatives and interrogatives is no longer the more common way of doing things, and that most people put it there “obligatorily” even in negatives and interrogatives than those like me who do not. There are still a few of us, however. See next.


Here are a few Google books citations that forego the to particle in negatives:

  • If a goal is wrong, we ought not try to achieve it. Only if it is right, ought we to try.
    Aristotle for Everybody, Mortimer J. Adler, 1997

  • Sometimes "ought" means what one should do, all things considered — as in "In these circumstances what you ought to do is start over" or "Given the situation, I think you ought not press your right to x."
    Ethics in Engineering Practice and Research, Caroline Whitbeck, 1998.

  • 1.Scientists ought not do research that causes unjustified risks to people. 2. Scientists ought not do research that violates norms of free informed consent. 3. Scientists ought not do research that unjustly converts public resources to private profits.
    Ethics of Scientific Research, Kristin Sharon Shrader-Frechette, 1994.

  • A normative rule (for example, "one ought to do A given B") is defeasible in a normative system S iff S contains another rule to the effect that one ought not do A given B&C or one is permitted not to do A ...
    Defeasible Deontic Logic, Donald Nute, 1997.

  • The lover of art who doesn't feel at ease when confronted with contemporary art ought not attack it, nor should he force himself to take pleasure in it.
    Soul of the Age: Selected Letters of Hermann Hesse, 1891–1962, Therodore J. Ziolkowski, 1991.

  • He cringed to remember them meeting in the Circle Ritz lobby this late morning, both of them looking like they had been for a long time somewhere they oughtn't admit to.
    Cat on a Hyacinth Hunt: A Midnight Louie Mystery, Carole Nelson Douglas, 1999.

  • Shouldn't she, if she believed that I took pleasure in cruelty, want not to be with me, oughtn't she be afraid of the pain I'd caused her?—
    First Love and Other Sorrows: Stories, Harold Brodkey, 1998.

  • Recognizing that a consistent and uniform approach to surrogacy was needed, the ministers agreed that "there were too many unknowns, too many uncertainties and that we oughtn't experiment ...
    Women as Wombs, Janice G. Raymond, 1993.

  • He put down his rifle, walked over, and grabbed the other end of the saw. In no time at all the two men zipped through the logs. "After all," he said later, "it was cold and one man oughtn't work a two-man saw.
    Stark Decency: German Prisoners of War in a New England Village, Allen V. Koop, 2000.

  • Hermann Hesse - 1971 Oughtn't so early an event as the Munich Putsch have shown them what he was?
    If the War Goes on: Reflections on War and Politics, Hermann Hesse, 1971.

I must point out that although examples of fully modal ought like these can be found in recent publishings, they are in the minority, and more people today now use a to-infinitive even in negative and interrogative context than, like these, do not.

Transatlantic Differences

As explained here, Quirk et al. in A Grammar of Contemporary English report that American English sometimes does exactly what I have just explained:

Ought regularly has the to-infinitive, but AmE occasionally has the bare infinitive in negative sentences and in questions (although should is commoner in both cases):
- You oughtn't smoke so much.
- Ought you smoke so much?

On the other hand, noted English authority :–) Barry England here reports that:

British English requires the to-infinitive. (I didn't know until reading the above comments that American English allowed its omission.)

I will leave it to others to trace the history of this.

  • This Q came up flagged for closure, and I voted to close as a "request for resources". Then I saw your lovely answer. Should I retract the vote? Feb 14, 2017 at 17:06
  • @Cascabel Como quieras. I’ve asked her for clarification as I may have misparsed her question, inviting her to clarify me it in Portuguese if she'd like. She seems to have liked my answer, so perhaps one can edit the question to make this the answer to it. :)
    – tchrist
    Feb 14, 2017 at 17:07
  • I will wait on it for a while. Ciao for now. Feb 14, 2017 at 17:09
  • Ought I add that the interrogative works too?
    – fectin
    Feb 15, 2017 at 1:34
  • Thank you for your answer. It has helped me learn the 'ought' usage well. Feb 15, 2017 at 4:28

What is special about the modal auxiliary 'ought' is that it takes the 'to infinitive' after it.

Ought to

from English Grammar Today (http://dictionary.cambridge.org/grammar/british-grammar/modals-and-modality/ought-to)

Ought to is a semi-modal verb because it is in some ways like a modal verb and in some ways like a main verb. For example, unlike modal verbs, it is followed by to, but like modal verbs, it does not change form for person:

I ought to phone my parents.

It ought to be easy now.

Ought to: form Affirmative

Ought to comes first in the verb phrase (after the subject and before another verb):

We ought to do more exercise.

Ought to cannot be used with another modal verb:

Medicine ought to be free.

Not: Medicine ought to can be free. or Medicine can ought to be free.


The negative is formed by adding ‘not’ after ought (ought not to). It can be contracted to oughtn’t to. We don’t use don’t, doesn’t, didn’t with ought to:

We ought not to have ordered so much food.

Not: We don’t ought to have ordered so much food.

You oughtn’t to have said that about his mother.

Not: You didn’t ought to have said that about his mother.

The negative of ought to is not common. We usually use shouldn’t or should not instead:

You shouldn’t speak to your father like that. (preferred to You oughtn’t to speak …)

Also from Practical English Usage by Michael Swan:

enter image description here

  • Are you alleging that the bare infinitives used after negative and interrogative instances of ought are somehow ungrammatical? Quirk et al. in A Grammar of Contemporary English write that “Ought regularly has the to-infinitive, but AmE occasionally has the bare infinitive in negative sentences and in questions (although should is commoner in both cases): - You oughtn't smoke so much. - Ought you smoke so much?”
    – tchrist
    Feb 14, 2017 at 17:20
  • 2
    @tchrist: No. The OP asked for any British 'literature' that says 'to' is indispensable after ought'. Although my citations do not say that it is indispensable, they do not accept using the bare infinitive after 'ought'. After all, grammaticality and acceptability need not always contradict! Feb 14, 2017 at 17:29
  • 1
    Oh I see. You've read her question more closely than I did, so +1 to you.
    – tchrist
    Feb 14, 2017 at 17:49
  • "Prove otherwise authentically"? What about all the citations I gave is inauthentic? What about Quirk is inauthentic? I've never ever heard anyone suggest that Quirk was invalid or non-authentic. You may be suffering from a locality illusion in which you misconstrue the rules of your own local dialect as ones governing the entire Anglosphere. That's a bad idea for many reasons.
    – tchrist
    Feb 14, 2017 at 18:01
  • 1
    Oh, sorry! I take back my words. I failed to go through your citations. Feb 14, 2017 at 18:10

Inspired by tchrist's outstanding answer, I did some research into a particular negative form of ought without to: the phrase "ought not be." I should first emphasize that "ought not be" is considerably less common than "ought not to be." Here is the Ngram chart for "ought not be" (blue line) versus "ought not to be" (red line) for the period 1650–2000:

Here is the effectively magnified view of the same chart for the period 1800–2000:

And here is the further effectively magnified view of the chart for the period 1940–2000:

The line for "ought not be" looks exceedingly weak. But in fact, Google Books searches find a substantial number of matches for the phrase, dating back to the 1500s and persisting despite hostility from grammar prescribers at least as early as T.O. Churchill, A New Grammar of the English Language; Including the Fundamental Principles of Etymology, Syntax, and Prosody (1823), who tolerates the loss of "the sign of the infinitive mood" (that is, the word to) in poetical settings but not in prose:

'What, know you not,/ That, being mechanical, you ought not walk,/ Upon a lab'r'ing day, without the sign/ Of your profession?' [—]Shaks., Jul. Caes.

'To wish him wrestle with affection. ' [—]Ib., Much Ado.

'Nor with less dread the loud/ Ethereal trumpet from on high 'gan blow.' [—]Milton, P. L., vi, 60.

These phrases are poetical, and by no means allowable in prose. To use ought, or cause in this manner is a Scotticism. 'Won’t you cause them remove the hares?'

The marginal status of "ought not [verb]" (without the to) in British usage is apparent in Ernest Gowers's update of H.W. Fowler, A Dictionary of Modern English Usage (1926, 1965):

Ought, the past tense of owe (now used as present also) is the only surviving form of the verb in its sense of be under a duty to, or be expected to. An auxiliary cannot therefore be used with ought as though it were an infinitive, and it must be negatived with a bare not in the old-fashioned way; you ought not to have done that.

Gowers's comment arises in the context of his opposition to the wording "you didn't ought to have done that," which he calls "a not uncommon colloquial vulgarism." But the necessity of including to in the form of the expression he prefers is so obvious to him that he doesn't bother addressing the alternative "you ought not have done that" at all.

What is dictated from on high is not always what is practiced down below, however. Following are some instances of "ought not be" from published works through the years (ending at 1900, because enough is enough).

From English Recusant Literature, 1558–1640, volume 244 (1969) [combined snippets]:

Finallie, Because Fiſʒher afffirmeth that the knowledge of Purgatorie came in predetentim, by little and little, therefore it ought not be admitted, nor esteemed. For by the same Logick he may proue, that S. Iames his epistle ought not to be admitted for Canonicall Scripture because (as S. Hierome doth witness) by little and little, in process of tyme, it obtained authoritie and credit.

From Stephen Batman, Batman vppon Bartholome: His Booke De Proprietatibus Rerum, Newly Corrected, Enlarged and Amended with Such Additions as Are Requisite Vnto Euery Seuerall Booke, Part 1 (1582):

For the eie ought not be strained too far out, neither lifted vp too high. For that betokneth disturbance of discretion:neither to deep in. For [it] betokeneth default of matter [or] of vertue.

From Robert Monro, Monro, His Expedition with the Worthy Scots Regiment (Called Mac-Keye-Regiment) Levied in August 1626 (1637):

A Commander ought not be ignorant of the circumstances belonging to the quartering of an Army , therefore for the better Information of the younger sort, who have not seene such Marches as I have bin at, with his Majesty of worthy memory , who quartered his Army Summer or Winter, according as the occasion or neerenesse of his enemy did offer, ...

From Robert Wilson, ‎Peter Talbot, The Friar Disciplind, or, Animadversions on Friar Peter Walsh His New Remonstrant Religion (1674) [one of five instances of "ought not be" in this work]:

For though an answer did not inuolue heresy or manifest error in the Catholick faith , yet if it inuolues nonsense , or a plain contradiction , it inuolues an error against natural reason obuious to euery man, ( except Peter Walsh ) and therfore it ought not be taken for a good answer ; ...

From John Fitz-Gerald, "The Narrative of Mr. John Fitz-Gerald, Late of the Order of St Francis, in the Kingdom of Ireland" (1681):

... I do expect your Reason and Understanding will lead you to grant me undeniably, when you read and consider the total course and whole passages of these three Suborners, and also when you understand to what intent and purpose they designed their Undertakings, in this their work of Subornation, that they should not, nor ought not, be received, or accepted, but as they are, and as they may be properly called, Perjurors, ...

This last example is interesting because it puts "ought not" in parallel with "should not," which (to my knowledge) never takes a to before the following verb.

From John Donne, "Elegie XVII: The Expostulation," in Poems on Several Occasions (1719):

All which were such soft pastimes, as in these/ Love was as subtily catch'd, as a disease;/ But being got it is a treasure sweet,/ Which to defend is harder than to get:/ And ought not be prophan'd on either part,/ For though 'tis got by chance, 'tis kept by art.

From remarks by the Earl of Cholmondeley in the House of Lords at the First Session of the Third Parliament of King George II (December 4, 1741), in The History and Proceedings of the House of Lords, volume 8 (1743):

That the justest Intentions may be sometimes defeated, and the wisest Endeavours fail of Success, I shall readily grant ; but it will not follow that we ought not to acknowledge that Wisdom and Integrity which is exerted in the Prosecution of our Interest, or that we ought not be grateful for the Benefits which were sincerely intended though not actually received.

From "Summary of the Proceedings in the House of Commons of Ireland, during the Present Session of Parliament" (March 15, 1787) in The Gentleman's and London Magazine (October 1787):

He [Mr. Curran] then proceeded to argue in support of the petition.——This clause enacted by reference a foreign act. Where was the act to be found, if pleaded in our courts? This mode of adoption might have answered the reign of Hen. 7. when the power of England to bind us was admitted ; it was necessary from the urgency of the occasion, 1782—but it was not now necessary, and therefore ought not be done.

From Charles Putt, Essay on Civil Policy, or the Science of Legislation (1830):

If, therefore, it can be shown, that a promise to marry ought not to bind, it follows, that damages ought not be given for the breach or non-performance.


But even supposing the law remodeled, and writing made essential, I am still of opinion that even such a promise ought not be deemed obligatory.

From William Forsythe & Appleton Morgan, History of Trial by Jury (1852):

SECTION V. Question of new Trial in Cases of Conviction of Felony.

A question of great importance has often been raised, whether in criminal cases there ought not be an appeal from the verdict of the jury on' matters of fact. In the English and Scotch law it is unknown, and a conviction of felony can not be questioned by any form of legal process, on the ground that the verdict was not warranted by the evidence.

From a petition to Queen Victoria, quoted in "Destitution in England an Emigration," in the Warwick [Queensland] Examiner and Times (January 22, 1870):

We respectfully submit that your Majesty's colonial possessions were won for your Majesty, and settled by the valor and enterprise and treasure of the English people, and that, having thus become a part of the national freehold and inheritance of your Majesty's subjects, they are held in trust by your Majesty, and ought not be surrendered, but transmitted to your Majesty's successor whole and entire as they were received by your Majesty.

From Prince Josef Lubomirski, The Ace of Clubs: A Romance of Russia and Siberia, serialized in the [San Francisco California] Morning Call (May 5, 1890):

"I think," said the doctor, "as the Countess has the express permission of the Emperor, she ought not be delayed on her Journey."

And from "The Ex-Slave Pension Scheme Ventilated in U.S. Senate," in the [Fredericksburg, Virginia] Free Lance (December 16, 1899):

"The bill ought not be even presented to Congress," was the public declaration of Senator Gallinger, who produced a remarkable certificate. It was signed by an aged colored woman in Nashville, where the Ex-Slave Bounty and Pension Association has its headquarters, who promised to pay 25 cents for the privilege of becoming a member and 10 cents a month for dues. "I hereby certify that I was born a slave," wrote the old negro woman, "and am entitled to all the benefits included in said bill."


The objections of T.O. Churchill, Lindley Murray, and other proponents of grammarosity during the 1800s seem to have succeeded in making "ought not be" and similar phrases disreputable, at least in book publishing houses: occurrences of the phrase in Google Books search results drop sharply around 1835 and are very rare indeed during the second half of the nineteenth century.

But the phrase is not at all uncommon in newspaper articles during the same period. In fact, an Elephind search for "ought not be" across the period 1850–1920 (they heyday of prescriptive grammar) yields 4,436 matches from U.S. and Australian newspapers. Even allowing for a sizable number of duplicate occurrences of the same instance in multiple newspapers, that’s a lot matches.

Why did newspaper usage resist the push to abjure use of "ought not be"? The simplest (and in this case, perhaps, most probable) explanation is that newspapers were echoing spoken English, where the wording remained common despite the best efforts of Churchill, Murray & Co.

The upshot of all this is that the form "ought not be" has probably been in everyday use in English for more than four centuries. To say that the to in "ought not to be" is indispensable is to deny the persistent use of "ought not be" across that entire period.

  • I have added a third magnification of the chart (for the period 1940–2000), to show what looks to me much more like persistent use of "ought not be" than statistically insignificant noise. What I find most striking about the third chart is the decline in frequency of use of "ought not to be" by more than half since 1940, while the frequency of use of "ought not be" has remained remarkably steady. The decline in frequency of use of "ought not to be" since 1800 is really quite impressive.
    – Sven Yargs
    Feb 15, 2017 at 20:24
  • Please see this Elephind search page for links to 5,094 matches for "ought not be" found in newspapers published in Australia and the United States between 1800 and 1939. The Library of Congress offers data from its huge newspaper database only through 1924. As I said in my answer, there are a fair number of duplicates, but there are still many, many unique instances.
    – Sven Yargs
    Feb 15, 2017 at 21:15
  • 2
    @MilesRout That is utterly false. Ought not be is used, in both speech and writing, in American English. To many AmE speakers (such as tchrist) ought not to be is even ungrammatical: the negation mandatorily cancels out the infinitive marker just as it does with need (it doesn't need to be, but it need not be; there is no *it need not to be). Plus, even if it were just erratic noise, it would still be expected to decline along with the ‘real’ form. But the to-less form now accounts for almost one fifth of all occurrences; that's not noise. Feb 15, 2017 at 23:22
  • 1
    @MilesRout That's not how language works. Feb 16, 2017 at 7:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.