Substitution with So, Do, One, Not

In answers to a positive question, so usually refers to the positive possibility, while not refers to its negation: Will it happen? — I think so / I hope not. With negative questions and complex contexts, a short answer may be ambiguous; in that case, it is better to state the idea in full.

Therefore, choosing so or not is not a question of politeness or length but of meaning: which version of the clause—the affirmative or the negative one—is the speaker confirming? Confusing so and not in an answer means confirming the opposite of what the speaker actually means.

Typical patterns are I think so, I’m afraid not, and it appears so. You do not need that before the substitute: say I think so or I think that it will happen. Do not combine the two patterns as I think that so.

So/not replace the intended content, but you can add a comment or explanation to the answer: I think so, at least for now. This does not prohibit all parenthetical words around the short answer.

This construction is lexically restricted: only verbs and expressions that introduce a whole clause expressing an idea about a possible fact allow that clause to be replaced by so/not. They include think, believe, expect, guess, hope, imagine, suppose, presume, reckon, assume, appear, seem, and the expression be afraid (meaning “unfortunately”).

Not every verb allows this substitution. Want requires other patterns: I want it to happen or I want her to. The verbs promise and decide can take a that-clause, but that alone does not make I promise so or I decide so ordinary short answers.

Check the specific established pattern in a dictionary. The ability to take a that-clause is not a sufficient test for so: for example, I know that… does not become the ordinary short answer I know so.

You can negate the idea in two ways: negate the verb expressing the idea (don't think so) or negate the substitute (think not). These two methods are NOT interchangeable for all verbs in the same way—each verb has its own usual pattern.

In modern conversational English, believe, expect, and think are negated with the auxiliary do: don't believe so, don't expect so, and don't think so. The forms believe not, expect not, and think not occur, but as literary or rhetorically formal forms, not as neutral everyday language.

By contrast, be afraid, guess, hope, and suppose naturally attach not directly to the verb: I'm afraid not, I guess not, I hope not, and I suppose not. With hope, this is not a stylistic preference but practically the only option: negating hope itself with don't is hardly used in this meaning—I don't hope so sounds like a mistake, not a neutral alternative.

Suppose allows both I suppose not and I don’t suppose so. Other verbs also have regional and context-dependent variants, so this list shows convenient typical answers, not every possible negation.

The forms believe not, expect not, and think not are not grammatical errors—they can be found in classical literature and in very formal or rhetorical modern contexts, for example in a polite but firm refusal during a public speech. In everyday conversation, they sound artificial or theatrical where the natural alternatives are don't think so, don't believe so, or don't expect so.

The difference should be understood as a matter of register, not as “correct versus incorrect”: both forms are grammatical, but the choice of form communicates something about the tone of the conversation. Learners should be able to recognize think not in a text and understand its meaning, but for their own neutral speech, don't think so is the more natural choice.

So here is not the same So that introduces a result (“It started raining, so we left”), nor is it the intensifier before an adjective (“so tired”). This is a separate, third function of so: a substitute for an entire clause. Context always indicates which function is intended—substitute so usually comes immediately after a verb of thought in a short reply.

You should also not confuse this construction with two neighboring types of substitution: do replaces a verb phrase (“Did you call her?” “I did”), while one/ones replaces a noun phrase (“Which jacket?” “The blue one”). Here, so/not replace the WHOLE clause—the subject and predicate together—not just the verb or just the noun.

Do not add a full clause without a grammatical link after so: I don’t think so he’s coming does not combine these patterns correctly. Choose I don’t think so or I don’t think he’s coming. A separate Yes, I think can be a conversational comment after the answer, but Yes, I think so is the clear pattern for this lesson.

The most natural place for this construction is a short reply to another speaker’s question or supposition, rather than an extended monologue. That is why it occurs so often in dialogues: the question states the clause, and the answer replaces it.

There is also a fronted version of the same idea: so can come at the beginning of a remark with reporting verbs—so I hear, so I'm told, so they say. Here, so also replaces an entire clause (“I hear that + clause”), but the source clause comes before so rather than after the verb. It is the same substitution, only with a different word order, and is typical when passing on information from another source with the suggestion “that is what I was told; I have not checked it myself.”

A verb phrase is recovered, but not necessarily every adverbial from the original sentence. The new part can have a different time or place: She called on Monday, and I did on Tuesday.

Do not add an object again when it is already included in the recovered meaning: choose did so or did it instead of did so it. Additional adverbials and the full emphatic pattern did + base verb are possible, however; these are other grammatical patterns.

The pro-verb changes form just like ordinary do: its tense form (do/does/did) agrees with the time and person it refers to, rather than mechanically copying the antecedent's form.

After have/has/had, use done; after a modal, use the base form do. With a to-infinitive, the safest options are to alone or to do so / to do it / to do that. Bare to do as a substitute is not equally natural in every variety and context.

Modal in the formula means a modal verb, such as can, will or should. The dots separate alternative patterns; this is not one long sentence.

Do, do so, do it and do that often convey related meanings, but register is not their only difference. Do so normally refers to an action or event; auxiliary do with ellipsis can also recover a state: She knows him, and I do too.

Do so is the most formal option, typical of reports, instructions and official texts. Bare do is neutral and natural in most contexts, including comparisons (than he does). Do it and do that are more conversational: they refer to an action as a specific thing already mentioned and sound natural in an everyday reply.

Check the particular sentence. Do not automatically add so to every short auxiliary: with know in the example, ordinary I do too does not require do so. Several alternatives may be acceptable when their structure and meaning fit the context.

When another auxiliary or modal verb already comes before do (have, will, might, should…), British English often adds do: I have done, I might do, I should have done.

In American English, this additional do is usually omitted, leaving the sentence to end with the auxiliary itself: I have, I might, I should have. Both alternatives are grammatically correct; this is a dialect difference, not an error in one of them.

Aux + do is characteristic of many British uses, but British English also allows the auxiliary alone. The label “British variant” does not mean that do must be added to every answer.

The preceding text provides a basis for recovering an action or state. The new subject can be different, and some pronouns can change their reference: She washed her hands, and I did too usually means that each person washed their own hands. Do not require every participant to match literally.

Tense and person agree with the new subject and the time of speaking rather than being copied mechanically: if the antecedent is in the Past Simple and the new action is also in the past, use did; if you mean another person's present habit, use do or does according to the subject.

If the first sentence is negative and the second affirmative, or vice versa, do does not change polarity by itself: negation or its absence must be marked separately (didn't do so, doesn't do it).

Auxiliary do forms questions, negatives and emphasis: Does she work here? I don't agree. I do agree. In short answers, it can remain without a repeated verb: Yes, she does.

Word order therefore does not tell you whether this is substitution or an auxiliary: both Did she? after a preceding statement and so did I are possible. Distinguish a full verb following do from an unspoken predicate recovered from context.

Do substitution replaces a verb phrase or predicate: “performed this action”. Clausal substitution is a different phenomenon: so or not, without do, replaces a whole idea or statement after verbs such as think, hope, believe or guess: “I think so”, “I hope not”.

The distinction is easy to check: if you remove so and leave only do in think so, the expression no longer works; do cannot function like that after think. Conversely, this verbal construction do so cannot be shortened to so alone: “did so” cannot be replaced by “so” without do.

In comparisons (than, as) and after coordinating conjunctions (and, but), bare do/does/did is usually enough; it sounds neutral in any register.

In business correspondence, instructions, legal and academic texts, do so contributes to a formal tone, especially after expressions such as failure to, in order to and be required to.

In conversation, do it and do that sound more natural when one person has just named a specific action and the other agrees, refuses or confirms it; do that can add a slight sense of distance or surprise (“You actually did that?”).

Aux + do often occurs in conversational British English. In writing, the choice depends on the genre and the particular sentence; do not automatically cross this form out just because the register is formal.

Use one for one item and ones for several. The earlier noun supplies the class, but its number can change: These chairs are nice. I'll buy one. Or: I need a chair. Are there any cheap ones?

Mentally restore the noun in the new phrase: a newer report → a newer one; newer reports → newer ones. Do not automatically make the substitute agree with the nearest word or the number in the previous mention.

One can stand alone: Do you have one? You can also add determiners and descriptions: a bigger one, the red ones, this one, my old one. When no description is supplied, some or any usually does the work instead of bare ones: Have you got pens? — I've got some.

When the substitute has no adjective and the reference is indefinite, do not put the indefinite article a/an before bare one. One itself means one member of the class: I need a pen — do you have one?, not a one. With an adjective, the indefinite article returns: Do you have a spare one?

Compare I lost my pen. Have you got one?, where any pen will do, with I found it, where the same pen was found. However, the one or this one can identify an already known item: This is the one I lost. One therefore does not always mean a different item.

Choose it to continue referring to a particular item. One/ones are useful when you recover the class and add a selection or description: the damaged one, another one, the ones on the shelf.

In their uncountable meanings, advice, luggage, traffic and information cannot be replaced with one/ones. Repeat the noun or use a suitable expression such as some or it. Not every adjective can naturally stand alone as a replacement.

Countability also depends on meaning. Coffee can name a substance, while a coffee can mean a serving: I ordered a coffee, but they brought the wrong one. Do not conclude that one is never possible merely because a word is also used with much.

Some words can already stand alone: the first, three, some and each, as well as the possessive forms mine, yours and hers. Do not simply drop a noun after my or her without changing the form: my umbrella → mine; her umbrella → hers. His can stand alone without a change.

Adjective + one/ones is a reliable pattern: the quieter one, the red ones. Shorter forms are also possible in a clear context, especially with a superlative or a choice between two: the best, the cheaper of the two. It would therefore be too strong to say that every adjective requires one.

These and those can stand alone or be followed by ones: these/these ones, those/those ones. Both patterns are used. The shorter form is often sufficient, while ones can help single out items. Its presence is not automatically an error or a sign of careless writing.

Mine, yours, hers, ours and theirs stand alone: This is mine, not mine one. Before an adjective, use a possessive determiner: my new one. Informal speech also includes my one, his one and similar forms; do not confuse these with mine one.

Compare the different units of substitution: so/not refer to a whole idea, as in I think so; do avoids repeating a predicate, as in I did; one/ones recover an item's name, as in a different one. All reduce repetition, but their structures differ.

One also has a separate use as a generic pronoun meaning a person in general: One should always check the facts first. Here it is the subject and does not replace a previously mentioned noun. The form is the same, but the grammatical function differs.