## "is" vs. "are" and when to use them

5

2

I have been debating the usage of these two forms of the word "To Be":

1. Is the number and address valid?
2. Are the number and address valid?

Which form of the verb is correct for the sentence?

1

– tchrist – 2014-12-18T23:16:02.763

@tchrist that answer is not applicable as the subjects in question are two completely distinct things, not a single unit. – None – 2014-12-19T00:15:43.420

@F.E. this debate started from a Code Review Answer about a method name. I am comfortable with the first version (with is) but someone has suggested it should be are – Malachi – 2014-12-22T19:39:17.873

there really isn't a standard here so I want to make sure that it is readable, and I want my code to be almost as though I am reading plain English when I look through the code. that is why I want to know what is right on this particular instance – Malachi – 2014-12-22T20:07:04.297

@Malachi Just read your last comment. If this is programming, then you'd want to follow the style used by the exiting code. How do they word similar functions/methods? (If it was my code and I had the freedom to name it as I wanted, then would probably use something like "Is_number_and_address_valid()" or "IsValid_number_and_address()" or "Is_valid_NumberAddress()" or etc. But you are in your own situation, and should follow the workplace style.) – F.E. – 2014-12-22T20:12:42.390

5

1. Is the number and address valid?
2. Are the number and address valid?

Which form of the verb is correct for the sentence?

Your 1st version ("is") sounds okay to my AmE ear. The 2nd version ("are") sounds awkward to me. If you're not comfortable in using the 1st version ("is"), then you could reword your example sentence into another form, one which you'd feel more comfortable with.

I'd think it reasonable to think that the 2nd version ("are") would not be fully acceptable to many people when it is heard in an informal or casual setting.

A big problem with the 2nd version ("are") is that the listener first hears the plural verb "Are" which is then followed by a singular noun phrase "the number", and that is jarring due to the clash in grammatical number. The listener was expecting to hear a plural subject, but instead gets what seems to be a singular subject.

Though the subject in its entirety would end up being plural in number, that doesn't help during the initial stages when the listener is hearing the spoken sentence. For example, consider:

• A. "Are a dog--"

This immediately sounds awkward, due to a violation of proximity concord. (Note: proximity concord is when the verb agrees in number to the closest noun phrase. In general, the listener will initially assume that the closest noun phrase is the intended subject, and often that noun phrase is located right before the verb.)

Let's continue and finish the rest of that spoken sentence:

• B. "Are a dog and a cat on the gift list?"

We can see that version #B does have formal concord with respect to subject-verb agreement. Both the subject and the verb are plural in number, but that can only be seen after the whole sentence has been processed. Notice how this problem (in #A/B) does not exist in the corresponding declarative clause version:

• C. "A dog and a cat are on the gift list." -- (declarative clause)

There's no corresponding problem here because the listener gets to process the whole subject ("A dog and a cat") first, and gets to realize that the subject is plural. And so, the listener is now expecting a plural verb to come along soon.

But unfortunately, in version #A/B the verb comes first before the subject. That same problem of verb arriving before its subject is in the OP's version #2 ("Are the number and address valid?"). And that is why the OP's version #2 is awkward to many people -- the violation of proximity concord.

### LONG VERSION:

Some common abbreviations:

• NP == noun phrase

NOTE: The rest of this here post is gonna be something somewhat like a rainy day post. You'll see what I mean by how I ramble here and there, as I attempt to answer the OP's question. But there's a reason for the meandering, maybe. And the rambling is probably necessary, due to what all has to be touched upon. And so, here we start . . .

The subject in the OP's examples is the noun phrase (NP):

which is in the form of a coordination of nouns. In general, the number of the verb (in the OP's case, the verb "is" or "are") will often match the number of the subject, but not always. When they do match, then subject-verb agreement will have formal concord (concord is explained further on down below).

There is a misconception out there that subject-verb agreement can be defined by a few simple rules, and that it is, in general, easy to figure out. But it isn't definable by a few simple rules, and there are many so-called exceptions, which are commonly found in the wild. A major factor in determining subject-verb agreement, w.r.t. grammatical number, is in determining the subject's grammatical number, but that is not a simple task in itself. It too has many so-called exceptions, which are also commonly found in the wild.

Most of this answer post contains info that is related in someway to these two problems: figuring out what the subject's grammatical number is, and figuring out what the corresponding verb's grammatical number should be. A big assist in figuring out those two problems is a native English speaker's "ear". That ear has learned the grammatical rules over many years, or decades, and basically, it intuitively understands and knows English grammar better than the person that is attached to that ear. Anyhow . . .

Types of concord: There's three main kinds of concord, also commonly known as kinds of agreement: formal/grammatical concord, notional concord, proximity concord. Sometimes other terms are used, such as the principle of proximity (or as blind agreement or attraction). You can probably get a good feel for their definitions from your favorite grammar sources. And hopefully, when I use these terms later on, the surrounding context will make their meanings clear.

The context style: Since the OP's two examples are questions in the form of interrogative clauses, they would probably be more likely used in an informal context, that is, printed in fiction or narrative prose, or spoken in a casual or neutral style, than be printed in a very formal context like a dry textbook. (Though, textbooks are supposedly using more of the techniques found in narrative prose, so as to not bore the students to sleep.) So, let's keep that in mind: the context in which the OP's example sentence would be used is more likely to be informal to neutral in style.

Default number and person: It's probably practical to consider that the default number and person for subject-verb agreement of a clause is singular and 3rd person. There would usually have to be some reason or reasons for it to be something else; and often there are such reasons.

Type of clauses: There are many different types of clauses, and the clause type of the matrix clause and the construction types of its clausal dependents often have strong influences on subject-verb agreement (such as copular-like main clauses, which often are acceptable with singular verbs even when the subject is plural in number, and then there's pseudo-cleft constructions which have a default of singular verb).

But grammar usage books seem to mostly use, or only use, declarative non-copular clauses when they do most of their discussion on subject-verb agreement. And then, generally they use clauses with prototypical subject-verb-complement order, which often means that interrogative clauses usually aren't used as examples. (Reader, remember that the OP's examples are closed interrogative clauses.)

Many kinds of overrides: There can be overrides of formal concord, and those overrides can occur at clause level, and there also can be respecification of the NP when the NP is a measure phrase. A discussion and examples of overrides are found in CGEL pages 354, 504-7.

An example of a measure override is:

• [14.ii] [Twenty dollars] seems a ridiculous amount to pay to go to the movies. -- (CGEL page 504)

The subject is a plural NP which is taking a singular verb. The speaker is interpreting that subject NP, which is a measure phrase, as being notionally singular. Since the verb is singular, that means that notional concord has overridden formal concord.

"And" coordination of NPs: Sometimes these will take singular verb. CGEL page 507:

[26]

• i. [Eggs and bacon] is/are my favourite breakfast.

• ii. [The hammer and sickle] was/were flying over the Kremlin.

Note that only the singular verb is acceptable in [i-ii], while both singular and plural verb is acceptable in [iii].

Coordination of distributive NPs: Usually they take singular verb. CGEL, page 508:

Finally, coordinations of NPs containing distributive each or every take singular verbs:

[28]

• i. [Each dog and each cat] has/have to be registered.

• ii. [Every complaint and every suggestion] was/were thoroughly investigated.

Though, there might be some disagreement on this, such as with "every" where MWCDEU finds both singular and also some plural usage: page 318,

• Every single word and meaning of great ancient writers like Geoffrey Chaucer were recorded in the OED --Robert Burchfield, U.S. News & World Report, 11 Aug. 1986

"Or" coordination of NPs: The resolution rules will only give fully acceptable results when the NPs are either all singular or all plural. If the NPs are mixed, then many speakers will find any subject-verb agreement to be doubtful. CGEL page 508-9:

[31.ii ] [(Either) Mary or the twins] ?is/?are sure to go. -- [both are of questionable grammaticality]

[31.iii ] [(Either) the twins or Mary] ?is/?are sure to go. -- [both are of questionable grammaticality]

and CGEL explains:

Usage manuals generally invoke the principle of proximity, saying that the verb should agree with the nearest coordinate. This rule would select are in [ii] and is in [iii]. In practice, however, many speakers tend to feel uncomfortable with both forms and will typically find ways of avoiding the conflict, e.g. by using a modal auxiliary, which has no agreement properties: (Either) Mary or the twins will be sure to go.

In a way, this shows that proximity concord can influence the acceptability of a sentence to a speaker (and to style manuals).

The power of "one": When "one" is the head of a partitive, the presence of "one" can often attract a singular verb, and it can do that even when the semantics call for a plural verb. CGEL page 506 (and especially notice version [22.ii ]):

. . . The relativized element in these examples is object. Where it is the subject that is relativized, the expectation would be that the number of the verb would be determined by the antecedent, giving a plural verb in Type I, and a singular in Type II. In practice, however, singular verbs are often found as alternants of plurals in Type I:

[22]

• i. He's [one of those people who always want to have the last word]. -- (Type I )

• ii. He's [one of those people who always wants to have the last word]. -- (Type I )

• iii. He's [one of her colleagues who is always ready to criticize her]. -- (Type II )

Examples [i] and [iii] follow the ordinary rules, but [ii] involves a singular override. It can presumably be attributed to the salience within the whole structure of one and to the influence of the Type II structure (it is in effect a blend between Types I and II ). But it cannot be regarded as a semantically motivated override: semantically the relative clause modifies people. This singular override is most common when the relative clause follows those or those + noun.

Coordination of clauses: When the subject is an and-coordination of clauses, then usually subject-verb agreement will be singular, but sometimes it can be plural. CGEL, page 508:

And-coordination of clauses

Subjects with the form of an and-coordination of clauses generally take singular verbs:

[29]

• i. [That the form was submitted on the very last day and that the project had not been properly costed] suggests that the application was prepared in a rush.

• ii. [How the dog escaped and where it went] remains a mystery.

It is nevertheless possible to have a plural verb when the predicate treats the coordinates as expressing separate facts, questions, or the like:

[30]

• i. [That the form was submitted on the very last day and that the project had not been proper costed] are two very strong indications that the application was prepared in a rush.

• ii. [How the dog escaped and where it went] are questions we may never be able to answer.

Notice how the latter two examples, which use plural verbs, also have plural predicative complements (PCs).

The above and-coordinations were of finite clauses, which will usually take singular agreement, and that is also usually true when the clauses are infinitival clauses. When the clauses are -ing clauses, and because an -ing clause can often syntactically behave somewhat as a noun phrase, an and-coordination of them can often be even less clear as to subject-verb agreement, in that the coordination can sometimes be functioning more similar to a coordination of NPs than of finite clauses.

IN SUMMARY: There are many other types of cases where subject-verb agreement is not so straightforward or as simple as traditional grammar rules can lead a person to think. But I think you've got the picture, and if you're interested in seeing more of these cases, then a vetted grammar source, such as a reference grammar like CGEL, would be a good place to start.

THE ANSWER: And now, let's see if we can put that above rambling to good use.

First of all, let's look at some declarative clauses, which use the prototypical subject-verb-complement order, that correspond to the OP's examples:

1. The number and address [ is / are ] valid. -- (corresponds to OP's examples)

2. The number and the address [ is / are ] valid. -- (has an extra "the")

The speaker might prefer to use version #3 with "is" if they are seeing "the number and address" as being singular in concept, where "number" and "address" are closely tied to each other as though they make up one entity (e.g. "pencil and paper"). That is, the speaker is notionally seeing the subject as being singular for subject-verb agreement. And consistent with that, the speaker would probably use only one determiner ("the") for that coordination of "number" and "address".

But if the speaker is seeing the "number" and "address" as two separate things, where the coordination of them two would be plural in concept, then the speaker might prefer to use version #4 with "are". That is, the speaker is notionally seeing the subject as being plural for subject-verb agreement, which would also support the formal concord of interpreting that subject as plural. And consistent with that interpretation, the speaker would probably also use a pair of determiners (two instances of "the") for that coordination of singular nouns "number" and "address".

Notice that the OP's examples are using only one determiner ("the") in the subject NP (similar to #3), not two determiners. That will help support the speaker's notional concord of singular.

The above discussion was not laying out "rules", like those often taught in traditional grammars, but rather tendencies and preferences, sorta more like guidelines.

Notice that both #3 and #4 are in the form of copular clauses. Sentences in the form of copular clauses often accept singular verbs, even when that would seem to contradict formal concord. You'll often see them use the template "X is Y", even when the subject is formally a plural NP.

When prototypical declarative clauses are used, the subject is read or heard first, and if the subject is interpreted by the reader or hearer to be plural, then they will usually expect to hear a plural verb (for the matrix verb). And accordingly, if the subject is interpreted to be singular, then a singular verb is expected. For example:

• [That ten days we spent in Florida] was fantastic. -- (CGEL page 354, [71.i])
• [Those ten days we spent in Florida] were fantastic. -- (CGEL page 354, [72.i])

In the first version, the singular subject is signaled to the reader by the use of the singular "That", and that supports the singular verb "was". And correspondingly, in the second version the plural subject is signaled to the reader by the use of the plural "Those", and that supports the plural verb "were".

But for a sentence that is a question which is in the form of a closed interrogative clause, like that of the OP's examples, the verb happens to come first, and we'll naturally use the number of the verb when we then parse the following NP, for we'll be naturally assuming that the closest NP is the subject. If that NP doesn't match in number, then that will often be jarring. The OP's examples:

1. Is [the number] and address valid? -- (okay, the verb and the closest NP match in number)
2. Are [the number] and address valid? -- (awkward, the verb and the closest NP don't match in number)

The OP's version #1 with its singular "is" matches the number of its nearest NP, which is proximity concord in action, and that singular is consistent with the writer's intended singular notional concord. (And also consistent with the singular interpretation is the use of only one determiner "the" in the subject NP.)

But the OP's version #2 with its plural "are" will be jarring to many hearers due to the mismatch with the number of its closest NP. (In other words, version #2 violates proximity concord.)

Notice how there would be no problem for version #2 if the closest NP had been plural, for instance:

• Are [the boys] and girls going to show up at the same time?

In that last example, all three concords (formal, notional, proximity) match in having a plural interpretation.

CONCLUSION: A speaker or writer constantly juggles the three main types of concord--formal, notional, proximity--as they construct their sentences, and when the three main types of concord align up with each other, together supporting the same number agreement, then the resulting sentences will usually sound okay. But when there is a mismatch, then sometimes that can cause an awkward sentence which might not be acceptable to many native English speakers.

Styles and registers will weight the different concords differently. For instance, there are some very formal registers that weight formal concord very heavily, at the expense of notional and proximity concord, which when combined with their adherence to a style guide that uses rules based on traditional grammar's misunderstandings of how English grammar works (rules that cause an overuse of "whom", a rule for not splitting infinitives, a rule for not stranding prepositions, etc.), will often end up producing sentences that are unacceptable to many or most native English speakers when those sentences are spoken in an informal or neutral environment.

These three main concords are often in competition with each other, and grammatical discussions of that competition done in grammar usage books will usually use examples that are in the form of declarative clauses with prototypical subject-verb-complement order. In these types of examples, proximity concord will have a certain range of weight. But when the example sentence is in the form of a closed interrogative main clause, such as in the OP's examples where the verb comes first before the subject NP, then it seems that proximity concord has even more of an influence on subject-verb agreement than it would have in a corresponding declarative clause version. And that mismatch with proximity concord is a big factor in making the OP's 2nd version ("are") sound so awkward.

NOTE: CGEL is the 2002 reference grammar by Huddleston and Pullum et al., The Cambridge Grammar of the English Language.

NOTE: MWCDEU is the 2002 Merriam-Webster's Concise Dictionary of English Usage.

2Thank you for this Answer. I found it very informative, and explained things that I could not explain but knew from Having a Native English Speaking Ear. – Malachi – 2015-01-08T18:19:53.413

1+1 This is excellent. I happen to disagree on one point, viz., I find "Are the number and address valid?" perfectly acceptable colloquially: it amounts to "Did you check both of them?" But that's trivial. I particularly value the paragraph on "Context Style", which is something that reely reely reely needs to be hammered on around here. – StoneyB on hiatus – 2015-01-10T15:08:51.507

1

This needs further work.

If you are a native speaker, go with your gut feeling. Non-native speakers should probably stick to using are except in a phrase with the contracted there + is or here + is, as in Here's the name and number. If you're a native speaker and doubt yourself, I offer the following. Your mileage may vary.

If you consider or are referring to number and address as a single unit, you can use is. For example,

Here is my name and my number. Call me.

Some native speakers may not find this natural. It may be in part due to dialectal differences.

To identify the two even more as a single unit, use one determiner for both:

Here is my name and number.

(This is exactly what is done with the number and address.)

Since both number and address and name amd number are often together on an order form, application, or other such context, they can definitely be seen as a unit and take is. But they do not have to. However, some native speakers might find that is sounds strange or that are sounds strange. This probably largely depends on personal preference, dialect, and perhaps to their attitude toward the matter.

But even if two nouns are seen as a unit and described by one determiner, they can still take are.

'Here are my mom and dad.' This would probably be more common than 'Here is my mom and dad.'

If you use here's or there's (contracted forms with the singular is), you can say

'Here's or there's mom and dad.'

This usage is common in spoken English. In written English, the tendency is to be more 'careful' regarding grammatical agreement and to write