The relative pronouns

A relative pronoun is a pronoun that marks a relative clause. An example is the word which in the sentence «This is the house which Jack built.» Here the relative pronoun which introduces the relative clause, which modifies the noun house. A relative pronoun plays the role of a noun phrase within that clause. For example, in the relative clause «which Jack built,» «which» is a pronoun functioning as the object of the verb «built.»

The element in the main clause that the relative pronoun in the relative clause stands for (house in the above example) is the antecedent of that pronoun. In most cases the antecedent is a nominal (noun or noun phrase), though the pronoun can also refer to a whole proposition, as in «The train was late, which annoyed me greatly», where the antecedent of the relative pronoun which is the clause «The train was late» (the thing that annoyed me was the fact of the train’s being late).

In a free relative clause, a relative pronoun has no antecedent: the relative clause itself plays the role of the co-referring element in the main clause. For example, in «I like what you did», what is a relative pronoun, but without an antecedent. The clause what you did itself plays the role of a nominal (the object of like) in the main clause. A relative pronoun used this way is sometimes called a fused relative pronoun, since the antecedent appears fused into the pronoun (what in this example can be regarded as a fusion of that which).

Even within languages that have relative pronouns, not all relative clauses contain relative pronouns. For example, in the English sentence «The man you saw yesterday was my uncle», the relative clause you saw yesterday contains no relative pronoun. It can be said to have a gap, or zero, in the position of the object of the verb saw.

Other arguments can be relativised using relative pronouns:

Hunter is the boy who helped Jessica.

Hunter is the boy whom Jessica gave a gift to.

Jack built the house in which I now live. (Similarly with prepositions and prepositional phrases in general, for example, These are the walls between which Jack ran.)

Jack is the boy whose friend built my house.

In English, different pronouns are sometimes used if the antecedent is a human being, as opposed to a non-human or an inanimate object (as in who vs. that).

(1) This is a bank. This bank accepted my identification.
(2) She is a bank teller. She helped us open an account.

With the relative pronouns, sentences (1) and (2) would read like this:

(3) This is the bank that accepted my identification.
(4) She is the bank teller who helped us open an account.

In sentences (3) and (4), the words that and who are the relative pronouns. The word that is used because the bank is a thing; the word who is used because the teller is a person. Alternatively, which is often used in defining (or restrictive) relative clauses in either case. For details see English relative clauses.

• Michael Cysouw (2011). «Quantitative explorations of the worldwide distribution of rare characteristics, or: the exceptionality of northwestern European languages». In Horst J. Simon & Heike Wiese (ed.). Expecting the unexpected: Exceptions in grammar. Berlin: Mouton de Gruyter. p. 411-431.
• Kordić 1999, pp. 36–37.
• Kordić 1999, pp. 16–19.
• Dayal, Veneeta (1996). «Locality in WH Quantification: Questions and Relative Clauses in Hindi». Studies in Linguistics and Philosophy. 62. doi:10.1007/978-94-011-4808-5_5.

English also uses free relative clauses, which have no antecedent and can be formed with the pronouns such as what («I like what you’ve done»), and who and whoever.

• The basic relative pronouns are considered to be who, which and that; but see an alternative analysis of that below.
• When that is used in a restrictive relative clause and it is not the subject of the relative clause, it may be omitted entirely. For example: («The dentist that I saw» or «The dentist that I spoke to») may be rendered simply («The dentist I saw» or «The dentist I spoke to»). But any relative pronoun when used in a non-restrictive relative clause must not be omitted («My dentist, whom I saw», or «My dentist, who spoke to me»); nor when its preposition is fronted («The dentist to whom I spoke»); nor when its antecedent is the subject of the relative clause («The dentist that saw me”, or “The dentist who saw me»).
• A relative clause whose antecedent is a whole proposition—that is, a matter (or person or thing) to be dealt with—is formed with which («The cake was burnt, which made me angry»); here which refers to the whole circumstance of the cake’s being burnt.

The words used as relative pronouns have other uses in English grammar: that can be a demonstrative or a conjunction, while which, what, who, whom and whose can be interrogatives. For other uses of whoever etc., see -ever.

The choice of relative pronoun typically depends on whether the antecedent is human or non-human: for example, who and its derivatives (whom, whoever, etc.—apart from whose) are generally restricted to human antecedents, while which and what and their derivatives refer in most cases to things, including animals.

English also makes the distinction between human vs. thing in personal pronouns (he, she vs. it) and certain other pronouns (such as someone, somebody vs. something); but some particular things—such a navy ships and marine vessels—are described with female pronouns, and pets and other animals are frequently addressed in terms of their gender or their (anthropomorphic) ‘personhood’. Typically, it is when these things-as-human become antecedents to relative clauses that their relative pronouns tend to revert to that or which—for things—rather than taking the regular who, whom, etc., for human referents. See Gender in English.

Restrictive or non-restrictive relative clauses

(1) The builder, who erects very fine houses, will make a large profit. (non-restrictive)
(2) The builder who erects very fine houses will make a large profit. (restrictive)

The first expression refers to an individual builder (and it implies that we know, or know of, the builder, which is the referent). It says that he builds «very fine» houses, and that he will make a large profit. It conveys these meanings by deploying a non-restrictive relative clause and three short intonation curves, marked-off by commas. The second expression refers not to a single builder but to a certain category, also called a set, of builders who meet a particular qualification, or distinguishing property: the one explained by the restrictive relative clause. Now the sentence means: it is the builder who builds «very fine» houses who will make a large profit. It conveys this very different meaning by providing a restrictive relative clause and only one intonation curve, and no commas. Commas are, however, often used erroneously, probably because the rule is taught based on logic and most people are not aware that they can trust their ear in deciding whether to use a comma. (English uses commas in some other cases based on grammar, not prosody.)

Thus, in speaking or writing English prose, a restrictive rather than non-restrictive meaning (or vice versa), requires the correct syntax by choosing the appropriate relative clause (i.e., restrictive or non-restrictive) and the appropriate intonation and punctuation.

To determine whether a relative clause is restrictive or non-restrictive a simple test can be applied. If the basic meaning of the sentence (the thought) is not changed by removing the relative clause, the relative clause is not essential to the basic thought and is non-restrictive. Alternatively, if the essential meaning of the thought is disturbed, the relative clause is restrictive.

Restrictive relative clauses are also called integrated relative clauses, defining relative clauses, or identifying relative clauses. Conversely, non-restrictive relative clauses are called supplementary, appositive, non-defining, or non-identifying relative clauses.

Integrated clauses that are not restrictive

Although the term «restrictive» has become established as joined with integrated clauses, there are integrated clauses that do not necessarily express a distinguishing property of the referent. Such a (so-called) restrictive clause, actually a non-restrictive clause, is so completely integrated into the narrative and the intonation of the main sentence that it falsely appears to be restrictive.

These examples of integrated relative clauses in that sense are not truly restrictive:

When the «restrictive» relative clause is removed from either of the above sentences, the antecedent («the father» and «the clergyman») is not placed in question. In the first example, for instance, there is no suggestion that the narrator has two fathers because the relative clause does not express a distinguishing property of the subject. Instead, the relative clause is integrated but is not truly restrictive.

That or which for non-human antecedents

The distinction between the relative pronouns that and which to introduce restrictive relative clauses with non-human antecedents is a frequent point of dispute.

For clarity, we can look at the case of non-human antecedents using the previous example:

(1) The building company, which erects very fine houses, will make a large profit. (non-restrictive)
(2) The building company that (or which) erects very fine houses will make a large profit. (restrictive)

Equivalently, the two cases would be applied where the statements are logically:

(1) «which», non-restrictive: (The building company erects very fine houses) AND (The building company will make a large profit).
(2) «that», restrictive: (The building company erects very fine houses) IMPLIES (The building company will make a large profit).

Zero relative pronoun

English, unlike other West Germanic languages, has a zero relative pronoun (denoted below as Ø)—that is, the relative pronoun is implied and not explicitly written or spoken; it is «unvoiced». This measure is used in restrictive relative clauses (only) as an alternative to voicing that, which or who, whom, etc. in these clauses:

Jack built the house that I was born in;
Jack built the house Ø I was born in;

He is the person whom I saw;
He is the person Ø I saw.

In other words, the word «that» (or «who» or «which», etc.) as a relative clause connector is optional when it would not be the subject of the relative clause; even when it would be required in other languages.

The zero relative pronoun cannot be the subject of the verb in the relative clause; that is, that or who, etc., cannot be omitted (unvoiced) if the zero pronoun would be a subject. Thus one may say:

Jack built the house that sits on the hill;
She is the one who encouraged me;

but never (except in some varieties of colloquial English):

*Jack built the house Ø sits on the hill;
*She is the one Ø encouraged me.

Relative clauses headed by zeros are frequently called contact clauses in TEFL contexts, and may also be called «zero clauses».

(If that is analyzed as a complementizer rather than as a relative pronoun the above sentences would be represented differently: Jack built the house that I was born in Ø; Jack built the house I was born in Ø; He is the person I saw Ø. (see § That as relativizer instead of relative pronoun)

‘What’ relative pronoun

Some varieties of English use what as a relative pronoun. For example, in Guardians of the Galaxy Vol. 2, a Ravager says, «For it is a name what strikes fear into the hearts of anyone what hears it.»

What as a relative pronoun appeared on the front-page of United Kingdom newspaper The Sun on 11 April 1992 in the headline «It’s The Sun Wot Won It.»

Standard Englishes proscribe the use of what as a relative pronoun, preferring who or that.

Relative pronoun as the object of a preposition

A relative pronoun often appears as the object of a preposition, both in restrictive and non-restrictive clauses, as in

«Jack is the boy with whom Jenny fell in love.»

«Yesterday, Jenny met Jack, for whom she no longer has any feelings.»

It is not unusual to place the preposition at the end of the relative clause, while the relative pronoun that it governs is placed at the beginning of the clause or omitted, so

«Jack is the boy that Jenny fell in love with.»

is also possible.
A preposition is never placed in front of the relative pronoun that, but preposition stranding is possible when there is an explicit that, or when the relative pronoun representing the object of the clause is omitted.
So

«Jack is the boy Jenny fell in love with.»

are possible but

* «Jack is the boy with that Jenny fell in love.»

* «Jack is the boy with who Jenny fell in love»

«Jack is the boy who Jenny fell in love with.»

especially in informal style.
Use of the objective case with a stranded preposition, as in

«Jack is the boy whom Jenny fell in love with.»

The word that, when used in the way described above, has been classified as a relative pronoun; however, according to some linguists it ought to be analyzed instead as a subordinating conjunction or relativizer. This is consistent with that used as a conjunction in (I said that I was tired), or implied in (I said I was tired).

There are some grammatical differences between that and the (other) relative pronouns: that is limited to restrictive relative clauses, and it cannot be preceded with a preposition. There are also similarities between the (purported) relative pronoun that and the ordinary conjunction that: the weak pronunciation is (almost invariably) used in both cases, and both of them are frequently omitted as implied.

What he did was clearly impossible.

Here «What he did» has the same sense as «that which he did», or «the thing that he did». Thus the noun phrase the thing and the relative pronoun that are ‘fused’ into what; and the resulting relative construction «What he did» functions as the subject of the verb was. Free relative constructions are inherently restrictive.

English has a number of «fusible» relative pronouns that initiate relative constructions, including what, whatever and whoever. But these pronouns introduce other clauses as well; what can introduce interrogative content clauses («I do not know what he did») and both whatever and whoever can introduce adverbials («Whatever he did, he does not deserve this»). See -ever.

Some non-finite clauses, including infinitive and participial clauses, may also function as relative clauses. These include:

• infinitive clauses containing an ‘explicit’ relative pronoun (argument)—generally, but not always, fronted with a preposition—that takes an antecedent to that ‘explicit’ argument: She is a woman whom to beat; He is the man on whom to rely. (The infinitive verbs are ‘to beat’ and ‘to rely’; the antecedents are ‘woman’ and ‘man’, respectively.)
• infinitive clauses presenting an ‘implied’ (and unvoiced) relative pronoun, or zero object argument, that takes an antecedent to that ‘implied’ argument: She is a woman to beat Ø; He is the man to rely on Ø.
• infinitive clauses modifying the subject of the infinitive verb: She is the person to save the company.
• present participle clauses having an unvoiced zero subject argument that takes an antecedent to the argument: The man Ø sitting on the bank was fishing. (These clauses are the least likely to be recognized as relative clauses.)
• past participle clauses having an unvoiced zero object argument that takes an antecedent to the argument: The body found Ø here yesterday has now been identified. (This is the «reduced object passive relative clause»; see Reduced relative clause § Non-finite types.

For further examples see Uses of English verb forms § Uses of nonfinite verbs.

Some adverbial clauses can function as relative clauses, including:

They were just towed across the Midway onto the bridle path, where they were just sitting there peacefully

In this case, removing the underlined resumptive pronoun results in an acceptable gapped relative clause:

They were just towed across the Midway onto the bridle path, where they were just sitting ___ peacefully

In other cases, the resumptive pronoun is used to work around a syntactic constraint:

In this example, the word it occurs as part of a wh-island. Attempting to extract it gives an unacceptable result:

*They have a billion dollars of inventory that they don’t know where ___ is.

a book that you wish the author was a terrific friend of yours

A bound relative clause, the type most often considered, qualifies an explicit element (usually a noun or noun phrase) appearing in the main clause, and refers back to that element by means of some explicit or implicit device within the relative clause.

Restrictive and non-restrictive

Bound relative clauses may or may not be restrictive. A restrictive relative clause is a relative clause that functions as a restrictive modifier. A non-restrictive relative clause is a relative clause that is not a restrictive relative clause. Whereas a non-restrictive or non-defining relative clause merely provides supplementary information, a restrictive or defining relative clause modifies the meaning of its head word (restricts its possible referent). For example:

In speaking, it is natural to make slight pauses around non-restrictive clauses, and in English this is shown in writing by commas (as in the examples). However, many languages distinguish the two types of relative clauses in this way only in speaking, not in writing. Another difference in English is that only restrictive relative clauses may be introduced with that or use the «zero» relative pronoun (see English relative clauses for details).

A non-restrictive relative clause may have a whole sentence as its antecedent rather than a specific noun phrase; for example:

Here, which refers not to the bed or the cat but to the entire proposition expressed in the main clause, namely the situation of the cat being allowed on the bed.

Finite and non-finite

Relative clauses may be either finite clauses (as in the examples above) or non-finite clauses. An example of a non-finite relative clause in English is the infinitive clause on whom to rely, in the sentence «She is the person on whom to rely».

Languages differ in many ways in how relative clauses are expressed:

• How the two clauses are joined together.
• Where the embedded clause is placed relative to the head noun (in the process indicating which noun phrase in the main clause is modified).

• The clauses are joined by the complementizer «that».
• The embedded clause is placed after the head noun «the person».

• Gap strategy or gapped relative clause
• Relative pronoun
• Pronoun retention
• Nonreduction

Gapped relative clause

Only a very small number of languages, of which the best known is Yoruba, have pronoun retention as their sole grammatical type of relative clause.

Dialects of some European languages, such as Italian, do use the nonreduction type in forms that could be glossed in English as «The person just passed us by, she introduced me to the chancellor here.»

In general, however, nonreduction is restricted to verb-final languages, though it is more common among those that are head-marking.

Strategies for joining the relative clause to the main clause

• Use of an indeclinable particle (specifically, a relativizer) inserted into the sentence, placed next to the modified noun; the embedded clause is likewise inserted into the appropriate position, typically placed on the other side of the complementizer. This strategy is very common and arguably occurs in English with the word that («the woman that I saw»), though this interpretation of «that» as something other than a relative pronoun is controversial (see below). In the modern varieties of Arabic (using illi placed after the modified noun); in Chinese (using de placed before the modified noun).
• Directly inserting the embedded clause in the matrix clause at the appropriate position, with no word used to join them. This is common, for example, in English (cf. «The person I saw yesterday went home»), and is used in Classical Arabic in relative clauses that modify indefinite nouns.
• By nominalizing the relative clause (e.g. converting it to a participial construction). Generally, no relative pronoun or complementizer is used. This occurs, for example, in reduced relative clauses in English (e.g. «The person seen by me yesterday went home» or «The person planning to go home soon is my friend»). Formal German makes common use of such participial relative clauses, which can become extremely long. This is also the normal strategy in Turkish, which has sentences equivalent to «I ate the potato of Hasan’s giving to Sina» (in place of «I ate the potato that Hasan gave to Sina»). This can be viewed as a situation in which the «complementizer» is attached to the verb of the embedded clause (e.g. in English, «-ing» or «-ed» can be viewed as a type of complementizer).

Position of the head noun with respect to the relative clause

Various possibilities for ordering are:

• Relative clause preceding the head noun, as in Turkish, Japanese, or Chinese.
• Adjoined relative clause. These languages have the relative clause completely outside the main clause, and use a correlative structure to link the two. These languages also have nonreduced relative clauses. Hindi, the most well-known such language, has a structure similar to «Which person I saw yesterday, that person went home» or (without fronting of the relativized noun in the relative clause) «I saw which person yesterday, that person went home». Another example is Warlpiri, which constructs relative clauses of a form similar to «I saw the man yesterday, which he was going home». However, it is sometimes said these languages have no relative clauses at all, since the sentences of this form can equally well translate as «I saw the man who was going home yesterday» or «I saw the man yesterday when/while he was going home».

Ergative–absolutive languages have a similar hierarchy:

This order is called the accessibility hierarchy. If a language can relativize positions lower in the accessibility hierarchy, it can always relativize positions higher up, but not vice versa. For example, Malagasy can relativize only subject and Chukchi only absolutive arguments, whilst Basque can relativize absolutives, ergatives and indirect objects, but not obliques or genitives or objects of comparatives. Similar hierarchies have been proposed in other circumstances, e.g. for pronominal reflexes.

English can relativize all positions in the hierarchy. Here are some examples of the NP and relative clause usage from English:

Some other examples:

Languages that cannot relativize directly on noun phrases low in the accessibility hierarchy can sometimes use alternative voices to «raise» the relevant noun phrase so that it can be relativized. The most common example is the use of applicative voices to relativize obliques, but in such languages as Chukchi antipassives are used to raise ergative arguments to absolutive.

For example, a language that can relativize only subjects could say this:

These languages might form an equivalent sentence by passivization:

These passivized sentences get progressively more ungrammatical in English as they move down the accessibility hierarchy; the last two, in particular, are so ungrammatical as to be almost unparsable by English speakers. But languages with severe restrictions on which roles can be relativized are precisely those that can passivize almost any position, and hence the last two sentences would be normal in those languages.

Modern grammars may use the accessibility hierarchy to order productions—e.g. in Head-Driven Phrase Structure Grammar the hierarchy corresponds to the order of elements on the subcat list, and interacts with other principles in explanations of binding facts. The hierarchy also figures in
Lexical Functional Grammar, where it is known as Syntactic Rank or the Relational Hierarchy.