In linguistics, a copula (plural: copulae) is a word used to link the subject of a sentence with a predicate (a subject complement). The word copula derives from the Latin noun for a link or tie that connects two different things.[1]
A copula is often a verb or a verb-like word, though this is not universally the case.[2] A verb that is a copula is sometimes called a copulative or copular verb. In English primary education grammar courses, a copula is often called a linking verb. Other copulae show more resemblances to pronouns. This is the case for Chinese and Guarani, for instance. In highly synthetic languages, copulae are often suffixes attached to a noun that may still behave otherwise like ordinary verbs, for example -u- in Eskimo languages. In some other languages, such as Beja and Ket, the copula takes the form of suffixes that attach to a noun but are distinct from the person agreement markers used on predicative verbs[3]. This phenomenon is known as nonverbal person agreement or nonverbal subject agreement and the relevant markers are always established as deriving from cliticised independent pronouns.
The term copula is generally used to refer to the main copular verb(s) in a language. In the case of English, this is the verb to be. The term can also be used to refer to some other verbs in the language which fulfill similar functions. Other English copulae include to become, to get, to feel, and to seem. Other verbs can have secondary uses as copulative verbs. In the following example, the past tense of the verb to fall is used as a copula: "The zebra fell victim to the lion." These extra copulae are sometimes called "semi-copulae" or "pseudo-copulae".
Most languages have one main copula, but some languages, like Spanish or Thai, have more than one, and some have none.
Contents |
Copulae have multiple uses.
Several uses of the copula can be categorized:
Note that the auxiliary verb function derives from the copular function; and, depending on one's point of view, one can still interpret the verb as a copula and the following verbal form as being adjectival. Abelard in his Dialectica made an argument against the idea that the copula can express existence based on a reductio ad absurdum.[4]
Along with copular sentences where the canonical order of predication is displayed - that is, the subject precedes the predicate - as in "a picture of the wall is the cause of the riot" there can also be "inverse copular sentences" where this order is mirrored as in "the cause of the riot is a picture of the wall" (cf. Everaert et al. 2006). Although these two sentences are superficially very similar it can be shown that they embody very different properties. So, for example it is possible to form a sentence like "which riot do you think that a picture of the wall is the cause of" but not "which wall do you think that the cause of the riot was a picture of". The distinction between these two types of sentences, technically referred to as "canonical" vs. inverse copular sentences, respectively - and the unified theory of copular sentences associated to it - has been proven to be valid across-languages and has led to some refinement of the theory of clause structure. In particular it challenges one of the major dogmas of the theory of clause structure, i.e. that the two basic constituents of a sentence Noun Phrase and Verb Phrase are associated to the logical/grammatical functions of subject and predicate (cf. phrase structure rules and sentence (linguistics)).
In fact, copular sentences show that this axiom is not adequate on empirical grounds since the Noun Phrase that cooccurs with the Verb Phrase in a copular sentence can be the predicate and the subject be contained in the Verb Phrase. Interestingly, it has been suggested that inverse copular sentences appear to play a sharp role in setting the pro-drop parameter. In Italian, for example in sentences of the type Noun Phrase verb Noun Phrase, the verb generally agrees with the Noun Phrase on the left with one exception: inverse copular sentences. One can construe minimal pairs like the cause of the riot is/*are these pictures of the wall vs. la causa della rivolta sono/*è queste foto del muro: the two sentences are one the gloss of the other with only one difference: the copula is singular in Italian and plural in English. If one does not want to give up the idea that agreement is on the left, then the only option is to assume that pro occurs between the copula and the Noun Phrase on the left. That pro can occur as a predicate must be in fact independently assumed to assign a proper structure to sentences like sono io (am I: "it's me") which can by no means be considered a transformation of *io sono, which has no meaning.
In informal speech of English, the copula may be dropped. This is a feature of African American Vernacular English but is also used by a variety of English speakers in informal contexts.[5] An example: "Where you at? We at the store."
The double copula is the use of two successive copulae when only one is necessary. Use of the double copula is one of the disputes in English grammar.
As in most Indo-European languages, the English copula is the most irregular verb because of constant use. Most English verbs (traditionally known as "weak verbs") have just four separate forms, e.g. "start", "starts", "starting", "started". A sizable minority (traditionally known as "strong verbs") have five separate forms, e.g. "begin", "begins", "beginning", "began", "begun". "To be" is a very special case in having eight forms: "be", "am", "is", "are", "being", "was", "were", "been". Historically it had even more, including "art", "wast", "wert", and, occasionally, "beest" as a subjunctive.
From one perspective, the copula always relates two things as subsets. Take the following examples:
Example 1 includes John in the set of all doctors. Example 2 includes John and Mary both in the set of all doctors. Example 3 includes the set of doctors in the set of those who are educated.
Example 4 is different. Example 4 includes Mary's state at the time of utterance in the set of states consistent with running. Example 5 then includes the set of states consistent with running in the set of states consistent with fun.
A copula and an action verb can generally be differentiated by trying to substitute the verb with a form of "to seem" or "to be".
Note that this approach falters, in part, with the verb "to appear". In the sentence "Sam appears to be happy", "appears" is a copula. Yet, "seems" but not "is" can be substituted: "Sam is to be happy" means something else entirely.
With certain writing styles, authors and editors often consider an action verb to be a stronger device than a copula.[6][7]
Taking this to the extreme, some writers have championed a movement of copula-free writing touted as E-Prime, forcing replacement with active verbs or even sentence constructions.
In Indo-European languages, the words meaning "to be" often sound similar to each other. Due to the high frequency of their use, their inflection retains a considerable degree of similarity in some cases. Thus, for example, the English form is is an apparent cognate of German ist, Latin est and Russian jest', even though the Germanic, Italic, and Slavic language groups split at least three thousand years ago. The origins of the Indo-European copulae can be traced back to four different stems *es- (*h1es-), *sta- (*steh2-), *wes- and *bhu- (*bʰuH-) in most Indo-European languages.
Just like in English, the verb "to be" (qopna) is irregular in Georgian (a Kartvelian language); different verb roots are employed in different tenses. The roots -ar-, -kn-, -qav-, and -qop- (past participle) are used in the present tense, future tense, past tense and the perfective tenses respectively. Examples:
Masc'avlebeli var. | "I am a teacher." | |
Masc'avlebeli viknebi. | "I will be a teacher." | |
Masc'avlebeli viqavi. | "I was a teacher." | |
Masc'avlebeli vqopilvar. | "I have been a teacher." | |
Masc'avlebeli vqopiliqavi. | "I had been a teacher." |
Note that in the last two examples (perfect and pluperfect) two roots are used in one verb compound. In the perfective tense, the root qop (which is the expected root for the perfective tense) is followed by the root ar, which is the root for the present tense. In the pluperfective tense, again, the root qop is followed by the past tense root qav. This formation is very similar to German (an Indo-European language), where the perfective and the pluperfective are expressed in the following way:
Ich bin Lehrer gewesen. | "I have been a teacher", literally "I am teacher been." | |
Ich war Lehrer gewesen. | "I had been a teacher", literally "I was teacher been." |
Here, gewesen is the past participle of sein ("to be") in German. In both examples, just like in Georgian, this participle is used together with the present and the past forms of the verb in order to conjugate for the perfect and the pluperfect aspects.
In some languages, copula omission occurs within a particular grammatical context. For example, speakers of Russian, Hungarian, Hebrew, Arabic and Quechuan languages consistently drop the copula in present tense: Russian: я — человек, ya — chelovek "I (am) a human"; Hungarian: ő ember, "he (is) a human"; Hebrew: אני בן-אדם "I (am a) human", Arabic: أنا (أكون) إنسان ənæ (əku:n) ensæn "I am human"; Southern Quechua: payqa runam "s/he (is) a human". This usage is known generically as the zero copula. Note that in other tenses (sometimes in other persons besides third singular) the copula usually reappears.
In Quechua —Southern Quechua used for the examples—, zero copula is restricted to present tense in 3rd person singular only (kan): Payqa runam — "(s)he is a human"; but: (paykuna) runakunam kanku "(they) are human".
In Hungarian, zero copula is restricted to present tense in 3rd person singular and plural: Ő ember/Ők emberek — "s/he is a human"/"they are humans"; but: (én) ember vagyok "I am a human", (te) ember vagy "you are a human", mi emberek vagyunk "we are humans", (ti) emberek vagytok "you (all) are humans". The copula also reappears for stating locations: az emberek a házban vannak, "the people are in the house," and for stating time: hat óra van, "it is six o'clock." However, the copula may optionally get omitted in colloquial language: hat óra (van), "it is six o'clock."
Hungarian uses a copula to say Itt van Róbert "Bob is here" (and this not only with regard to third person singular/plural), but not to say Róbert öreg "Bob is old". This is to relate a subject to a more temporary condition/state taking place in space (very often in the sense of Lojban zvati: la rabyrt. zvati ne'i le zdani "Robert is in the house").
Further restrictions may apply before omission is permitted. For example in the Irish language, is, the present tense of the copula, may be omitted when the predicate is a noun. Ba the past/conditional cannot be deleted. If the present copula is omitted, the following pronoun é, í, iad preceding the noun is omitted as well.
Romance copulae usually consist of two different verbs that can be translated as "to be", the main one from the Latin esse (derived from *es-), and a secondary one from stare (derived from *sta-) . This is found in Spanish, Catalan and Portuguese and to a lesser extent in Italian. The difference is that the first usually refers to essential characteristics, whilst the second refers to states and situations, e.g. "Bob is old" versus "Bob is well". (Note that the English words just used, "essential" and "state", are also cognate with the Latin infinitives esse and stare.) In Spanish, the high degree of verbal inflection, plus the existence of two copulae (ser and estar), means that there are 105 separate forms to express the eight in English, and the one in Chinese.
Copula | Language | ||
---|---|---|---|
Italian | Spanish | English | |
Sum-derived | Bob è vecchio. | Bob es viejo. | "Bob is old." |
Sto-derived | Bob sta bene. | Bob está bien. | "Bob is well." |
In some cases, the verb itself changes the meaning of the adjective/sentence. The following examples are from Portuguese:
Copula | Example 1 | Example 2 | ||
---|---|---|---|---|
Portuguese | English | Portuguese | English | |
Sum-derived | O Bob é bom. | "Bob is good." | O Bob é parvo. | "Bob is foolish." |
Sto-derived | O Bob está bom. | "Bob is feeling good." | O Bob está parvo. | "Bob is acting/being silly." |
In certain languages there are not only two copulae but the syntax is also changed when one is distinguishing between states or situation and essential characteristics. For example, in Irish and Scottish Gaelic, describing the subject's state or situation typically uses the normal VSO ordering with the verb bí. The copula is, which is used to state essential characteristics or equivalences, requires a change in word order so that the subject does not immediately follow the copula (see Irish syntax).
In Slavic languages, a similar distinction is made by putting a state in the instrumental case, while characteristics are in the nominative. This is used with all the copulae (e.g. "become" is normally used with the instrumental). It also allows the distinction to be made when the copula is omitted (zero copula) in East Slavic languages (in other Slavic languages the copula is not omitted).
Haitian Creole, a French-based creole language, has a reputation as being rather exotic linguistically when compared to French and the other Romance languages; and it lives up to this reputation with its copula system. It has three forms of the copula: se, ye, and the zero copula, no word at all (the position of which will be indicated with Ø, just for purposes of illustration).
Although no textual record exists of Haitian at its earliest stages of development from French, se is obviously derived from French c'est ([se]), which is the normal French contraction of ce (that) and the copula est (third-person singular of the present indicative of the verb être, ultimately from Latin sum). There appears to be no trace of Latin sto.
The derivation of ye is less obvious; but we can assume that the French source was il est ("he/it is"), which, in rapidly spoken French, is very commonly pronounced as y est (IPA [je]).
The use of a zero copula is unknown in French, and it is thought to be an innovation from the early days when Haitian was first developing as a Romance-based pidgin. Coincidentally, Latin also sometimes used a zero copula.
Which of se/ye/Ø is used in any given copula clause depends on complex syntactic factors that we can superficially summarize in the following four rules:
1. Use Ø (i.e., no word at all) in declarative sentences where the complement is an adjective phrase, prepositional phrase, or adverb phrase:
Li te Ø an Ayiti. | "She was in Haiti." | (she past-tense in Haiti) |
Liv-la Ø jon. | "The book is yellow." | (book-the yellow) |
Timoun-yo Ø lakay. | "The kids are [at] home." | (kids-the home) |
2. Use se when the complement is a noun phrase. But note that whereas other verbs come after any tense/mood/aspect particles (like pa to mark negation, or te to explicitly mark past tense, or ap to mark progressive aspect), se comes before any such particles:
Chal se ekriven. | "Charles is writer." | |
Chal se pa ekriven. | "Charles is not writer." cf. with the verb kouri ("run"): Chal pa kouri, not Chal kouri pa. | |
Chal, ki se ekriven, pa vini. | "Charles, who is writer, not come." |
3. Use se where French and English have a dummy "it" subject:
Se mwen! | "It's me!" French C'est moi! | |
Se pa fasil. | "It's not easy," colloquial French C'est pas facile. |
4. Finally, use the other copula form ye in situations where the sentence's syntax leaves the copula at the end of a phrase:
Kijan ou ye? | "How you are?" | |
Pou kimoun liv-la te ye? | "Whose book was it?" | (of who book-the past-tense is?) |
M pa konnen kimoun li ye. | "I don't know who he is." | (I not know who he is) |
Se yon ekriven Chal ye. | "Charles is a writer!" | (it's a writer Charles is; cf. French C'est un écrivain qu'il est.) |
The above is, however, only a simplified analysis.[8][9]
Japanese and Korean have copulae that would most often be translated as one of the so-called be-verbs of English, the two languages being grammatically very similar but with entirely different vocabulary and some difference in phonetic range.
The following is for description of the Japanese copulae, but, if simply exchanged with corresponding Korean words in Korean Hangul script, it would equally serve well for the Korean ones, except the last paragraph which is concerned with a habit peculiar to Japanese people.
The Japanese copula has many forms. The words da and desu are used to predicate sentences, while na and de are particles used within sentences to modify or connect.
Japanese sentences with copulae most often equate one thing with another, that is, they are of the form "A is B." Examples:
私は学生だ。 | Watashi wa gakusei da. | "I'm a student." | (lit., I TOPIC student COPULA) |
これはペンです。 | Kore wa pen desu. | "This is a pen." | (lit., this TOPIC pen COPULA-POLITE) |
The difference between da and desu appears simple. For instance desu is more formal and polite than da. Thus, many sentences such as the ones below are almost identical in meaning and differ in the speaker's politeness to the addressee and in nuance of how assured the person is of their statement. However, desu may never come before the end of a sentence, and da is used exclusively to delineate subordinate clauses. Additionally, da is always declarative, never interrogative.
あれはホテルだ。 | Are wa hoteru da. | "That's a hotel." | (lit., that TOPIC hotel COPULA) |
あれはホテルです。 | Are wa hoteru desu. | "That is a hotel." | (lit., that TOPIC hotel COPULA-POLITE) |
Japanese sentences may be predicated with copulae or with verbs. However, desu may not always be a predicate. In some cases, its only function is to make a sentence predicated with a stative verb more polite. However, da always functions as a predicate, so it cannot be combined with a stative verb, because sentences need only one predicate. See the examples below.
このビールはうまい。 | Kono bīru wa umai. | "This beer is delicious." | (lit., this beer TOPIC be-tasty) |
このビールはうまいです。 | Kono bīru wa umai desu. | "This beer is good." | (lit., this beer TOPIC be-tasty POLITE) |
*このビールはうまいだ。 | *Kono bīru wa umai da. | This is unacceptable because da may only serve as a predicate. |
There are several theories as to the origin of desu; one is that it is a shortened form of であります de arimasu, which is a polite form of である de aru. Both forms are generally used only in writing and more formal situations. Another form, でございます de gozaimasu, which is the more formal version of de arimasu, etymologically a conjugation of でござる de gozaru and an honorific suffix -ます -masu, is also used in some situations and is very polite. Note that de aru and de gozaru are considered to be compounds of a particle で de, and existential verbs aru and gozaru. です desu may be pronounced っす ssu in colloquial speech. The copula is subject to dialectal variation throughout Japan, resulting in forms such as や ya (in Kansai) and じゃ ja (in Hiroshima).
Japanese also has two verbs corresponding to English "to be": aru and iru. They are not copulae but existential verbs. Aru is used for inanimate objects, including plants, while iru is used for people and animals, though there are exceptions to this generalization.
本はテーブルにある。 | Hon wa tēburu ni aru. | "The book is on a table." |
キムさんはここにいる。 | Kimu-san wa koko ni iru. | "Kim is here." |
Japanese people, when learning English, usually drop the auxiliary verbs "be" and "do" due to believing incorrectly that "be" is a semantically empty copula equivalent to "desu" and "da".[10]
N.B. The characters used are simplified ones, and the transcriptions given in italics reflect standard Mandarin pronunciation, using the Pinyin system.
In Chinese languages, both states and qualities are generally expressed with stative verbs (SV) with no need for a copula, e.g. in Mandarin, "to be tired" (累 lèi), "to be hungry" (饿 è), "to be located at" (在 zài), "to be stupid" (笨 bèn) and so forth. These verbs are usually preceded by an adverb such as 很 hěn ("very") or 不 bù ("not"). It could also simply be the pronoun and verb. e.g. 我饿。 wǒ è. "I am hungry."
Only sentences with a noun as the complement (e.g. "this is my sister") use the verb "to be": 是 shì. This is used frequently: for example, instead of having a verb meaning "to be Chinese", the usual expression is "to be a Chinese person", using 是 shì. Some scholars call this verb form an equative verb (EV), as published in some Yale Chinese textbooks.
The history of the Chinese copula 是 is a controversial subject. Before the Han Dynasty, the character served as a demonstrative pronoun meaning "this" (this usage survives in some idioms and proverbs, as well as in Japanese). Some linguists argue that 是 developed into a copula because it often appeared, as a repetitive subject, after the subject of a sentence (in classical Chinese we can say, for example: "George W. Bush, this president of the United States" meaning "George W. Bush is the president of the United States).[11] Other scholars do not completely accept the explanation, proposing that 是 served as a demonstrative pronoun and a copula at the same time in ancient Chinese. Etymologically, 是 developed from the meaning of "straight"; in modern Chinese, however, it can be combined with the modifier 的 de to mean "yes" or to show agreement. e.g. Question: 你的汽车是不是红色的? nǐ de qìchē shì bú shì hóngsè de? "Is your car red?" Response: 是的。 shì de. "Yes." (Shows positive agreement towards the question, in the sense that the responder is saying "It is.")
A more common way of showing that the person asking the question is correct is by simply saying "right" or "correct", which would be 对 duì which would drop the modifer 的 de.
Cantonese uses 係 (Jyutping: hai6) instead of 是.
In Irish, the copula is used for things that are in a permanent state.
Is fear Liam | "Liam is a man" | (lit., is Liam man) |
Is leabhar e sin | "That is a book" | (lit., is book is that) |
To describe non-permanent states, "to be" is used, eg. Tá mé ag rith "I am running"
In Siouan languages like Lakota, in principle almost all words—according to their structure—are verbs. So, not very unlike in Lojban (see below), not only (transitive, intransitive and so-called 'stative') verbs but even nouns often behave like verbs and do not need to have copulae.
For example, the word wičháša refers to a man, and the verb "to-be-a-man" is expressed as wimáčhaša/winíčhaša/wičháša (I am/you are/he is a man). Yet there also is a copula héčha (to be a ...) that in most cases is used: wičháša hemáčha/heníčha/héčha (I am/you are/he is a man).
In order to express the statement "I am a doctor of profession," one has to say pezuta wičháša hemáčha. But in order to express that that person is THE doctor (say, that had been phoned to help), one must use another copula iyé (to be the one): pežúta wičháša (kiŋ) miyé yeló (medicine-man DEF ART I-am-the-one MALE ASSERT).
In order to refer to space (e.g. Robert is in the house), various verbs are used, e.g. yaŋkÁ (lit.: to sit) for humans, or háŋ/hé (to stand upright) for inanimate objects of a certain shape. "Robert is in the house" could be translated as Robert thimáhel yaŋké (yeló), whereas "there's one restaurant next to the gas station" translates as "owótethipi wígli-oínažiŋ kiŋ hél isákhib waŋ hé".
The constructed language Lojban has copulae, but they are rarely used, and are sometimes viewed with distaste in the Lojban community, because all words that express a predicate can be used as verbs. The three sentences "Bob runs", "Bob is old", and "Bob is a fireman", for instance, would all have the same form in Lojban: la bob. bajra, la bob. tolcitno, and la bob. fagdirpre. There are several different copulae: me turns whatever follows the word me into a verb that means to be what it follows. For example, me la bob. means to be Bob. Another copula is du, which is a verb that means all its arguments are the same thing (equal).[12]
The E-Prime language, based on English, simply avoids the issue by not having a generic copula. It requires instead a specific form such as "remains", "becomes", "lies", or "equals".
Esperanto uses the copula much as English. The infinitive is esti, and the whole conjugation is regular (as with all Esperanto verbs). Additionally, adjectival roots can be turned into stative verbs: La ĉielo bluas. "The sky is blue."
Similarly, Ido has a copula that works as English "to be". Its infinitive is esar, and, as is the case in Esperanto, all of its forms are regular: the simple present is esas for all persons; the simple past is esis, the simple future is esos, and the imperative is esez, among a few more forms. However, Ido also has an alternative irregular form for the simple present ("es"), which some Idists frown upon. The possibility to turn adjectives and even nouns into verbs also exist, although this is mostly done by means of an affix, on top of the verbal endings. The affix is "-es-". So, "The sky is blue." can be said as "La cielo bluesas". As can be seen, the suffix "-es-" plus the verbal desinence "-as" are simply the verb "to be" annexed to the adjectival or nominal root.
Interlingua speakers use copulae with the same freedom as speakers of Slavic, Germanic, and Romance languages. In addition to combinations with esser ('to be'), expressions such as cader prede ('to fall prey') are common. Esser is stated, rather than omitted as in Russian.
The existential usage of "to be" is distinct from and yet, in some languages, intimately related to its copulative usage. In language as opposed to formal logic, existence is a predicate rather than a quantifier, and the passage from copulative to existential usage can be subtle. In modern linguistics one commonly speaks of existential constructions - prototypically involving an expletive like there - rather than existential use of the verb itself. So for example in English a sentence like "there is a problem" would be considered an instance of existential construction. Relying on unified theory of copular sentences, it has been proposed that there-sentences are subtypes of inverse copular sentences (see Moro 1997 and "existential sentences and expletive there" in Everaert et al. 2006 for a detailed discussion of this issue and a historical survey of the major proposals).
For example:
Other languages prefer to keep the existential usage entirely separate from the copula. Swedish, for example, reserves vara (to be) for the copula, keeping bli (to become) and finnas (to exist, lit. to be found) for becoming and existing, respectively.
In ontology, philosophical discussions of the word "be" and its conjugations takes place over the meaning of the word is, the third person singular form of 'be', and whether the other senses can be reduced to one sense. For example, it is sometimes suggested that the "is" of existence is reducible to the "is" of property attribution or class membership; to be, Aristotle held, is to be something. Of course, the gerund form of "be", being, is its own (vexed) topic: see being and existence.
|