In order to understand the objection expressed in the title, we first have to understand the word ‘accusative’. ‘Accusative’ is the name of a case – so we also need to understand about case. Once that has been clarified, we need to understand a little about Latin, because the objection to It is me is based on Latin grammar. Then we need to ask whether English grammar is like Latin grammar in the relevant ways. Finally, we need to ask why the grammar of Latin is taken to be the model of ‘good’ grammar by some people.
Let us begin with the notions of ‘case’ and ‘accusative’. There are many languages (though modern English is not one of them) where nouns have endings to show the roles they play in sentences. These different endings are called ‘cases’. Since we shall need to make reference to Latin later, let us consider what happens in Latin (though the notion of case in another language is discussed in Myth 19: ‘Aborigines Speak a Primitive Language’). A noun like agrícola, ‘farmer’, has this form if it is the subject of the verb, the person or thing performing the action of the verb (for example in agrícola labor at, ‘the farmer works’). However, a different form is used as the direct object of the verb, the person or thing undergoing the action of the verb (for example, puella agricolam monet, ‘the girl warns the farmer,’ literally ‘girl farmer warns’). The form ending in -a is called the nominative form of these nouns. The form ending in -am is called the accusative form. The English names of these cases are borrowed from the Latin. These are just two of the six case forms that nouns have in Latin, but we needn’t worry about the vocative, genitive, dative or ablative cases here.
Not only nouns in Latin have case endings, but adjectives (which we need not worry about here) and pronouns, too. So if you wanted to say ‘the goddess warns her’ (meaning ‘the gir’), for example, you would say dea illam monet (literally ‘goddess [nominative] that-one [accusative] warns’). ‘She warns the farmer,’ by contrast, would be illia agricolam monet (literally ‘that-one [nominative] the farmer [accusative] warns’). The main function of the nominative case is, as has been stated, to show which noun is the subject of the verb. One of the main functions of the accusative case is to show which noun is the direct object of the verb. The cases in a language like Latin are far more important in showing this than the order of the words, so that dea puellam monet and puellam dea monet and even puellam monet dea all mean ‘the goddess warns the girl.’ The cases show the function of the nouns, independent of their position. This is different from English where the goddess warns the girl and the girl warns the goddess mean different things. In English, where there is no case marking for ordinary nouns, the position in the sentence shows the function, and so the position is fixed.
Although showing what is subject and what is object are two of the main functions of the nominative and accusative cases in Latin, they are not the only ones. Another function of the nominative case in Latin is to mark a subject complement. A subject complement is a phrase like the teacher in sentences such as Miss Smith is the teacher. A subject complement refers to the same person as the subject of the sentence (so the teacher is the same person as Miss Smith, but in the goddess warns the girl, the girl is not the same person as the goddess). Subject complements occur only with a small set of verbs like to be (that is, the is in Miss Smith is the teacher), to become, and so on. So if you wanted to say, in Latin, ‘Flavia is a girl,’ you would say Flavia puella est (literally ‘Flavia [nominative] girl [nominative] is’).
Now let us consider whether English has nominative and accusative cases. It has been stated above that English nouns do not have case endings. But English pronouns show a system similar to that in Latin. In English you say I warn him but he warns me, using I and he as the subject of the verb and me and him as the direct object of the verb. We might, therefore, conclude that I, he, she, we, they are nominative case pronouns, and me, him, her, us, them are accusative case. It and you can be either.
If English works in just the same way as Latin, then we would expect to find It is I and not It is me. So how do pronouns in subject complements really work in English? In the King James Bible (for instance in Matthew 14:27, Mark 6:50, John 6:20) we find It is I, with the pronoun I in the subject complement, just as in Latin. However, we need to bear in mind that the King James Bible was written in English that was rather old-fashioned at the time (1611). In Shakespeare’s Twelfth Night (II.v) Sir Andrew Aguecheek uses both me and I in this context within two lines:
MALVOLIO: You waste the treasure of your time with a foolish knight -
SIR ANDREW: That’s me, I warrant you.
MALVOLIO: One Sir Andrew.
SIR ANDREW: I knew ‘twas I, for many do call me fool.
The construction It is me was well established by this time and has been gradually gaining at the expense of It is I ever since. It is me/ him/her can be found in the works of great writers of English such as Christopher Marlowe, Daniel Defoe, Jane Austen, Charlotte Brontë, Charles Dickens and Aldous Huxley, to name but a few.
Ironically, perhaps, Latin did not use the equivalent of either It is I or It is me in contexts like this. When the Latin playwright Plautus has one of his characters ask ‘Who is at the gates?’, the answer comes back Ego sum (‘I am’). In English until the fifteenth century, a construction with I am would also have been used. The construction was usually I am it (though not necessarily with the words in that order). The same construction is still used in modern German: Ich hin es (literally ‘I am it’). It would take us far beyond this chapter to try to explain why such a change should have taken place, but it did. By Shakespeare’s time It is me was frequently heard, even if it is not the majority form in the plays of Shakespeare, Jonson and Marlowe. By the eighteenth century this construction was common enough for some grammarians to feel it was worth trying to discourage it. They pointed to the (supposed) Latin pattern and demanded It is I. Partly as a result of this, both constructions survive today, It is I having a distinctly formal ring to it. Consequently, it is used especially by those who are very conscious of their language use.
If a particular case is used in a construction in Latin, does it follow that the same case must be used in the parallel construction in English? More generally, does the structure of English (or of any language) have to be the same as the structure of Latin? The answer is very clearly ‘no’.
If you look back at the examples of Latin sentences given above, you will see that puella, for instance, is translated as ‘the girl’. But there is no word in Latin corresponding to English ‘the’. No one has ever suggested that English should follow Latin in this respect and omit every occurrence of the word the. English does not follow Latin in that grammatical pattern and need not in others. Or consider what happens in French. The French equivalent of it is me is c’est moi (literally ‘that is me’), and it would be totally impossible to say c’est je (literally ‘that is I’), because je can occur only as a subject in French. Not even the French Académie has suggested that French speakers should say c’est je, even though French derives directly from Latin in a way that English does not. French does not follow Latin in this particular grammatical pattern. Why should English be expected to follow the Latin pattern when French does not? Why don’t we say of English that I can only occur as a subject, as they do in French?
More generally, it is not true that all languages have the same set of grammatical constructions or patterns. It is true that there are probably no languages without nouns and verbs, no languages (except sign languages) without consonants and vowels, no languages which do not have verbs with direct objects. But the number of such absolute language universals is relatively small While there are some languages (such as Latin and Zulu) where verbs have to be marked to show what their subject is, there are others (such as Danish and Mandarin) where there is no such marking. While there are some languages (like Latin and English) which force you to state whether a male or a female person is involved when you use a singular third-person pronoun (i.e. he or she), there are others (for instance, Finnish and Maori) which have no such requirements. Matters such as what case will be
used for a particular function are very definitely in the variable class and not in the universal class.
Despite that, it is quite clear that people’s view of what English should do has been strongly influenced by what Latin does. For instance, there is (or used to be – it is very infrequently observed in natural speech today) a feeling that an infinitive in English should not be split. What this means is that you should not put anything between the to which marks an infinitive verb and the verb itself: you should say to go boldly and never to boldly go. This ‘rule’ is based on Latin, where the marker of the infinitive is an ending, and you can no more split it from the rest of the verb than you can split -ing from the rest of its verb and say goboldlying forgoing boldly. English speakers clearly do not feel that to and go belong together as closely as go and -ing. They frequently put words between this kind of to and its verb.
Why should the patterns of Latin dictate what is acceptable in English? The reason is to be found in the role Latin played in the history of Western Europe. Latin was the language of the powerful and the learned in Western Europe for a thousand years. In Italy, Dante wrote a piece c. 1300 praising the use of Italian rather than Latin. He wrote it in Latin. In France a royal decree of 1539 prescribed the use of French rather than Latin in the courts of law. Erasmus, a Dutchman who died in 1536, wrote entirely in Latin. English did not become the language of the law in England until the seventeenth century.
Against this background, Latin was seen as the language of refinement and education into the eighteenth century. The prestige accorded to the churchmen, lawyers and scholars who used Latin was transferred to the language itself. Latin was held to be noble and beautiful, not just the thoughts expressed in it or the people who used it. What is called ‘beauty’ in a language is more accurately seen as a reflection of the prestige of its speakers. For parallel comments, see Myth 4: French is a Logical Language.
Because Latin had this prestige, people thought that English would gain similar prestige by following the patterns of the language which already had prestige. From a more detached point of view, we can say that this is making a mistake about the source of Latin’s prestige.
Latin gained its prestige not from the grammatical patterns it used but from the speakers who used the language and the uses to which it was put. The Australian aboriginal language Dyirbal shares many of the linguistic features of Latin, but does not have the same social prestige because its speakers do not have powerful positions and the language is not used for highly respected functions in our society. If Dyirbal speakers had sailed around the world and colonized Great Britain and held governmental power in Britain, then Dyirbal might have high prestige – but because of its use, not its structures.
To sum up, Latin has (or had; its prestige is waning as fewer educated people use it) high prestige because of the way it was used for such a long period of time. Some people think that English would be improved if it followed the patterns of this high-prestige language more closely. One such pattern is the use of the nominative as the case of the subject complement. These people think that English is in some sense ‘better’ if it follows Latin grammatical rules about subject complements, and this involves saying It is I rather than the usual modern English pattern (and the usual pattern of a number of other languages of Western Europe) of It is me. To the extent that such people’s opinions mold actual usage, this has now become something of a self-fulfilling prophecy.
But there is another school of thought which says that there is no real point in avoiding the normal English pattern. People who adhere to this view – and I am one of them – believe that even if languages sometimes borrow patterns from each other voluntarily, you cannot and should not impose the patterns of one language on another. To do so is like trying to make people play tennis with a golf club – it takes one set of rules and imposes them in the wrong context. It also follows that you should not impose patterns from older versions of the same language as people do when they try to insist on whom in Whom did you see? And if anyone asks who told you that, you can tell them: it was me.
For a discussion of the various constructions, examples of their use and comments on them, see F. Th. Visser, An Historical Syntax of the English Language, Volume I (Leiden: Brill, 1963, pp. 236–45).