Standards of ‘correct’ English

Within the practice and study of lexicography, dictionaries are often divided into two groups: prescriptive dictionaries, which recommend how language should be ‘properly’ used, and descriptive dictionaries, which record how language is actually used. Today, most professional lexicographers strive to be descriptivists, collecting evidence of real linguistic usage from books, newspapers, blogs, and other sources, and reporting what they find without opinion or bias.

For English dictionaries in particular, the descriptivist approach is often traced back to the OED. When the project to create what would become the OED was in its earliest stages of development by the London Philological Society, one of its members, Richard Chenevix Trench, famously declared that a lexicographer was ‘not a critic’ but ‘an historian’. As he saw it, the task of the dictionary-maker was to gather together the words of a language, ‘whether good or bad, whether they commend themselves to his judgment or otherwise’ (Trench 1857: 4).1 Applying methods of objective empiricism would allow lexicography to rise to the status of a science—and the Society was intent on creating a ‘more Scientific Dictionary than any at present existing’ (Philological Society 1859: 2).

However, the break between prescriptivism and descriptivism is not as clear-cut as this might suggest. Trench argued that dictionaries should not omit a word simply because their compilers did not like it, but he had no qualms about compilers making their dislikes known. ‘Where [a lexicographer] counts words to be needless, affected, pedantic, ill put together, contrary to the genius of the language, there is no objection to his saying so,’ he remarked; ‘on the contrary, he may do a real service in this way’ (Trench 1857: 5).2 How could this allowance for personal taste be reconciled with the aims of a scientific dictionary?

This question did not need to be fully confronted for some years, as at first work on the new dictionary consisted mostly of collecting lexical evidence rather than editing it. However, when James Murray became chief editor in 1879, he needed to decide exactly where the line between the historian and the critic lay. The task would not be made easier by the dictionary’s future users. Even when Murray and his colleagues endeavoured to describe language objectively, there were readers who insisted on taking their definitions as prescriptive statements. Soon after the first part of the OED, A–Ant, was published in 1884, one user wrote to complain of the dictionary’s inclusion of advertisemental, a word he thought was ludicrous. Murray shot back: ‘The delusion is wide-spread that the Dictionary by recording facts as to the use of words, presses readers to use them […but] the Dictionary does not advise you to say [advertisemental], it merely records the fact that such has been said’ (840222A).

Still, the letters poured in. Apart from those objecting to what had been published in the dictionary (and what had been left out), there were inquirers who anxiously sought the private opinion of its chief editor on matters of ‘correct’ speech and writing. In his replies, Murray sometimes vacillated between recording the facts and dispensing advice. When one inquirer asked in 1894 if a 700th anniversary could be called a septingentenary, Murray dismissed the word as ‘a mere imitation of centenary, formed on false analogy, and without proper etymological basis’ (940430A). To another correspondent who sent him a list of neologisms and asked if they were legitimate English words, he answered in 1911, ‘I am not a judge of the language—only its historian’ (111021A). Nonetheless, he went on to pass judgement on each of the neologisms. Cynicize was simply ‘a nonce-word’, but auto-biophony ‘seem[ed] very unnecessary, & meaningless’, while evolenter ‘could only have been composed by an illiterate person, who should not thus expose his illiteracy’.

Murray was aware that verdicts like these depended on individual preference rather than objective truth. When the same correspondent pressed him on how to pronounce acoustics and laboratory, he replied, ‘it is a matter of taste. Some people wear turned-down collars, & some wear stand-up collars; why should they not? Is not speech as free as dress, when the pronunciations are equally well-grounded?’ (111021A). Sixteen years before, Murray had posed a similar question to another inquirer when she had asked about the pronunciation of pronunciation itself: ‘We do not all think alike, walk alike, dress alike, write alike, or dine alike; why should not we use our liberty in speech also, as long as the purpose of speech, to be intelligible, and its grace, are not interfered with?’ (950105A).

Of course, the provisos in Murray’s questions are significant. One way of speaking may be more or less well-grounded, graceful, or intelligible than another. But if it is the duty of lexicographers to decide these cases, then can they really do so on a scientific basis? Murray’s co-editor Henry Bradley thought so, at least when it came to the language of science itself. When Murray wrote to him about the difficulty of settling the pronunciation of chemical terms that were found more often in writing than in speech, Bradley replied that Murray’s own conclusions ‘must be absolutely followed’ (901212A):

It is very unlikely that there is any one else, even among professional chemists, who has worked out the whole question as you have, & it seems to me that this is one of the few matters in which a dictionary-maker, if properly qualified, may claim to make law instead of merely recording usage. There will be fashions in the pronunciation of these as of other words; but […] the authority of a lexicographer, whose decisions are based on a well-reasoned body of principles, will prevail over any irregularities of usage.

In Bradley’s view, the lexicographer’s credentials as a scientist of language trump the credentials of the chemists who actually use the terms. Presumably, this is in part because the lexicographer has amassed a body of evidence about the word’s history and its current use, to which the chemists do not have access. And yet evidence does not in itself allow a lexicographer to make impartial decisions; in fact, the more evidence there is, the more complex decisions can become. In the preface to the first volume of the OED, Murray (1888: x–xi) mentioned that he had ‘once [been] present at a meeting of a learned society, where, in the course of discussion, he heard the word gaseous systematically pronounced in six different ways by as many eminent physicists’. How was the lexicographer to decide which pronunciations were worth registering? In the end, the OED recorded only two.3 Nor were these instances of ‘law-making’ restricted to technical terms or pronunciations. For example, although in the 1880s the spellings coco and cocoa were both used to refer to the coconut palm by many scientists and laypeople alike, Murray assured the botanist Isaac Bayley Balfour that he would ‘treat cocoa as an incorrect by-form’ in the dictionary (871017A). Years later, he would quarrel with his co-editor William Craigie about how the dictionary should represent the alternative spellings rhyme (the dominant form) and rime (the older form, which Murray insisted the OED should continue to ‘recommend’: 080518A, 080522A).

The point of citing these examples is not to reproach the OED’s editors for failing to live up to their descriptivist ideals, but to recognize that such ideals were (and remain) impossible to achieve in full. Even if the OED had not overtly condemned any words, spellings, or pronunciations, it would still not have been able to encompass every word, spelling, and pronunciation in use in a living language. Decisions had to be made about what to include and exclude in the limited time and space available, and those decisions were influenced by social factors as well as scientific facts.4 As Murray (1888: xvii) himself acknowledged in his ‘General Explanations’ to the OED, lexicographers must ‘draw the line somewhere’. In doing so, they unavoidably become critics.

Further reading

Brewer, Charlotte. (2007). ‘Pronouncing the “P”: Prescription or Description in 19th- and 20th-century English Dictionaries?’ Historiographia Linguistica, 34(2/3), 257–280.

Brewer, Charlotte. (2010). ‘Prescriptivism and Descriptivism in the First, Second and Third Editions of OED’. English Today, 26(2), 24–33.

Mugglestone, Lynda. (2016). ‘Description and Prescription in Dictionaries’. In Philip Durkin (Ed.), The Oxford Handbook of Lexicography (pp. 546–560). Oxford: Oxford University Press.

Finegan, Edward. (2020). ‘Description and Prescription: The Roles of English Dictionaries’. In Sarah Ogilvie (Ed.), The Cambridge Companion to English Dictionaries (pp. 45–57). Cambridge: Cambridge University Press.

2. Trench also allowed that a dictionary of the standard language could omit words that occurred only in dialectal or technical use. Obscene words were another special case: see ‘Defining Obscenity’.
3. The 1898 entry for gaseous, edited by Bradley, gave the forms (gæ·sǐəs) and (gēi·sǐəs), with the initial vowel in each sounded as in mass and mace respectively. Another word whose pronunciation proved to be slippery was the noun content: see letter 910922A.
4. See for instance the issues of androcentrism and Anglocentrism in the OED: ‘Women and the Dictionary, Part II: Authors’ and ‘English around the World’.