This is NOT a request for a recommendation of a "best" book. I'm looking for a definitive authoritative source to address a specific question...

I recently had a discussion with a friend of mine who is a retired high-school English teacher. We were discussing the correct usage of "a" vs "an".

In her opinion the article must always match the subject. Thus she says "an emergency" is correct but so is "an dire emergency", she claims.

I personally think she is wrong, that the correct usage of the indefinite article is never based on the subject per se but rather only on the vocal pronunciation of the most proximate word following the article.

Any suggestions for a source that would generally be well-respected by school teachers?

  • 8
    There's no authority such as the Academies of France and Spain. As for 'an dire emergency', that's plain wrong. Whether to use 'a' or 'an' is a purely phonetic decision: when the word begins with a vowel sound, use 'an'. Hence 'a university' [university begins with a /j/ sound] and 'a one-time chance' [one begins with a /w/ sound]. Aug 1 '15 at 16:55
  • 6
    In order to resolve the matter you raised about a and an you only need to read a book or a newspaper. NOBODY interprets the rule in a way which would produce an dire emergency. It is entirely erroneous to think of it in that way.
    – WS2
    Aug 1 '15 at 17:02
  • 3
    @O.M.Y.: All I can say is I'm gobsmacked to see that your friend was apparently an English teacher! On the plus side, at least she's retired. Aug 1 '15 at 17:05
  • 3
    There are two definitive grammars of modern English, and many lesser ones. The two big ones are the Cambridge Grammar of the English Language by Huddleston and Pullum, which runs about 1500 pages and requires two hands. The other is McCawley's The Syntactic Phenomena of English, which is less than a thousand pages and has a paperback edition. Neither is light going. We're talking college textbooks. Aug 1 '15 at 17:27
  • 6
    The 2002 reference grammar by Huddleston and Pullum (et al.), The Cambridge Grammar of the English Language, discusses that issue on page 1618: "The other case is specific to one grammaticised word, the indefinite article, which has an as a liaison form before vowels: … An is used when the next word begins with a vowel. The choice between a and an depends purely on the phonological context. The liaison form occurs before a vowel sound, not before particular letters."
    – F.E.
    Aug 1 '15 at 18:41

There is no central authority.

  1. Search online for a versus an, you will see that the rule is If the next word begins with a vowel sound then use 'an' Here is the first example that I found http://www.quickanddirtytips.com/education/grammar/a-versus-an?page=all

Note that it says the next word, not the next noun and it also says a vowel sound not a vowel.

  1. Search the works of some famous and well respected authors. Find examples. You can start with Shakespeare, continue with Dickens, and then search online for famous English authors, e.g https://en.wikipedia.org/wiki/Category:20th-century_English_writers

  2. Buy any book, any newspaper, any magazine, any packet of soap-powder. Read it and mark the examples you find.

  3. Use Microsoft Word (or your favourite word-processor) - type various combinations. Use the grammar/spelling check. You will see that incorrect use of 'a' or 'an' will be indicated.

enter image description here

  • What I am looking for is a scholarly and peer reviewed reference. While I am sure I could find hundreds of Google hits that demonstrate the "correct" usage I have no doubt I could find an equal number of variant and just plain wrong demonstrations. Look at how often people use the word "not" to end a sentence in order to negate the precedent. As you are from the UK you know that examples of "an hospital" are prevalent in the British use of English but in the US it will be "a hospital" because we always pronounce the "H" in that word.
    – O.M.Y.
    Aug 1 '15 at 17:39
  • PS: What is a "packet of soap-powder" ?
    – O.M.Y.
    Aug 1 '15 at 17:54
  • @O.M.Y - I think it is highly unlikely that you will find a scholarly work on the subject. In English, our authorities are our forebears. I did not suggest using Google search, that will turn up all sorts of nonsense. I recommended looking at published work. If you wish to do research then Google Scholar maybe a good place to start. I don't intend to do the research for you. Aug 1 '15 at 18:19
  • @O.M.Y. - search online for "packet of soap-powder", you will find it. google.co.uk/?gws_rd=ssl#q=%22packet+of+soap-powder%22++:us. But that's off-topic ;-) Aug 1 '15 at 18:21
  • Google contains lots of "nonsense" but it also contains the texts of many of the very publications you suggest I review and I am capable of filtering out the wheat from the chaff. My point was that given any subjective source (which all first person usage would be) one can find numerous valid and invalid examples. Sheer numbers of "good examples" is the literary equivalent of mob rule. Even the dreaded Wikipedia does not allow quantities of citations to override quality of citations. Thus my search for an objective (peer reviewed or at the very least well vetted) reference work.
    – O.M.Y.
    Aug 2 '15 at 2:28

Chasly is correct. I am amazed that anyone could recommend "an dire ..." I have never heard or read such a construct and would unhesitatingly correct it when editing. The only thing I can add is that "an" may be used when a word starts with h, as in "an historical account". Even this usage begins to sound pedantic in contemporary prose.


One such study is available here (for a price): A Corpus-Based Sociolinguistic Study of Indefinite Article Forms in London English, in Journal of English Linguistics, December 2010, vol. 38, no. 4, 297-334.

I don't claim that is the definitive study (and I won't even touch authoritative, for fear of revenge and retribution from the anti-nazi-grammar nazis), and I don't expect I (or anyone) can produce anything other than an opinion on which that definitive study might be...and even then, that opinion would have to hedge around which of the possible scholarly domains the study was definitive in and for what time period. If the realm is scientific (broadly speaking, to include sociology, psychology and etc.), the notion that there is, and if so, which of the available is, a definitive study, will always be subject to debate.

In any case, the study cited may provide useful insights into general patterns of and underlying influences on a/an choice, although it (perhaps usefully) restricts the data analyzed to a London corpus. The cited study does have the advantage of being a more recent analysis by 8 years than such petrified sources as the Cambridge Grammar (cited above in comments), and it notably takes into account more recent trends reflecting the rapid change in language usage over those 8 and the subsequent 5 years.

  • I am hesitant to select your Answer (A) primarily because this is a pay-to-access text which rather defeats the purpose of SE to provide detailed Answers to future readers, and (B) as you yourself noted this particular work is a very regionally confined usage study and certainly not even vaguely authoritative.
    – O.M.Y.
    Aug 2 '15 at 15:38
  • 1
    @O.M.Y. - You've had your answer. I think it's time to accept it. The idea that the indefinite article is spelled in accordance with the following noun in preference to the following word whatever it may be is utterly ridiculous. No-one in the history of the English language has ever followed that rule and no native speaker ever will. Aug 2 '15 at 23:54
  • No-one in the history of the English language has ever followed that rule ... "It was [Otto] Jespersen who first questioned the Bloomfieldian solution. In 1941, he proposed that the syntactic class of the following word determined the form of the indefinite article; specifically, an occurred before adjectives, and a before nouns." per this source provided (indirectly) by Mari-Lou A
    – O.M.Y.
    Aug 4 '15 at 18:57
  • The idea that the indefinite article is spelled in accordance with the following noun in preference to the following word whatever it may be is utterly ridiculous. ... "in 1959, [Noam] Chomsky [...] proposed the revolutionary theory that an is the indefinite article used with animate nouns and a that used with inanimates." per the same source as for my Jespersen comment.
    – O.M.Y.
    Aug 4 '15 at 19:06
  • 1
    @O.M.Y: Specgram is a satirical organization, in case that's not clear. Its pupose is humor, not truth.
    – herisson
    Aug 4 '15 at 22:03

Not the answer you're looking for? Browse other questions tagged or ask your own question.