Two questions for you wise people: 1) when using the indefinite article with an ancronym that begins /S/, should it be 'an' or 'a' considering that phonetically the acronym would begin with a vowel sound? 2) am I right in thinking that generally with words beginning with /h/, 'an' is used when that /h/ is silent such as in "I'll be there in an hour", and that 'a' is used when the /h/ is voiced such as in "I saw a horrible beast"? (In accents that drop /h/ such as cockney, the 'an' form would be seen more prevelantly of course) p.s. there are other examples on which I'm clearer such as with words that begin with an orthographic vowel but have a phonetic, intial consonant sound such as 'user'. Question 2 becomes more difficult with the pronounciation of 'a' in so much as the variant schwa sound vs the more definite dipthong /eI/. I just wish the correct way generally to approach the subject.
For question 1, it depends on how the acronym is typically (or specifically) pronounced. For example SOS is pronounced ESS OH ESS, so you send out an SOS. On the other SAM (surface-to-air missile) is pronounced like the name Sam, so you launch a SAM. Some acronyms are pronounced differently in different contexts or by different speakers, so your choice of w or an is determioned by how it is being pronounced in that instance: SQL (Structured Query Language) is pronounced either as SEE kwell or as ESS KEW ELL. SQL Server conventionally uses the first pronunciation, so it ius a SQL Server, but a SQL script may have eother pronunxciation, so whether you use a or an will tell the reader which pronunciation you are using. Is that sufficiently confusing?
Thanks Cog / Etan - in summary therefore whatever the scenario - choose the article which best fits. That is to say if 'an' feels right it probably is. I thought perhaps that grammar was inflexible on this point but it would appear to be fluid yet simple. 1 - 'an' where the next sound is a (cardinal) vowel 2 - 'a' where it is a consonant, inlucling /j/ /w/ /y/ where pronounced as a consonant. 3 - ignore orthographic spelling and any acronym or such construction and follow rule one and two. My heart was right but my headed wanted input. This was the result.
I'd actually look to the Chicago Manual of Style in this case for a more definitive answer; my copy is currently 40 miles away, though.
The difference is on the phonological level. The two indefinite articles are allomorphs. They have exactly the same meaning. Beyond that, graphemes are irrelevant to speech, which should always be the guide in such instances. It is always important to remember that writing is the secondary form of communication. So your basic concepts are correct. I'm not sure how "cardinal" vowels apply, though. Are you opposing them to syllabics or semi-vowels?
Terminology may have let me down with respect to cardinal vowels. What I intended to clarify above was if the <an> allemorph is used exclusively when preceeding phonetic vowel sounds as oppose to orthographic or graphemic representations of vowels. It would appear that this is case and that <an> is not adopted with semivowel / approximant construction. I can't think of an example in which a syllabic consonant would be affected by this though there may be of course.
Ready for a perplexer? Which is correct? I added a herb to the broth. I added an herb to the broth. LOL
all you have to do is say it out loud... how you say it will tell you whether 'a' or 'an' works best...
Like usual you're right...but I was looking for a more story-crafting answer. I have noticed that the pronunciation of herb as "urb" or "herb" varies between cultures and also within socioeconomic strata in the US. Consequently, the use of "a" or "an" can serve in writing as a way to help add a subtle hint of dialect to characters. For example, Julia Childs always pronounced the "h".
Wikipedia, in so far as it is to be trusted, offers the following and I see nothing wrong with it: "The form "an" is always prescribed before words beginning with a silent h, such as "honorable", "heir", "hour", and, in American English, "herb". Some British dialects (for example, Cockney) silence all initial h's (h-dropping) and so employ "an" all the time: e.g., "an 'elmet". Many British usage books, therefore, discount a usage which some Americans (amongst others) employ as being a derivative of the Cockney. The reason is that the indefinite article a is pronounced either of two ways: as a schwa, or as the letter itself is pronounced, "long a" (actually a diphthong, /eI/). Some words beginning with the letter h have the primary stress on the second or later syllable. Pronouncing a as a schwa can diminish the sound of the schwa and melt into the vowel. Pronouncing it as a "long a" does not do this, but as the pronunciation cannot be prescribed, the word is spelled the same for either. Hence an may be seen in such phrases as "an historic", "an heroic", or "an hôtel of excellence", which was the by-line in an advertisement in a New York City newspaper. Merriam-Webster's Dictionary of English Usage is more descriptive than prescriptive, but it advises, "You choose the article that suits your own pronunciation." Theodore Bernstein gives the straight vowel-sound vs. consonant-sound explanation but allows that one should indeed say "an hotel" if they think hotel is pronounced otel. The appearance of an or a in front of words beginning with h is not limited to stress. Sometimes there are historical roots as well. Words that may have had a route into English via French (where all h's are unpronounced) may have an to avoid an unusual pronunciation. Words that derived from German however would use a as the h's would be pronounced. There is even some suggestion that fashion may have had some influence. When England was ruled by a French aristocracy, the tradition may have been to exclusively use an, while when Britain was governed by a German-based monarchy the tide may have changed to a." So as for which is correct, a herb or an herb, the answer would be both depending on context. A herb is correct for those that voice the initial /h/ sound, and an herb will be correct liaison for those that drop the /h/ or do not recognise it as phoentic chunk of language at all. As for socio-variance: two types exist. Socio-marking variants and none socio-marking. Certain language traits will mark a speaker as hailing from a certain region, the canadian trill I believe is an example of this, or will mark them in some other way such as by race, possibly even by sex or religion but increasingly more common is by aesthetic. The way in which people choose to speak may mark them as belonging to some social group or another. Verlan I believe is an example of this, identifying (when at its origin) the speaker as belonging to the French underclass and started as a conscious choice to deform language and invert it. Verlan itself is the deformation of l'envers - the inverse. Some of these terms have since entered general French slang however, but still retain a marking quality in their usage. None socio-marking variants include such language traits as the verying pronounciation of either "ee-ther" and "eye-ther". The difference between to the two does not tend to mark out a speaker as hailing from one region of another and is learnt usually from one's peers. This does not stop some from trying to imprint this variant socially as Wikipedia again suggests, though it my opinion that it is none socio-marking: Either has two different pronunciations in modern English. The pronunciation "ee-ther" prevails in American English, and is the pronunciation of the majority of English speakers. The pronunciation "eye-ther" is associated with British English and Canadian English, but it is not universal in either place or in Australian English and other dialects that take their lead from British English. It is also found in the U.S., especially in New England, although many Americans will regard it as an affectation.
