Thursday 30 March 2017

An historic moment

The indefinite article of English is either <a> or <an> depending on whether the following word starts with a consonant or a vowel sound - so far so regular. But what do we make of book titles such as these?
How do you pronounce the phrase "an historical" or "an historic"? Should it be /ən ɪstɒrɪk(əl)/ or /ə hɪstɒrɪk(əl)/? Could one also say /ən hɪstɒrɪk(əl)/?
Well, Alex Rotatori has spotted an interesting spoken example of the latter version (though the indefinite article is used here in its strongform /æn/); it's taken from Theresa May's speech in the House of Commons announcing the delivery of the Brexit letter to His Excellency Mr Donald Tusk, president of the EU.


Is it something which occurs occasionally, even regularly, is it idiosyncratic or just a slip of the tongue/brain?

9 comments:

  1. Happens all the time. You have the same phenomenon in GenAm before vowels - it's a sign that the rule isn't productive but fixed. Most of the time, Americans seem to use "a" in front of vowels, and "an" may be a variant, possibly with a difference in register. That leads to it that you often hear the irritating [ən ʕ].

    ReplyDelete
    Replies
    1. "That leads to it that you often hear the irritating [ən ʕ]."

      Haven't heard many pharyngeal fricatives in the US, but I'll listen out for them :)

      Delete
  2. Omce upon a time RP had a thing about some initial /h/ that were regularly dropped, "hotel" comes to mind and there were others. Jones' EPD (1958) gives both hou'tel and ou'tel with a note that some people always say ou'tel, others occasionally. He noted ist.. occasionally for "historical". In Outline §787 (1932 and 1962) he wrote "Those who pronounce the h in hotel when said by itself would often drop it in a good hotel". Gimson (1962:187) follows Jones more or less, but he does confirm that dropped /h/ in "historic" gets "an". Wells (1982:286) writes "In hotel, historic, hysteria etc present-day RP has on the whole restored the /h/ which used to be dropped".

    So if T.M. had wanted a Jonesian RP pronunciation she'd've got away with "an istorical". Now T.M. doesn't speak R.P., but regional home counties SBE, and would've said "an istorical" or "a historical" depending on style. But she said "an historical", that I assume is a hyper-correction, suitable for this istoric occasion. Regarding T.M.'s accent, listen (in your link) to o in "historic" and o in "no", they're not RP LOT or GOAT.

    ReplyDelete
    Replies
    1. PS. Regarding your questions: written "an historical" is pronounced "an ist...", the "an" indicating a dropped /h/. The same goes for initials: "an EU directive" (although some editors might insist on writing "a EU"). So your students can continue to alternate a/an, remembering that an comes before a pronounced vowel (observing that "Union" starts with a pronounced consonant).

      Delete
    2. Your EU example raises the question of whether when we use abbreviations, we intend them to be decoded or not, hence 'an /i: ju:/...' vs. 'a /ju:r.../.
      It's something I find comes to mind whenever I write an/a Xmas card.

      Delete
  3. Jack Windsor Lewis:
    What I feel sure has caused folk to say 'an historic' instead of the more sensible and cumftable /ən ɪstɒrɪk/ is sheer horror of being perceived as uneducated. I grew up in a community where aitches were in rather short supply so I much prefer the latter. Fortunately, nobody can hear me when I write ‘an historic’.

    ReplyDelete
  4. I have to provide a different perspective on [ən hɪs...] to other commenters. I'm a - presumably idiosyncratic - user of this pattern. I've always done it, and since university days have wondered why it's in my speech. I'm not alone (clearly, Teresa ;) ), but I don't know the distribution of it here in Australia (without /h/-drop). I have heard it here. A complicating factor is that I might have picked it up in London as a child (I spent age 3-9 in the UK). I also do it for (some) other words of the pattern /hə'CV.../ (my dialect has unemphasised schwa, not ɪ).

    So, I don't do it (and didn't start doing it) as a class-related behaviour, nor as a conscious hypercorrection, but it's possible that was what adults around me were doing when I was acquiring my phonological rules ;)

    With regard to the class/education side of things, I must say that I noticed *immediately* when T.M. did it in a speech recently and felt a flush of happiness:P The fact that I noticed it, however, means that I hadn't expected it in her speech, so I perhaps have a sociolectal (or dialectal) model of where I regard /ən hV.../ as un/marked.

    ReplyDelete
    Replies
    1. Thanks Duncan. It's clear there's something else going on here that has nver been picked up by pronunciation books. The RP thing, almost a shibboleth in its day, can be traced through Daniel Jones' books. Let's ignore h-spellings that are hardly pronounced in any dialect of English - heir, hour etc. Apart from that, RP was not an h-dropping accent except perhaps away from stressed or focused syllables. But there was this small set of words with a weak initial syllable, pronounced without h and consequently with ən or ði as appropriate before the vowel. 'Outline' edn 1 (1918 §337, ready 1914) says a 'quite usual' pronunciation, and lists horizon, historical and hotel, repeated in the completely rewritten edn 3 (1932 §787 and 1967). Gimson (1962 p.187)) lists 'historical' and notes 'nowadays used by only a minority of speakers', but an h-less 'hotel' is 'fairly wide spread'. Cruttenen (2014 p.207) lists the same three words (but 'history' rather than 'historical') and 'many speakers' drop the /h/. So that all possibly means it's in decline in RP.

      The rule is generalized to unstressed word-initial syllables, but only three words are ever mentioned. Jones' dictionary (1957) has these three words and additionally 'hallucination, heretical, historian, historic' when not initial (i.e. preceded by a consonant, like 'good (h)istorian'). In contrast, there's an enormous number of words that fit the rule but are excluded, based on Greek or Latin roots like 'haemoglobin'. I don't have access to current dictionaries. So the shibboleth is down to three words, no wonder non-RP speakers were confused, especially when the pronunciation 'ən istorical' was written 'an historical' in text.

      So back to Theresa's 'ən historic' with pronounced /h/, the converse of RP practice. Your evidence indicates at least one generation acquiring it as children from adults around them. If there's any hyper-correction involved, it would be an earlier generation. There's a possible PhD topic here for someone, or at least an MA.

      Delete