Fortunately there are some rules that hold fairly well. In final clusters, after a voiced consonant we always get z: fibs, adds, begs, slams, hens, tells. The -s is almost always inflectional, but the rule still works when it isn’t: lens lenz.
This is why newsreaders and others tend to mispronounce the name of the city of Homs, now in the news. As the Oxford BBC Guide to Pronunciation reminds us, it is properly hɒms, rather than the hɒmz we often hear. The only way this could be clearly signalled in English spelling would be if we were to write Homss. In Arabic it’s actually حيمص Ḥimṣ ħimsˁ, with a final voiceless ‘emphatic’ (pharyngealized) alveolar fricative. The city was previously known as Emesa (Greek Ἔμεσα), again with a voiceless fricative.
I notice that in the Japanese Wikipedia it’s ホムス homusu, not ホムズ homuzu. Quite right.
I have known the word cowslip, the name of a wildflower, from an early age, and have always pronounced it ˈkaʊslɪp. As far as I know, so does everyone else. The s here does not trigger pre-fortis clipping, so it is natural to analyse the word as cow plus slip. Recently I had a sudden thought: is it really from cow’s plus lip? Does the flower or its leaf look like part of a bovine mouth? (There’s another, similar, flower called an oxlip.) That would make the fricative an inflectional z. Are there, or were there, people who say ˈkaʊzlɪp?
But no. The OED tells us that its etymology is
Old English cú-slyppe, apparently < cú cow + slyppe viscous or slimy substance, i.e. ‘cow-slobber’ or ‘cow-dung’ (compare German kuh-scheisse as a plant-name in Grimm)
It’s got nothing to do with a cow’s lip. Where the cow slips, there slip I.
Posts
The Survey of English Dialects, question II.2.10, records very few responses with [z] - one in Northumberland and otherwise a few in the southwest, where voicing of the initial s- in cow-slip is the likely explanation.
ReplyDeleteIsmail Tohari asks me to post the following comment:
ReplyDeleteIn Arabic we have حسن and حسان. The former is usually written as HASSAN. Yet, the latter, though being different in Arabic, is written the same as the former i.e. Hassan. When written in Arabic letters, we have no problem as they are completely two different names. But the problem arises when I want to transliterate them into English. Some use Hassan (double s) for حسن on the ground that if written with one (s), this (s) would be pronounced as 'z' Hazan. By doing this, writing Hassan with double s, they end up with a different name.
How could such a problem be solved?
The only way this could be clearly signalled in English spelling would be if we were to write Homss.
ReplyDeleteI can think of two other ways: Homse and Homce. They look a little strange, but not nearly as strange (to me) as the sequence -mss.
There's also a case for Homps or Hompse.
None of these signal a sound other than English-like s, but only a specialist transliteration system for Arabists can do that.
In any case, does the Classical Arabic pronunciation correspond to what Syrians actually say?
Since there is surely no native English word ending in /ms/, any choice of spelling will have detractors. I like Homse myself: it evokes the many native English words ending in /ns/ and spelled "-nse", such as dense, immense, license, rinse, sense, and tense (although "cleanse", ending in /nz/, is one annoying irregularity to even this pattern). Homce is another possibility, although it looks weirder to me.
DeleteHomps? yuck. There's no need to insert extra consonants, even if some people will pronounce them anyway.
When I try saying "homps" with a P, it comes out sounding the same as without the P. Inserting a P, phonemically, does not seem to change the phonetic outcome.
DeleteAs I said, some people will insert a [p] anyway, but others will not.
Deletevp
DeleteNeither -mse not -mce is used, as far as I can tell, although both are available. Both -mps and -mpse are actually used (e.g. amps, glimpse and, as a converse to your observation, some people will omit a [p] anyway, but others will not.
Not that I'm advocating any change to the status quo. The transliteration Homs is good enough for non-scholarly purposes, and is readable in any pretty well language that uses the Roman alphabet. And if one English speaker says hɔmz to another English speaker the hearer will know exactly what he or she means — assuming, of course, a knowledge that there is a city with the conventional spelling Homs.
"The only way this could be clearly signalled in English spelling would be if we were to write Homss."
ReplyDeleteEven this does not help if we consider the double ss in Missouri, which is pronounced Mi[z]ouri.
Or hussar
ReplyDelete'In Arabic it’s actually حيمص Ḥimṣ...'
ReplyDeleteWikipedia has it as just حمص.
Not in the English version I've been reading, Steve. And according to the article History, the text hasn't been changed since 17 Feb.
DeleteI presume the short i — modified by the consonants around it — came to Arabic from Syriac, which had the equivalent letters and sounds, and equivalent modifying effects. Wikipedia is not addressed to children or to learners of Arabic, so does not mark short vowels.
The Greeks wrote Ἔμεσα without a 'rough breathing' to signal an aspirate sound. Presumably they would have spelled it differently if they'r heard the local name earlier. But clearly they did hear a short front vowel in the penultimate syllable.
Wikipedia misleadingly says previously known as Emesa (Greek: Ἔμεσα, Emesa). Surely it's rather a question of being previously recorded as Emesa. The locals must have know it as something with an initial aspirate and a final 'emphatic'.
David: are we talking at cross-purposes? See mallamb's comment below. John's Arabic version has U+064A ARABIC LETTER YEH in between the initial U+062D ARABIC LETTER HAH and U+0645 ARABIC LETTER MEEM, making four letters in all, whereas the Wikipedia I'm looking at (which was indeed last edited 17 Feb) just has three letters: U+062D ARABIC LETTER HAH, U+0645 ARABIC LETTER MEEM, U+0635 ARABIC LETTER SAD.
DeleteSteve
DeleteYes we are at cross purposes. You included Ḥimṣ in your quote of John and not in your quote of Wikipedia — so I thought that was the difference you were writing about. I could never read Arabic well, and now I'm very stale and the print is impossibly small on my browser.
Mallamb's comment being below, I wasn't aware of the ي problem. Of course you're both right — and so is Wikipedia. And that's how Arabic Wikipedia spells it too.
I don't think John has noticed because nobody has said John, there's a spelling mistake in the OP.
Or possess, or scissors. Or electricity if you count that.
ReplyDeleteBut the rule about pronouncing single s as z doesn't apply to transliterated loanwords. Arabic Seen and Saad are always transliterated as s, while z is reserved for the voiced counterparts Zayy and (inconsistently) Zaad. Also, consonant length is distinctive in Arabic so we don't normally double a letter in English unless the corresponding letter is double in Arabic.
This applies only to modern loanwords, of course. Naturalised Greek forms, for example, are subject to normal English spelling conventions. And medieval Arabic loanwords have been through two or three languages on their way to English and often end up pretty garbled. For example the two s sounds in Saracen represent ʃ and q respectively (assuming the word is related to the Arabic root شرق "East").
Even within English initial S is always unvoiced (as John Wells parenthetically alludes to in the first sentence). So transliterations with initial S aren't really relevant, it seems to me.
DeleteAs for ss, isn't it always unvoiced at the end of a word?
With Sion or syne there is an alternative pronunciation with /z/
Delete1. That's why we have the spelling Zion.
Delete2. Few people in England have a clue what auld lang syne means. I think we started singing it as a simple meaningless word, which makes the sibilant voiced between g and a vowel. This goes for collective history and individual history — seeing the words in print came later in the culture and comes later in each lifetime.
Off-topic, spurred by your reference to pre-fortis clipping:
ReplyDeleteHow does this clipping interact with the vowel length distinction used by Australian English? That is, I see /e/ vs /eː/ in most works. I see several possible options:
1) Does pre-fortis clipping not operate there?
2) Does it leave AusE with three phonetic vowel lengths?
3) Do you analyze AusE as having a quality distinction like other dialects?
Minimal set: bed, bet, bayed, bait — [beˑd], [bet], [beːd], [beːt] ?
Are you saying that the vowel lengths in "bayed" and "bait" are identical? I'd find that a bit surprising.
DeleteAnyway, what I'd expect, on the basis of dialects such as RP, is a two-way phonemic length distinction (short DRESS vs. long FACE) overlaid with realization rules that cause a shortening of the vowel before a voiceless consonant. (There may also be morphological considerations that might make "bayed" even longer than e.g. "maid"). In other words, two phonemic lengths, but at least four (and probably more) phonetic lengths.
Well, the Australian FACE diphthong tends to be very broad, along the lines of [æi] so any contrast with DRESS will be based on quality. Consult anything by Felicity Cox if you want sources.
DeleteIf you want to argue for this, then START/STRUT is a much better example. According to most people, the quality is more or less the same, perhaps [ɐ]. So you'd have (assuming "short vowel" + voiced obstruent is marginally longer than "long vowel" + voiceless obstruent, which does not need to be true for AusE):
card - cud - cart - cut
Oh, and for [e], you may have meant DRESS/SQUARE. (Occurred to me only now.) But then you'd probably be hard pressed to find a minimal quadruple. My dictionary, at least, finds only scarce as a plausible word ending in SQUARE + voiceless obstruent.
DeleteYou are quite right, I only realized that just before I read your post. Being rhotic myself, I saw /eː/ on WP and ran with it.
Deleteحيمص is wrong. The ya would make the i long. But the usual spelling حمص with no vowelling could be almost anything – even the chickpea dish hummus without the shaddah. It could certainly be [ħomˁsˁ], and no doubt is in some dialects.
ReplyDeleteYou had me fooled with 'cow's lip'. But are you sure that the s here does not trigger pre-fortis clipping? It does for me, as in 'teaspoon', which you acknowledge somewhere, and that's obviously analyzable, so it's in spite of the morphology. Compare 'Where the cowslips grow' to 'Where the cow slips, there slip I'.
But oxlip had the entire English-speaking world fooled. According to the OED, that too comes from ox+slip, and who can say how much of that is phonetic slippage and how much is folk etymology? If the latter, the wonder is that the analogy with it didn't give us your ˈkaʊzlɪp!
I don't have any problem with the idea that the letter s can represent the sounds 's' and 'z'. isn't at least part of the problem here that many spellings in English represent more than one sound, an idea we've been struggling with for hundreds of years. As John Hart (1570) put it, the letters 'are misnamed much from their offices and natures, whereby the desirous are much more the hindered from learning to reade, though they were neuer so willing'.
ReplyDeleteIn fact, I like Albrow's (1972) approach to the question. If we start from the sounds of the language and look at how the spellings system relates to the sounds (i.e. a spelling can contain one, two, three or four letters, a sound can be spelt in multiple ways, and many spellings represent different sounds), we have a manageable, though never completely tidy, system for teaching and learning at school.
Thus, in the word 'answer', the sw, a (very infrequent) two-letter spelling, would represent 's'. In Saxon times, we might have spelt the word 'answeare', with each letter representing a sound, but then language changes over time and from place to place.
It was andswerian, actually, to swear against someone.
DeleteEarly in his career, Ernie Els had his last name pronounced as if it were "elz" pretty frequently.
ReplyDelete