A FRIEND who runs his own business recently asked for thoughts on the macron over the letter "e" in his company logo. He found the Wikipedia article on the macron a bit confusing, so I wrote him a version of what appears below about the macron generally, and in English phonetic transcription specifically.
In many languages (like the rōmaji transliterations for Japanese), the macron indicates vowel length—very literally length, as in how long a vowel is pronounced. In tonal languages such as Mandarin, it represents an even tone, neither rising nor falling. In linguistics, it's used for both purposes.
But in traditional English phonetic transcription—the kind you find in dictionaries—the macron is used very differently. Long ago, someone misanalysed modern English's vowels, calling the vowel in "peat" a "long e", and the "e" in "pet" a "short e". They're not different in actual length: "peat" and "pet" are equally long, but for the vowel in "peat" the tongue is higher in the mouth. (Why we use the letter "e" for the sound nearly every other European language writes with an "i" has to do with the Great Vowel Shift. If you need a primer on that, Wikipedia has a great long article.)
Persisting in this misanalysis of the relationship between the "peat e" and the "pet e", the macron was picked to designate the mis-named "long e" (the "peat" vowel, IPA [i]). The breve was picked to designate "short e". And so in most dictionaries, those labels have stuck. "Peat" is given as pēt and "pet" is pĕt.
So English speakers have learned to use these symbols differently from most everyone else who uses the Latin alphabet. That's why when you walk into Brooklyn's Ikea, under the big sign reading HEJ! ("hello" in Swedish) is a phonetic spelling, "Hā". This is meant to let you know that Swedish hej sounds like English hey. But when I see "Hā", I feel an urge pronounce it "haah", since I've gotten used to seeing the macron used to indicate length. The use of "ā" for the vowel in "face" is even more confusing than "ē" is for "peat", because the "face" vowel isn't a "long a" at all, but a diphthong, written [ei] in IPA. It's a mess.
I was reminded of all this by today's post by Nancy Friedman, letting us know that Kraft's shareholders have approved the name "Mondelez" for the company's global snack-food business. The name came out of an internal company contest. It's not only fanciful (and much-derided); its pronunciation is not obvious, which isn't great for a brand name. Kraft thus resorted to our friend, the macron. It is to let us know that the last syllable is pronounced leez. Confusing for some, but probably necessary in this case.
The macron's quirky use in English phonetic transcription is probably not going away, and in general, wishing language would just bēhāv properly never got anyone anywhere. So with that in mind, have a good wēk-ĕnd.
Readers' comments
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
Very interesting!
On a side note, you write, "It's not only fanciful (and much-derided)..." Does one hyphenate a compound modifier that appears AFTER the (pro)noun in modifies? I was under the impression, for some reason, that one only did so if the modifier appears before the noun.
This business might fancy a macron over the letter "e" in its company logo, but it's not going to appear in any editorial sentence in any publication I edit. The final straw was that New York department store wanting to be Macy*s -- down with commercial typographical decoration; if they want their logo, advertising rates will be supplied on request. So Adidas and BHS and Easyjet and Ebay and Iphone and ...
Well now, looky here. How many words did that take, for the usual combatants hereabouts, to disagree at length about the length and correct pronunciation of peat and pet?
Which, BTW, wasn’t actually what Johnson was on about, was it.
Yet we persist in the forlorn expectation that whatever we confidently say and write should mysteriously but unerringly convey a perfect facsimile of our intended, unambiguous meaning, from mouth to ear and page to eye.
Alas, correct usage, diction, syntax, spelling, grammar and whatever else we may care to include in a veritable lexicon of formalised linguistic rules and ordinances, cannot begin to facilitate the hoped-for nirvana: that two minds might one day think alike.
Remember when George W. Bush said he looked into Vladimir Putin’s eyes, when they met in Slovenia, three months before 9/11, and saw his soul?
Howls of derision?
No less implausible, I think, than when any two other people, anywhere, look ardently into each other’s eyes and aver, with trembling emphasis, in any language, “you know, I understand exactly where you’re coming from”. [OMG! Is it that obvious?]
If you ask me, and nobody has, with no way of knowing the precise pretext, subtext and context, the correct weight of peat, compared to the size of one's pet, for Pete’s sake, is just a fatuous red herring.
I sink, therefore I swim.
Mondelēz isn't the first English-language trade name to have a macron. That honour probably goes to Hōvis bread, as it was named when it was first sold in 1886.
Here is the macron, on an advertisement painted on a house wall in Islington High Street and photographed in 2008:
http://3.bp.blogspot.com/-h99CxedQpi0/T0H-pA_DzoI/AAAAAAAACNQ/fvmF5EYqiW...
The Hovis logo still uses the name in capitals with the 'o' smaller to make room for the macron, but strangely the mark has been replaced with a picture of a bicycle.
Long ago, someone misanalysed modern English's vowels.
Um, no. The vowels' lengths were and are different in British RP. They are not in General American. Although British dictionaries no longer use breve-macron respellings, their IPA pronunciation keys do indicate length. A phoneme's length depends on the phonological context, so "long e" in one context may be shorter than "short e" in a different one, but will be longer if the context is the same. Pétur Knútsson's page shows these patterns by graphing the length of "long e" with "short i" [not "short e", but still].
"peat" and "pet" are equally long
Not true!
The vowel sound in "peat" is long, and in IPA is written as /piːt/; that in "pet" is short, and is written as /pɛt/.
E.
You may want to re-read the article. Johnson states that in many languages, "the macron indicates vowel length—very literally length, as in how long a vowel is pronounced." So the length means the length of the pronunciation. In English, peat and pet are both the same length in duration - meaning that if you had a precise stopwatch, there's no reason the two words pronounced by a standard English speaker (meaning just about everyone but Paula Dean), would not be clocked as the same length. Yes, the vowel pronunciations are different, but the very point of this article is that the phrases "long" and "short" (and the accompanying diacritical marks) were applied incorrectly to designated a difference in vowel pronunciation rather than actual length.
I read the article through twice, and turned my keyboard seven times in my hands before beginning my reply...
I still disagree with the affirmation that the vowel sounds in "peat" and "pet" are the same length.
The confusion was introduced by the very sloppy terminology adopted by dictionaries many decades ago, and which some American dictionaries persist in using.
The difference between a long and a short vowel is exactly one of length, not of quality, whatever Merriam Webster might claim.
I'm not familiar with Paula Dean, so I'll not comment on her diction, but I would be interested to know why she is the only English speaker to not pronounce "standard English"
E.
Paula Dean: TV cooking show host and cookbook author from Georgia (USA), known for her strong Southern (US) drawl, and hence the object of humor. There are plenty of others whose accent is similar, but she's noted for her ability to make simple vowel sounds into multi-syllabic warbles.
Of course "standard English" is a loaded phrase to begin with, since regional differences can produce varied results, but the two words in question are not especially susceptible to alteration.
A point to consider: if the vowel sounds in peat and pet are indeed of different lengths, then it would be a safe assumption to say that the meaning of either word could be altered simply by lengthening or shortening the vowel. This is not the case; the alteration (as the blog points out), is in the position of the tongue, not in the length. The two vowel sounds share a kinship, but it is not one of length.
OK, I understand the Paula Dean reference, now. I've been around the South often enough and for long enough to imagine her saying something like /'peɪjɜːt/.
But still, I think you're missing the reason why I think that RLG chose those two words.
When I'm helping people to learn English pronunciation, giving examples where vowel length (and not quality) are vitally important to the meaning, I pick the pairs "sheep" and "ship", "cheap" and "chip", "neat" and "nit", and just for a laugh "sheet" and "shit".
The dictionary compilers have been using the macron sign and using the term "long vowel" and "short vowel" wrongly for such a long time that an enormous number of people now believe that the vowel sounds of "pet" and "peat" correspond to the short and long forms of the same vowel, and that in some kind of pronunciation notation this difference is marked by a macron sign.
To accept such barbarism would be akin to accepting that we should refer to a whale as being a kind of fish, just because there are enough people around who say that it's so.
E.
Again, the pairs you give don't differ in vowel length--you even admit this in your penultimate paragraph. Indeed, it is quite common to stretch out the pronunciaiton of "shit", yet it never sounds like "sheet."
Ok, as someone noted below, they are longer. But, as in my example, you can pronounce the "short" vowels longer and it won't be confused with the "long" vowel, at least in standard American English.
Maybe I'm missing something here, so I'll try one more time. First, it seems that you initially disagreed with the blog post. Now it seems that you are in complete agreement with it.
The confusion seems to be what is meant by "long" and "short." Since Johnson specifically wrote "very literally length," it is safe to assume that he meant how long the vowel sound is pronounced - a measurable length of time. In English, and specifically in the words "pet" and "peat," there is no measurable distinction in the length of the pronunciation of the vowel. If there is any perceptible difference, it would be the briefest fraction of a second, so is not worth considering - especially in terms of the meaning of the two words. I could say the word peat as "peet" or as "peeeeeeeet" and it would mean the same thing (although people might look at me funny). But as soon as I say "peht," the meaning of the word changes, even if I draw out the vowel - "pehhhhhhht." (Again, folks might look at me funny.) As Johnson states, the meaning of the word is clearly in how the position of the tongue affects the pronunciation of the vowel, not in the length of time pronounced.
Maybe I'm missing something here, so I'll try one more time. First, it seems that you initially disagreed with the blog post. Now it seems that you are in complete agreement with it.
Not quite... I commented on two things.
When I wrote in my original comment 'the vowel sound in "peat" is long, and in IPA is written as /piːt/; that in "pet" is short, and is written as /pɛt/', I was indeed contradicting RLG's affirmation that "peat" and "pet" are equally long.
When I later made a comment about dictionary compilers' barbarisms, I was in agreement with another part of RLG's post.
E.
Look up the famous joke about the Italian in a hotel, who asks for "a sheet on the bed". I know, it is a caricature, but it is an exaggeration in order to illustrate the point.
I mentioned in my comment that I choose these pairs when helping people learn English pronunciation. Vowel length and tonic stress are possibly the two most important elements of English pronunciation.
Quite a lot of latitude is allowed in vowel quality (just compare the difference between the vowels as spoken around Boston and around North Carolina, for example). But vowel length and tonic stress are less forgiving.
E.
This is getting frustrating. You seem to disagree with Johnson about the length of vowels and then go on to use long/short in just the way he is saying is wrong or misleading. But then you also say that "dictionary compilers have been using the macron sign and using the term "long vowel" and "short vowel" wrongly." Your use of long/short to describe the difference between /piːt/ and /pɛt/ is no different than using long/short to describe the difference between pēt and pĕt. Both are differences in the vowel sound, not the literal vowel length. regardless of how an Italian pronounces the vowel sound, Johnson's point is that calling the difference long vs. short is misguided and that the use of the macron in the English language is different from every other language using the same mark.
Paula Dean speaks 'suthun' - our southern accent, where even simple words acquire more syllables. Water becomes something like wawu-tuh along the coast of the Carolinas. In some areas, 'cow' has two syllables, sort of.
But for most Americans, Peat and Pet take exactly the same number of seconds to say. This is true in general among 'American' words where the only difference is what we Americans were taught to call long and short vowels. And taught incorrectly, apparently.
"But vowel length and tonic stress are less forgiving."
Tonic stress, yes, that is not so forgiving. But vowel length is VERY forgiving.
The word /piːt/ has the exact same spelling and meaning as the word /pit/. The word /pɛt/ has the exact same spelling and meaning as the word /pɛːt/. The "sheet on the bed" joke works not because of vowel length, but because the wrong vowel sound (wrong tongue position) is used.
This is rather vindicating to see. The "long/short" vowel distinction, in the conventional sense that stems from misanalysis, made some amount of sense to me in kindergarten, but none whatsoever since I've studied linguistics.
As an American, my first instinct would be to pronounce 'Mondelez' in Spanish fashion (or at least in the way Americans tend to pronounce Spanish, which would be with a short 'e').
As a 'Murrican, I know that those silly things you call the macron and the breve and actually the 'long sign' and the 'short sign'. Macron sounds like some miracle fabric, and Breve is a detergent, probably.
Without the long sign, I probably would have pronounced it MON-de-LAY. With it, MON-de-LEE. The Z (zed to y'all) is obviously silent.
;-)
Why not "Mondeleez"?
I respectfully disagree. The macron and all phonetic transcriptions for English that are keyboard hostile will disappear in the very near future. The 'English Phonetic Alphabet' designed for English sounds using standard keyboard symbols is one alternative that is gaining popularity. Suitable for all learners it circumvents linguistic terminology and associates vowel sounds with color names for accurate pronunciation. EPA takes about a 1/2 an hour for any student to learn.
For speaking English the two most important facts people need to understand are:
1) Every word in English is a color
2) English is a stress-based language
It is almost too simple to be true but the names of the colors in English each use a different vowel sound. 16 vowel sounds - 16 colors /Ay/ in made, rain, eight, great, they...are Gray words. /a/ in 'mad', 'calf', 'laugh', 'plaid'...are Black. /Ey/ in 'meat', 'feet', 'people', 'piece', 'sk'i...are green...and so on
It is a perfect system and easy to adopt because people usually know the names of the colors. Learners remember the sound of the word by color - not crazy spelling. They can speak confidently and infinite new words fit into the framework
English is a stress-based language (stress-syllable is where the meaning is not individual sounds so much) Words have one and only one most important syllable. That syllable determines the color (and the stress - therefore the pronunciation of the word)
'congratulations' and 'explain' are Gray, 'category' and 'distraction' are Black...
Apps and dictionaries are in production through Thompson Language Center and McMaster University.
Judy Thompson
This is what was claimed for the English phonemic alphabet when Kenyon and Knott was published by Merriam-Webster in the early 1950s. Oddly enough, not even Merriam-Webster uses it.
Why single out just the macron? Diacritics in general are a minefield: café in French but caffè in Italian, à is short in French but long in Hungarian...
We have to face the fact that spelling can only ever approximate to reflect pronunciation and slippage over time is inevitable for any living language.
'peat' and 'pet' equally long? I learn something new every day.
Unfortunately not true, "peat" is longer than "pet". The "tense" vowels are consistently longer than their "lax" counterparts.
Ok, but vowel length is not phonemic in American English, right? So what does it matter?
To be fair, 'dictionary respelling' predominates only in the US. In Australia, few people can even read it. Our dictionary-of-record, Macquarie, uses IPA.
I grew up in an Anglophone country, have a linguistics degree, love English, but I've never bothered to even attempt to understand dictionary respellings. Nor do I have any idea what the 'short/long difference' in English is meant to refer to. Diphthongs are not 'longer' versions of vowels (unless you're a faaahn Southern gentleman). Dictionary respellings are a bizarre attempt to reduce the complexity of English spelling by adding another whole layer of useless abstraction on top of it. IPA or bust.
Would the someone who misanalysed modern English's vowels be the Danish linguist, Otto Jespersen?
The macron-marked "long vowels" of Noah Webster's transcriptions represent vowels that used to be phonetically long in Middle English. Long vowels (and not short vowels) changed their pronunciations in the Great Vowel Shift that attended the switch from Middle (Chaucerian) to Early Modern (Shakespearean) English.
The spelling didn't change with it, though, because Caxton had fixed it at Middle English a century earlier. That's why the English names for vowel letters (which are the "long" vowel) are so different from other European languages'. Webster's system isn't really bad for the eighteenth century, given that modern linguistics was mostly developed a century later.
I haven't seen Webster's steampunk phonetic system in use anywhere but in the United States, though. The rest of the world (and all bilingual dictionaries) automatically use the standard IPA phonemic notation to indicate pronunciation.
Why do I have this urge to break into "On the road to Mondelez..."?
Great post. I always love reading Johnson, which is why I hate to be this person, but I can't help myself... aargh, please remove the apostrophe from that "it's" in the second-to-last paragraph. I know you didn't intend it.
Of course, and thank you. It's fixed.