Toby Morris has an amusing piece for The Spinoff (NZ/Aotearoa) about a common problem in reproducing Māori words typographically: the macron doesn’t come out right. It’s in comic form, so I can’t easily quote from it, but I’ll type out one “speech balloon” to give you an idea:
It can be complicated though: Tainui in the Waikato prefer double vowels, and many older speakers grew up not using macrons at all.
Check it out! (Via MetaFilter, where you will find discussion of computer fonts, the Hawaiian ʻokina, the Ming Kwai typewriter, and other vaguely related matters. Cf. also The World of Ā.)
Macrons are key to the proper spelling of Japanese in Hepburn romaji, but in practice they often get omitted. In in-Japan contexts where the reason for the omission is not that the Japanese-speakers lack cultural or economic power in the situation. Obviously the fact that the Japanese-speakers have their own “proper” scripts to use (and thus may be less emotionally invested in romanization being done “properly”) is one factor distinguishing the situations. But again (just as in Japan!) there are apparently multiple rival romanization systems extant in Aotearoa. So it’s one thing to be able to accommodate macrons technically, but making them a stylebook requirement may mean taking a side in some unresolved factional fight you may well not fully understand.
I remember handouts in primary school Irish lessons were mimeographed from a double spaced stencil typescript into which outsize fada accents had been scratched by hand.
I often see oh in the names of Japanese authors.
…and had to look up weka: yes, two short vowels. Phew!
It’s been many years (20+?) since someone thrilled the internet classicists’ mailing list by telling them they could save time putting macrons on Latin texts by switching to the Maori keyboard.
I’m afraid I never remember to do that, so I just put them in with ‘insert character’ or, if there are a lot of them, type the long vowels with acute accents, then do a quintuple search-and-replace. The only difficulty is long Y, which is found in a few Romanized Greek words (e.g. Lydia) and I’ve never been able to find in the Times New Roman insert-character display. I think I had to go to the Wikipedia macron article (disambiguated from the French politician) to copy and paste it, last time I needed one. Or just remember which of my already-macronized text files has a long Y I can copy.
Possibly worth noting for the international audience: “stink” is slightly outdated NZ kid’s slang for bad and/or unfair. You failed your maths test, and you should have passed, and that’s stink. Everyone else got good presents from the lucky dip but you got the stink one. We’re not having fish and chips tonight after all? STINK.
Beyond that, a word variously spelled “aye”, “eh”, “ay” but pronounced to rhyme with “day” is a normal tag in speech. It would be idiomatic, if someone says their name is consistently mis-spelled, to commiserate “That’s stink, eh.”
Thanks, I had vaguely wondered about the use of “stink”!
An alternative for Latin (and Old English) is just to use acute and pretend it’s an apex (which is not in Unicode), as Vicipaedia does. (Shouldn’t that be Vicipaideia, anyway?) It’s not like you need more than one mark, as you do for polytonic Greek. If Romanian emails used s-cedilla and t-cedilla instead of the proper comma below, what are the odds?
No; the Ancient Romans did not transliterate, they wrote down what they heard.
(And the Romanian under-commas were emergency workarounds for missing cedillas until the Academy decided the other way around for no good reason.)
by switching to the Maori keyboard.
Yes that’s the only way I can get mācrons for all the vowels/both cases. For comments in the kūmara thread, it’s been fiddly constantly switching between Māori/Polynesian and accented Latin for Cañari, Puná.
I notice PolLex uses doubled letters.
It’s in comic form,
Toby Morris is a total star. All through the pandemic, he liaised with public health expert Siouxsie Wiles to produce ‘explainers’ of all the complex issues. Get the boxed set.
Oh, to add to @Stephen’s context (thank you) — this was so obvious it didn’t occur to me: this weekend is the public holiday ‘Waitangi Day’, which marks the 1842 signing of the Treaty between The Crown and (most) Māori.
So the media have a focus on culture/language issues. Only most journalists/media (incl Toby) are Pākeha, so have to beware cultural appropriation.
No; the Ancient Romans did not transliterate, they wrote down what they heard.
That was a joke.
And the Romanian under-commas were emergency workarounds for missing cedillas until the Academy decided the other way around for no good reason.
Do you have a reference for this? My understanding is that comma below was first introduced in 1825 and stabilized in 1860 (the conversion from Romanian Cyrillic to Latin went letter-by-letter rather than all at once, so that Ш and Ц were retained side by side with Latin forms like D and G for a while). The cedilla forms first appeared in the first edition of ISO 8859-2 (1987) as a result of a misreading of Romanian requirements.
T-with-comma is historically short for tz; it’s a t with a zed-illa. Renderings as tz in French and other “western” languages were common in the 19th and early 20th centuries.
Tristan Tzara!
Interesting, I never thought of that. Wikipedia:
If the last sentence is true (the reference is a dead link), it’s a sad commentary on the excesses of Romanian Francophilia.
As an aside for any Windows user who struggles with writing accented letters in general, the latest version of Microsoft PowerToys contains a tool called QuickAccent that makes it much (much!) easier to write accented letters. Just press the base-letter you want and (while keeping it pressed) press space. By repeatedly pressing space, it shows a list of accented characters and it’s possible to select the one desired. Much easier than memorizing the ASCII or Unicode codepoint!
https://learn.microsoft.com/en-us/windows/powertoys/quick-accent
I often mix English, French and Swedish, and having to find those ‘å’ and ‘ô’ on a US-keyboard laptop is a mess. With the latest version it’s even possible to get sub- / super- script numbers (but not (yet) letters), Greek characters and sundry other signs.
“Racovitza (as he wrote his surname in French)“
Possibly already widely known, but in Word you can get some common diacritics by typing Ctrl + [character resembling your diacritic], releasing, then hitting the base character. So Ctrl-apostrophe followed by e gets you e-acute, Ctrl-caret then [a] gives you a-circumflex, etc.
Acute: apostrophe
Circumflex: caret
Cedilla: comma
Dieresis: colon
Grave: the backward apostrophe character (to the left of the key usually), but this is the odd one out because you hit it alone, no Ctrl. (That particular key behaves strangely for reasons John Cowan may understand).
(NB: most of these require hitting Shift alongside Ctrl, of course.)
Nothing for macron, unfortunately.
My wife keeps a Word document on her desktop containing nothing but the accented characters she needs and copies and pastes from it.
LH, 4th comment back (11:48am):
“Sad in the country” where sad is ‘Trist’ and the country is Romania? Is that at least partly an allusion to Ovid, who wrote his Tristia (=”Sad Things”) in Romania after Augustus relegated him? (Specifically in Constanța – there’s that ț again.) Or does Tzara’s “country” mean ‘rural area’ rather than ‘nation’?
Ah, an interesting suggestion! I like it, but I don’t know enough to have any idea if it’s plausible.
i like the ovid connection, which makes a certain kind of sense to me! all the more because i’m tempted to see the “țara” surname as a calqued nod to yiddish “amorets” [fool / dunce / uneducated person] < "am ha-arets" [person of the land]. that high/low multilingual double reference seems very like tzara and his crew.
“Possibly already widely known, but in Word” – there is also an “international” keyboard layout in Windows, where typing “a , ‘a, `a, ~a, ^a, ‘c gives you ä, á, à, ã, â, ç respectively. Still without macrons.
Try the PowerToys QuickAccent! It provides all the macronated characters, as well as ogonek and greek. (But not (yet) IPA or cyrillics.) And it works in all applications that take input from the keyboard.
I also considered installing https://en.wikipedia.org/wiki/AutoKey / https://en.wikipedia.org/wiki/AutoHotkey but I’m lazy.
The Windows Moby Latin keyboard driver is for the U.S. physical keyboard and supports about 1000 characters without preempting any keys except AltGr; it is reasonably mnemonic. Its variants Whacking {John, Taffy, Sandy, Mick} are for the UK physical keyboard and are meant for {English, English+Welsh, English+ScotsGaelic, English+Irish}.
The OED re-revised Māori, n. and adj. in March 2023 as part of its New Zealand-focused update this quarter, and I’m grinding my teeth once again at the lack of change-tracking, because now it’s impossible to know whether they had the macron in the headword in the original Third Edition revision of September 2000. (It wasn’t there in the old editions.) I suspect that they didn’t, and that this comment is new in 2023:
The qualification “when referring to the people or the language” applies since there are also compounds referring to animals/vegetables/minerals of New Zealand, e.g. maori wrasse (fish), Maori chief (fish), Maori onion, etc.
Yes, I can go look up publications by OUP and other publishers from 2000 to now to investigate the history of this spelling in detail. But the OED itself is part of that history, too, and they’ve just erased it.
Another typographical note: in the OED’s further comment on the 19th-century spellings Maodi, Maoude,
… presumably that’s supposed to be /ɾ/. They sometimes have font problems with IPA symbols; see also razzia, where they mention “the standard French voiced uvular fricative /#/”.
Poetry Foundation features a lot of poets from other cultures who write in English, and the diacriticized letters in their names or words often appear in a *slightly* different font — not anywhere near as bad as the ones in the Toby Morris comic, it often takes some magnification to spot it, but now I can’t unsee it. The ā has a very subtle mismatch (I missed it at first) in an essay by Māori poet Robert Sullivan. It’s a bit more noticeable with Vietnamese diacritics: the word nghệ ‘turmeric’ is discussed by Vietnamese poet Vi Khi Nao, writing about the Vietnamese poets who influenced her, but all the Vietnamese-specific accented letters, such as ệ, come out in a font with slightly different letter sizes or shapes (more noticeable in some browsers than others).
(I’m not blaming them, they’re probably doing about as good a job as anyone can in balancing a good overall page design with including all possible letters. The fallback fonts really are very close to the main one.)
That’s a shame. I agree about the Vietnamese fallback fonts, but they really should be able to do better with the ā.
Thanks for pointing that out @kt, ugh! I’m so used to seeing the macron fitting in smoothly (for example, all NZ newspapers and magazines use it nowadays), that really smacked me in the eye.
More awkward seem to be the upper-case forms. Quite often newspapers miss out the macrons there. (Fortunately they’re not often needed — so my “quite often” means within the rare occasions.)
Randall Munroe’s What If pointed out that liters per km is equal to mm², and can be visualized as the cross-section of the string of fuel consumed by the engine.
…It is.
+ 1 level in galaxy-brain meme
Another silly little thing that I can’t unsee: The American Heritage Dictionary’s web version has a very rare Stink A problem (it’s fine in print and in the iPhone app), affecting only the boxes for Word History, Usage Note, etc., and only in italics. Actually, it’s a reverse Stink A: the ā is fine, the letters *without* diacritics aren’t quite right. The difference is that in their body font, as with book fonts in general, the letter a has the “double-story” form in regular type, but the rounded script form in italics. (“Double-story” is the kind of a you see in the default font at languagehat.com; the rounded form is like ɑ.) But something in their stylesheet apparently disrupts the italic font in the Word History boxes, since the browser falls back on the cheap substitute of just slanting the regular letters — but letters with diacritics are surrounded by font tags that protect them, so they have proper italics. (Same for all 3 browsers that I can easily check.)
Probably the only page where this matters at all is orange, where the Word History box puts several of the ancestral words in italics: Tamil nāram, Sanskrit nāraṅgaḥ, Persian nārang, Arabic nāranj. All those words have a plain a and a macron-a in the same word, and they look different! There’s also a visible difference between the plain n and the dotted n in the Sanskrit word.
Top-notch etymology, too bad the typography makes it look very slightly sloppy.
Oh great, now I can’t unsee it either.
Firefox to the rescue – everything on that page looks as intended to me.
On Android it has the problem for me in both Chrome and Firefox. Things look fine in the etymology in square brackets immediately after the definition, but not in the Word History section.
Now I can’t decide whether to use on or in with an operating system.
David M, I see the problem in Firefox (Windows). Are you sure you looked at the Word History paragraph, not the etymology?
If I save the page to my computer and open that in Firefox, the Word History paragraph now has correct italics. That make me think the problem is with the site’s stylesheet, which the local copy can’t access.
I once fought with a different issue with someone’s fonts (the site actually relied on web fonts but the developers outsmarted themselves by downloading a pre-release unhinted (the effect was different from what I see in pictures in the article: additional black pixels adjacent to letters rather than blurring) version of Roboto, so the goal was making the browser use a system font or else use a different format for the web font, where the file offered by developers contained a hinted version from the official release) and was not able to untangle browser and operating system dependencies.
Oh… yes. The etymology looks right, the word history looks wrong. Yes, Windows.
The orange page looks fine to me in Chrome on Windows, even when I crank the magnification up to 500%.
No magnification is needed to distinguish a double-story a from a script-style one, so I suspect you made the same mistake as David M; I just checked in Chrome/Windows, and the problem is in fact there too. On the AHD orange entry, look at the phrase “alteration of Arabic nāranj” in the etymology: see the difference between the double-story lowercase a’s in “alteration” and “Arabic” and the script-style a’s in “nāranj”? That’s how it’s supposed to be. Now look at the word “nāram” in the Word History box: the ā is script-style, the a is double-story, within the same word. That’s where it looks weird.
(I’m a little surprised the Word History box didn’t explain the loss of the initial n via rebracketing; that’s a popular fun fact! They do have a box about it at umpire. Several languages borrowed the word for orange before it lost the n, e.g. Hungarian narancs.)
Yes, I was looking at the etymology. I copied the Word History paragraph into LibreOffice Writer to see what the fonts are set to. The italic letters other than ā have a font setting of “Minion New;times;clean;serif”, whereas the ā is set to “Minion New Italic;Bookman;URW Bookman L;Palatino Linotype;serif”.