Shibboleths

The hallmark of being an Israelite was being able to pronounce the “sh” sound. For American and British English, it seems to me, it’s being able to pronounce “the” or the short “i.” What is it in other languages?

When the Israelites under Jephthah wanted to tell friend from foe, they famously asked those claiming to be on their side to say “shibboleth.”  Those who couldn’t—who mangled it as “sibboleth” instead—weren’t Israelites (Judges 12:6).

In English there are a few such hallmark sounds that non-native speakers find famously difficult, but we don’t talk about them much.  I’m not referring to sounds that cause problems for native speakers of particular foreign languages, the way Japanese speakers find it difficult to tell “r” from “l,” or Germans have a problem with the “w” sound. I mean sounds that tend to baffle all non-native English speakers, with few exceptions.

On reflection, I think two are the most obvious.  One is the short “i,” which seems to exist in few European languages.  (This actually separates Americans, Canadians, and speakers of standard BBC English from natives of Australia and New Zealand: Aussies proverbially say “feesh and cheeps”; Kiwis, “fush and chups.”) The other is the “th” sound, which sounds in many other languages like “z.”  I’m not saying that no other languages possess these sounds. Greek obviously has a th, with its own letter to go with it, and if memory of a long-ago class in the rudiments serves, Russian has a short “i.” But in general, these are good shibboleths in standard English. Ask someone to say “this is a bit thick,” and you’ll separate the native speakers, and excellent foreign students of the language, from almost everyone else.

Am I right?  Have I missed some obvious ones?  And, to flip things: are there famous sounds in other languages that cause their speakers to laugh at almost all foreigners for failing to pronounce them correctly?  (The nasal vowels in French sound likely, as does the “r” in German, which isn’t rolled as “rr” is in Spanish but produced near the back of the throat.) Aspiring sentries want to know.

Author: Andrew Sabl

I'm a political theorist and Visiting Professor (through 2017) in the Program on Ethics, Politics and Economics at Yale. My interests include the history of political thought, toleration, democratic theory, political ethics, problems of coordination and convention, the realist movement in political theory, and the thought of David Hume. My first book, Ruling Passions: Political Offices and Democratic Ethics (Princeton, 2002) covered many of these topics, with a special focus on the varieties of democratic politics and the disparate qualities of mind and character appropriate to those who practice each of them. My second book Hume's Politics: Coordination and Crisis in the History of England was published in 2012; I am currently finishing a book on toleration, with the working title The Virtues of Hypocrisy, under contract with Harvard University Press. A Los Angeles native, I got my B.A. and Ph.D. from Harvard. Before coming to Yale I taught at Vanderbilt and at UCLA, where I was an Assistant, Associate, and Full Professor; and held visiting positions at Williams, Harvard, and Princeton. I am married to Miriam Laugesen, who teaches health policy and the politics of health care at the Mailman School of public health at Columbia, and we have a twelve-year-old son.

25 thoughts on “Shibboleths”

  1. How many (if any) phonemes are unique to English? I don’t know (though I suspect the number may be zero), but to stay closer afield: few other European languages seem to have the a in “flat.” The English r in its British, American, and Oceanic variants is also unlike the trilled (Italian, Spanish) or uvulur (French, German) rs found in most European tongues. The penchant of English for dipthongizing most long vowels is another departure most other European languages, though I don’t know how hard these long vowels are for a foreigner to reproduce.

    The voiced pharyngeal scrape in Arabic (represented by the letter ‘ayn) and some other Semitic langugages is notorious for being hard for non-Arabic speakers to utter.

  2. I’m not sure if it’s related to the “voiced pharyngeal scrape” ‘ayn Aaron mentions, but the chet in Hebrew can be unusual and difficult to English speakers.

    I thought the “th” sound in English was also supposed to be unusual, and the o-umlaut sound in German was a bit hard for me, as a native English speaker.

  3. I swear I heard of one such distinction made between Germans and the native Dutch speakers during WWII, some city name I think.

  4. MobiusKlein: Scheveningen. It’s a coastal town near Den Haag. The Dutch pronounce the initial consonant cluster as “Skkhh”, and almost drop the last syllable. Germans can’t cope.

    English speakers (and speakers of a lot of European languages) have trouble with the affricates in Chinese and Japanese. For example ‘tsunami’ is usually pronounced in English either as ‘sunami’ or with the ‘ts’ broken up into two separate sounds.

  5. When I was learning Danish, the phrase “rødgrød med fløde” (red porridge with cream) separated the men from the boys. I got pretty proficient at it (and at Danish, but by now it’s long gone): the “r” sounded like a gutturalized Hebrew chet, and the “d”s were all soft “th”s. Norwegians and Swedes used to compare Danish to a throat disease.

  6. On the short “i”: if you heard me say “this is a bit thick”, you would find nothing un-American in my pronunciation, even though I was born in Greece and learned English when I came to America at age 9. (I must note that anybody who says “That’s a bit thick” and is NOT talking about a shim or a syrup sounds British to me.) The Greek immigrants I know who DO have recognizable Greek accents all immigrated to the US when they were 12 or older. So I have long believed that something related to vocalization in the brain congeals at around age 10-11, at least as far as the Greek-to-English transition is concerned. I wonder whether that holds for other language pairs.

    By the way, ask a Greek who learned English after age 12 to say “chair”, and it’s a very good bet you’ll hear “tsair”. The “ch” sound is not found in Greek, so far as I know. Curiously, a sound very like it IS found in the rural Cretan accent which most Greeks find rustic at best and barbarous at worst. The modern Greek word for “head” is “kefali”, pronounced with the “k” of “key” or “keep”. An old Cretan will pronounce it something like “chefali”. Pronouncing a leading “k” (actually, “kappa”) as “ch” like in “chair” comes naturally to him, but he will STILL say “tsair” when speaking English. Of course, the Greek root word for “head” is recognizable in “cephalo-pod” or “en-cephalo-graphy”, so who the hell knows what the RIGHT pronunciation is?

    –TP

  7. Hate to sound skeptical, but I have to go there on two counts. First, doesn’t the German short “i” sound an awful lot like the English? Similarly with the Yiddish short “i”, now that I think of it.

    Second, I can’t buy the Aussie transliteration as “feesh and cheeps.” It’s definitely a different sound than the American short “i”, but if it was rendered in the International Phonetic Alphabet I can’t believe it’d be in there with the German “ie” or the Italian and French “i”. It isn’t held nearly as long and doesn’t require the same widening of the mouth.

    When I’ve heard other English-speakers mock American English, they’ve tended to pick up on the nasal tonal quality we typically have and on the diphthongizing Aaron Baker mentions. Those may be the most distinctive aspects of English as she is spoken in American.

  8. ‘Aussies proverbially say “feesh and cheeps”’

    Well that’s what Kiwis say, and to someone with a fush and chups accent that’s probably what it sounds like, but it’s actually the same sound as in RP and General American.

  9. Chess-players know this particular German tongue and spelling twister that separates the native speakers from the not (and the Austrians and Swiss from the Germans, for that matter):
    zugzwang (being forced to move although any move is disadvantageous)

    Non-native speakers either overemphasize the initial stop-sibilant (TTTsoog) or leave the stop entirely out of the middle consonant grouping (ghtsv). Or both.

  10. In German, “ch” (both versions) and “ü” are probably good candidates. (Scots get a pass on the “loch” version of “ch”, of course.)

    Personally, French has caused me no end of grief, and I am in the weird situation that I am technically a native speaker of German and English, but speak English with a slight German and German with a slight Midwestern accent that I’ve never managed to disguise.

    C.E. Petit: Try this. “Zwischen zwei Zwetschgenzweigen sitzen zwei zwitschernde Zwergzeisige.” 🙂

  11. Hi folks,

    This is fun.

    Katya: I left out the hard German “ch” (which I now find is called a “voiceless uvular fricative,” http://en.wikipedia.org/wiki/Voiceless_uvular_fricative) precisely because it’s common in Scotland (and of course in Hebrew–where it’s the modern pronunciation of chet, *unlike*, I find, the older-Herbrew and Arabic pronunciation, which it seems is used by Arabic speakers “as a shibboleth” [sic: http://en.wikipedia.org/wiki/Heth%5D). But maybe those are obscure exceptions in the scheme of things. The soft “ch,” on the other hand, seems very hard for this foreigner anyway, and after a fair amount of time in Germany, where I acquired an otherwise decent accent, I’m sure I never got it. The German ü sound is probably very hard if you’re not taught it as a child by a native speaker (as I was).

    Tony P.: You’re right that “a bit thick” sounds non-American. “A little thick” would be more U.S. in tone (but applied only to the consistency of a substance, not to something’s or someone’s being stupid, which is not a meaning “thick” carries here).

    Mark Barton: I have to differ, I’m afraid—with respect to an old-fashioned, rural Aussie accent at any rate. (Educated urbanites speak something closer to RP a.k.a. BBC, sort of.) Altoid is right that it isn’t a drawn-out “ee,” exactly,” but it sure isn’t an RP short “i” either: more like a Spanish “i,” not drawn out but definitely closer to rhyming with “tweet” than with “twit.”

    Altoid: you’re right about the German “i.” (In fact, “a bit” comes from the German “ein Bisschen,” the diminutive of “bite.”) Can anyone weigh in regarding other Germanic languages, and perhaps Slavic ones? Are Romance languages perhaps the outliers in *not* having this sound?

  12. It’s not often I get to participate in an RBC discussion, but as linguistics undergrad I couldn’t resist taking a break from my paper to add my two cents.

    About the German “r” or the uvular trill [Ê€] in IPA. I can’t speak for Austrian or Swiss German, but I know that in Bavaria they pronounce their “r’s” similarly to how they’re pronounced in Spanish or Italian. I learned this while living with a Bavarian in Spain, she could pronounce the Spanish “r” without a problem, but her German friends pronounced the uvular trill. The band Rammstein seems to always pronounce their “r’s” this way.

    Adding to what Aaron Baker and Warren Terra said sounds far back in the throat, epiglottal, pharyngeal, uvular, and velar sounds will generally be notoriously hard for English speaker. The chet and the ayn are both pharyngeal, though according to wikipedia the chet in modern Hebrew is almost always pronounced as a uvular fricative due to European influence, most likely German spoken by Ashkenazi Jews. Also, wikipedia says the ayn is a shibboleth for distinguishing Arabic speakers from non-native Arabic speakers.

    Russian doesn’t have the our short “i” [ɪ], it does have a vowel that’s very similar, the [ɨ]. It’s pronounced between the our “eeee” sound and our “ooooo” sound. I don’t really know about other Slavic languages. I know Brazilian Portuguese has our short i, but I haven’t found any other major Romance languages that have it or any of our other lax vowels. I think Mr. Sabl is right about Romance languages being the outliers in this situation.

    Other shibboleths that I can think of off the top of my head are vowel tones,European languages don’t use them but I’m willing to bet speakers of Southeast Asian languages can distinguish a Mandirin, Cantonese, Thai, or Vietnamese speaker from how they pronounce tones in the target language. Clicks, languages tend to borrow clicks very readily, but borrowed clicks are very simplified compared to languages that had them “originally”. The final sound in native pronunciation of the language Nahuatl, a sound that is very similar to the one found in Icelandic. Remember the volcano Eyjafjallajökull? The double “l” is trickier than it appears. Finally, r-sounds and l-sounds in general are hard for a lot of people, because most languages tend to do, if not unique, rare things with them.

    Back to my paper.

  13. My mom was from Scheveningen. And I grew up with the story about how that name always gave German spies and such away. It is now part of Schavenhaag(sp?) legally having merged with Den Haag. Thomas’s point about dropping the last consonant seems correct. Never having learned to spell in Dutch I didn’t know it had a consonant ending. But I don’t like the ‘kk’ in “Skkhh” as a way of getting across the sound of the beginning. It’s softer than that. (But perhaps Thomas is using a linguistic convention I’m unaware of.) Of course I want to put in a ‘g’, but it is the Dutch g-sound that I’m thinking of. I don’t know an English equivalent. Fun to see this!

  14. “Those who couldn’t—who mangled it as “sibboleth” instead—weren’t Israelites (Judges 12:6).”

    Actually, both sides in that war were Israelites. This was the period of the Judges, when the Israelite tribes were not unified under a single kingdom. The Gileadites (a subgroup of the tribe of Menasseh) had defeated the Tribe of Ephraim and captured the Jordan River crossing points, and used the “shibboleth” test on the fleeing Ephraimites as they reached the fords. Both GIleadites and Ephraimites were native speakers of Hebrew.

    The point being that a “shibboleth” is a test of people who are very similar and can be distinguished only on specific and typically trivial points of difference.

  15. Many years ago, in more easy-going times, I was returning to the US at Tijuana and, being of a racial and ethnic mix that’s perhaps more common in Spanish- than English-speaking lands, attracted the intense attention of the Customs officer. He gave up, however, and waved me through, when I told him (correctly, and correctly pronounced) that I’m from Chicago. I’ve never heard anyone from a Latin background who learned English after puberty say that name correctly, and presumably neither had he; not only the short i, but also the meter (anapest?) eludes them.

  16. I think US-English speakers have trouble pronouncing any “r” that’s not their own – the Spanish trill, or the French roll, or the Hebrew palatalized, or the Japanese…whatever the hell that is – but they don’t have problem hearing and distinguishing it. On the other hand, I’ve met very few English natives who could distinguish between the aspirated and nonaspirated versions of unvoiced consonants, such as the initial sounds in French “haricot” vs. “homard”.

    As long as we’re talking about Hebrew, you know what drives me totally batshit about learning it? Here’s a language whose verb tenses, and the trasnformation of verbs into other parts of speech, are marked almost entirely through vowel shifts…and yet it is written WITHOUT VOWELS. Arg. Or “oy-va-voy”.

  17. I would guess the initial consonant in Qatar must be pretty unusual outside of Arabic. I’ve heard that it is somewhere between a K and a G in English, but I can’t imagine what that would be like and I’ve never been able to catch on from hearing Arabic speakers.

    English-speakers from India often have trouble with the English V (as in “I’m a wery bad wegetarian.”) Those who live abroad long enough become self-conscious about it, but more than a few times I’ve seen the main payoff from their efforts is a Teutonic tendency to V the W (without altering their tendency to W the V).

  18. My favorite version of the story:

    guard: “Are you an Ephraimite?”
    stranger: “No.”
    guard: “Say shibboleth.”
    stranger: “Sibboleth.”
    guard: “Ah ha!”
    stranger: “Oh, sit.”

  19. @davidj: Colorado is even worse than Chicago.

    Swedish has a sound “sj” that can only be approximated by English speakers as “h”. It’s really weird sounding. If you meet a Swede, ask them to say seventy-seven starfish in Swedish. I knew an unlucky Irishman whose Swedish social security number was full of 7s, and so just after moving there, his first job was to say this unpronounceable number to people over and over.

  20. The short “i” sound in “hit” or “sit” is not that uncommon in European languages. The fact that people who have learnt English as a foreign language tend to get their vowel sounds mixed up may not be so much that their native language doesn’t have these sounds but rather that they haven’t gotten English vowels figured out yet. On the other hand, English does have rather a lot of vowel sounds, somewhere between twelve or thirteen up to twenty or twenty-four if you count “R-colored” vowel sounds, and even more irritatingly, they’re completely different depending on dialect. The one my students seem to have the biggest problem with is the U sound in “dug”. A fairly tricky consonant sound that certainly exists in other languages but is hard for English learners to get is the L in “pool”. If you say that word, you’ll notice that the L at the end isn’t the same as the L in “lemon”.

    Czech actually does have a very good shibboleth, the Ř sound. It’s something like a rolled R combined with a “zh”, and needless to say, it’s very rare, and Czechs are convinced that no other language in the world has it. I have no idea if this is true. If the Czechs ever went to war with the Slovaks, this would be very useful, since despite knowing many Slovaks who speak good Czech, and despite the fact that Czech and Slovak are only different languages in the “a language is a dialect with an army and a navy” (although neither of those countries have much use for navies) sense, I haven’t met a single one who can pronounce Ř properly.

  21. “The fact that people who have learnt English as a foreign language tend to get their vowel sounds mixed up may not be so much that their native language doesn’t have these sounds but rather that they haven’t gotten English vowels figured out yet.” Yes, absolutely, and they have that trouble with both vowels and consonants because of interference between native and second languages, which I believe is much-studied in second-language acquisition. There are clear patterns of interference between language pairs not only in sounds but in grammar. It’s the basis of dialect comedy.

    One contrastive example I’ve noticed relates to a sound mentioned above: if you listen to native Hebrew-speaking Israelis interviewed about Hamas, they’ll almost invariably pronounce it as “chamas” (like in “challah”) even though if you listen to Arabic-speakers, it’s clearly much closer to an ordinary aspirated English “h”, a sound that in fact exists in Hebrew. So I wondered why not “h”, and what I was told is that Arabic has three related sounds and this particular one is the middle one. Hebrew has only two related sounds and to Hebrew-speaking ears this one is closer to the “ch” than to the “h” in Hebrew, so they render it as “ch”.

  22. My two favorites (primarily with people whose native tongue is Spanish)

    Mosque v. Mask

    Beach v. Bitch

    So I guess my example would be to ask someone to say, “A mosque next to a beach”

  23. English-speakers (Americans, at least) have a hard time making long vowel sounds in unaccented syllables – for us, they tend to turn into schwa (either eh or uh).
    I read somewhere that during WWII, Italian women who consorted with GI’s were derogatorily known among Italians as segnorinas, because the Americans couldn’t quite say the “ee” sound of the initial “i” in signorina.

  24. Polish has the short i, written ‘y’. It also has nasal vowels: ‘Ä…’ and ‘Ä™’. The Polish sounds that I haven’t encountered in other languages are the double versions of sh, ch, and zh: sz/Å›, cz/ć, and rz/ź which are harder and softer respectively than the English versions.

Comments are closed.