ON MANY’S THE NIGHT.

John Patrick Shanley, the playwright (best known for Doubt), has an op-ed piece in yesterday’s NY Times about a visit to his ancestral Ireland that perhaps tries too hard to be Irishly eloquent—but never mind that, the bit I want to discuss comes in the second paragraph:

I am not Irish. I am Irish-American. Some say I have the gift as well. If I do, it is because I listened to my father and my uncles and some of my aunts as they gave as good as they got in my living room in the Bronx. On many’s the Saturday night, they would drink rye and ginger ale, and smoke and talk and sing and dance, and I would sing, too, and dance with my aunts, and listen through the blue air. And because I listened to so much talk and so much music, perhaps I was spared somehow from the truly unfortunate fate of being an uneloquent Irish-American.

“On many’s the Saturday night” sounds completely wrong to me; it has to be either “On many a Saturday night” or “Many’s the Saturday night.” But I’ve long since learned not to trust my own intuitions about a various and changing language, and I’m curious how it strikes the Varied Reader. If you are Irish or Irish-American, I’ll be particularly interested in your reaction. And may the wind be always at your back!

Comments

  1. This eloquent Irish-American thinks that “On many’s the Saturday night” is (as a Norwegian-American friend of mine eloquently says) “wrong from every conceivable point of view.”

  2. Well, it got though an editorial process, so there must be at least two people who think it means something. Maybe it’s an attempt at elevated diction. Or inebriated diction.

  3. David Marjanović says

    Or nobody noticed. The things that get published in scientific papers sometimes… *headshake*

  4. mollymooly says

    You might have got a more natural reaction from me if you hadn’t primed me with the key phrase in the post title. My feeling is that it’s affected rather than merely colloquial; I wouldn’t go so far as to dismiss it as stage-Oirish or pure fakery.
    Google books says that the sequence “for many’s the” occurs four times in Christopher Nolan’s “The Banyan Tree”. I’m not sure how fastidious he was about the accuracy of the dialect.

  5. mollymooly says

    Och! Inver Bay on a harvest day,
    And the sun goin’ down the sky;
    When with many’s the laugh the boats put off,
    And many’s the merry cry.

  6. Ah, that does make it sound like an actual thing. Thanks!

  7. (Although that actual thing I take to be a contamination of the two phrases I mentioned, but it’s not purely a personal contamination.)

  8. I’ve an Irish gene or two, but it’s the grammarian in me that doesn’t want a sentence to be introduced by a preposition. Making the relative a little more explicit, “Many’s the night that we did that”, makes it clearer that the main subject is “many”, the main verb “is”, the predicate “the night that we did that”. Although it doesn’t quite work to it here, if one were to make “on” explicit, it wouldn’t pop out at the beginning of the sentence, but rather before its object “which”: ??”Many were the nights on which we did that.”

  9. marie-lucie says

    GLee: many’s the night …
    “Many” is not the subject of “is”, “the night [that …]” is the subject. This is a case of inversion, as also in “Home is the sailor” or “Here comes X” (although “many” is not the same type of word as “home” and “here”).
    The reason that “On many’s the night” is not standard English is not because it starts with a preposition but because a preposition cannot be followed by a complete clause (one with a verb) like “many'[i]s the night …”. A phrase starting with “on” should be followed by a noun phrase such as “many nights” or “many a night”.

  10. m-l is right. I like her reasoning.

  11. John Roth says

    @marie-lucie
    I think Geoffrey Pullum would disagree strenuously with the notion that a preposition cannot have a clause as its complement. He sets that position out in the Cambridge Grammar of the English Language, and credits the insight to Otto Jesperson (1924).
    Whether the specific preposition on licenses a clause as its complement is a subsidiary question that I can’t answer. I can’t offhand think of a case, but then I don’t keep huge lists of this kind of thing.

  12. As a part Irish-American who lived in Boston for many years, and often goes drinking with Irish friends to this day, I agree with Hat and Cowan 100%.

  13. David Marjanović says

    it’s the grammarian in me that doesn’t want a sentence to be introduced by a preposition

    Confuses me.

    Jesperson

    Jespersen, a Dane.

  14. mollymooly says

    Hiberno-English uses the definite article in lots of idiosyncratic ways; see e.g. section 4.1 here for some. I guess this is why “many’s the” is being preferred for the job standardly done by “many a”.

  15. Many’s the time I’ve heard “many’s the night”: “on many’s the night” never.

  16. I don’t see anything particularly Irish about “many’s the” or “many a”.

  17. mollymooly says

    Standard English allows
    (1) a sentence of form
    many’s the + NP + WH-clause
    or
    (2) a noun phrase of form
    many a + N
    but not
    (3) a noun phrase of the form
    many’s the + N
    It’s (3) that seems to be particularly Irish.
    It’s most obvious when (3) occurs after a preposition, as in “on many’s the night”, “with many’s the laugh”, “for many’s the day”. But I would imagine non-Irish people would find the following direct-object equally unacceptable:
    – I’ve eaten many’s the steak in that restaurant.
    It’s trickier to construct an example in subject position; e.g….
    – Many’s the man would disagree with you
    …might be analyzed as eliding the relative pronoun:
    – Many’s the man would who(/that) disagree with you
    …which is a different colloquial construction, not specific to Ireland.

  18. David Marjanović says

    The Gattò thread is closed, but I was asked for a reference about the hypothesis of tones in Proto-Indo-European. I hadn’t heard of Guy Jucquois, but Google tells me he wrote an etymological dictionary of Hittite…
    My sources are this Wikipedia chapter, evidently written by a Russian, and the pdfs the references section links to.
    Of these, the second is by S. L. Nikolayev, the Slavicist and Caucasianist. It’s from 1989, in splendid isolation behind the Iron Curtain, and mostly boring (the lists of examples must amount to half the vocabulary of Lithuanian), and in Russian – my Russian isn’t quite good enough to understand all of it. Anyway, most of the paper is about comparative Balto-Slavic accentology; then it suddenly (or so it seems to me) talks about “primarily dominant”, “secondarily dominant” and “recessive” morphemes determining (at least in part) the position of the pitch accent in Balto-Slavic (it’s on the first dominant morpheme); and then, in the somewhat misnamed conclusions, it returns to the topic of the introduction, which is the fact that the position of the stress in Greek and Vedic Sanskrit sometimes doesn’t line up (for no discernible reason). A table is presented: when both root and suffix are dominant, Vedic Sanskrit has oxytones, but Greek has barytones; when both morphemes are recessive, both languages have oxytones (that’s fancy for “stress on the last syllable”); in the other two cases, both languages have barytones (stress on the first syllable). In Balto-Slavic words where the first syllable has “acute accent” (next table), when the first syllable is dominant, the first syllable gets a “dominant acute”; when the first is recessive and the second dominant (so that the first becomes “secondarily dominant), the first syllable gets a “circumflex”; when both are recessive, the first gets a “recessive acute”. Between the two tables, it is suggested that “possibly” the dominant morphemes had high tone and the recessive ones low tone, “as V. A. Dybo suggests for typological reasons”. Unfortunately, this claim does not come with a reference, but V. A. Dybo is the one of Dybo’s law.
    So, that gets us an extremely complex pitch accent system in Balto-Slavic from an extremely simple tone system + length + laryngeals in PIE.
    (The “secondarily dominant” morphemes are apparently a special development of Balto-Slavic, appearing when a recessive morpheme is followed by a dominant one. The morphemes are usually syllables, but the nominative ending -s is dominant.)
    The first, in English translation but with all bibliographic information and all references cut off, finds very similar cases all over the Old World, in particular West Caucasian and the Avar-Andi languages in Daghestan (of which Tindi retains tones). As the original is from 1978, this paper may be what Nikolayev implicitly cited.
    What would certainly be good to know here is Kapović’s 2008 book “Introduction into Indo-European linguistics”; though with such a modest title, it’s probably brick… and it’s in Croatian. But it probably doesn’t belong to the Moscow School.

    中国很大 zhōngguó hěn dà; very cool!
    There’s a Middle Chinese version too: the Emperor asks “What are these four tones of which you scholars speak?”, and the scholar Shen Yue replies “Tien1 tsi2 shiang3 jieat”, meaning “The Son of Heaven is holy and wise”, or freely rendered “Whatever Your Majesty chooses to make them.” “But,” we are told, “the Emperor never would follow them”, presumably because he didn’t get the joke. (Note that MC tones 2 and 3 are Mandarin tones 3 and 4 respectively.)

    Awesome. Too bad for the emperor, though.

    Indeed, I wonder: could the popularity of substrate theories outside of Romance be a form of “survival in the periphery” of an idea once dominant in the center? In the late nineteenth/early twentieth century substrate theories were very popular among Romance scholars, and Romance linguistics was an influential field within historical linguistics more generally. By the second half of the twentieth century, when substrate theories were systematically disproved within Romance, the field had lost its former prestige and as a result the credibility of substrate hypotheses in other subfields remained unaffected by their being discredited among Romance scholars.

    Wouldn’t surprise me.

    Do Nostraticists entertain the possibility of PIE tones?

    It may be interesting in this context that they reconstruct Proto-Altaic with pitch accent (high or low tone on the stressed syllable of a word) from the pitch accents of (Middle) Korean and Japanese and from the long vowels of Turkic and Tungusic.

    2-In Lhasa Tibetan a tonal distinction arose (inter alia!) through the loss of voicing as a phonological feature in stops: so that a phonemic distinction between (say) /ta/ and /da/ in Classical Tibetan (a non-tonal language) is now, in Lhasa Tibetan, a phonemic tonal distinction made on the /a/, with both syllables now both beginning with /t/.

    This also happened in Chinese except Wú (incl. Shanghainese), where this loss split the 4 Middle Chinese tones into 8. Various mergers have happened since then, but the 1st and 2nd tones of Mandarin result from this split.
    (It’s still allophonic in Shanghainese, where the voiced stops are still voiced but cause low tone.)

    in some recorded accents of French a tonal distinction was found to exist between LAC and LAQUE

    I sit in awe.

    Kam has fifteen tones by usual reckonings, nine in syllables that end in vowels or nasals and six in syllables that end in stops.

    The table in the Wikipedia article shows nine tones in total, of which only six can occur in syllables that end in stops.

    In the emeticity thread I got this idea that the process of phonological reduction that led to the tonogenesis in Chinese was similar to what we now see in Danish, and that Danish may well develop a tonal system. But the thread got closed before I came around to adding it.

    A common mechanism for tonogenesis is the loss of a glottal stop that caused allophonic tone distinctions. Danish seems to have done the reverse, sort of: most dialects have turned the usual Scandinavian pitch accent into the presence vs. absence of some kind of glottalization in a syllable, the infamous stød.
    I’ve wondered if Verner’s Law was reverse tonogenesis, too, with the low pitch of unstressed syllables causing voicing…

  19. David Marjanović says

    Forgot to mention that the WP article on Proto-Indo-European accent uses the term “pitch accent” the way Indo-Europeanists use it, “at most one syllable in a word was distinguished by height (rather than prominence)”, rather than the way phoneticians use it, which would be that the stressed syllable of each word has one of at least two phonemic tones while the pitches/contours of the unstressed syllables are predictable.

  20. marie-lucie says

    JRoth: I have not seen Pullum’s monumental grammar, but he is probably right in so far as some prepositions (such as after and before) are also used as subordinators. However, introducing subordinate clauses is not the typical function of a preposition.

  21. David Marjanović says

    …Uh. It’s also not written by a Russian, but by Ivan Štambuk, an obvious Croatian.

  22. “I’ve eaten many’s the steak in that restaurant.”
    This sounds like what should be a cleft sentence – “Many’s the steak” > “It’s many the strak….” that is a half-assed loan translation from a sentence like this one:
    “whereas many of the said jurisdictions have long since fallen into disuse or been superseded” = de bhrí go bhfuil a lán de na dlínsí sin imithe as úsáid nó curtha as ionad le fada…
    …where “many of the said jurisdictions” translates “a lán de na dlínsí”. So you have “a lán” = “it’s fullness”, a singular noun, translated with “many” an adjective. “many” always has a plural reference, obviously, so it’s pretty easy to see how this can get tangled up.
    BTW this is the kind of problem that almost never occurs in Chinese because doing plurals takes some extra effort and the issue just wouldn’t come up. When will these people ever learn…..?

  23. Wait, wait, I found some more:
    “there is many a … = is iomaí …(number)
    there is many a … = is minic ….(frequency)
    there is many a … = is iomaí nair …
    “Minic” is an adverb, but “iomaí” looks like a singular abstract noun to me.

  24. mollymooly says

    Obviously my
    – Many’s the man would who(/that) disagree with you
    should have been
    – Many’s the man who(/that) would disagree with you

  25. “I’ve eaten many’s the steak in that restaurant.”
    Shurely shome mishtake.
    “I’ve eaten many a steak in that restaurant.” Or “Many’s the steak I’ve eaten in that restaurant.”

  26. Off topic, I just want to mention a phrase used in court today by a barrister, John Kelsey-Fry QC, defending the former cabinet minister Chris Huhne in a big criminal case in Britain. He said Huhne “refutes in the most generous terms” (the idea that he pressured his wife into having an abortion). Refutes in the most generous terms means, then: “strongly denies”.

  27. J.W. Brewer says

    “Refutes” is used differently, I have come to learn, by US lawyers than it is by English barristers. US usage assumes that “refuting” the other side’s factual assertion is actually done via contrary evidence or something like that that tends to establish that the assertion is not, in fact, the the most likely account of what actually transpired historically, whereas over there “refuting” can apparently be no more than a conclusory and self-serving denial. Such denials might well happen to be true, but the mere fact that such a denial has been made doesn’t get you very far in determining that.

  28. For completeness, there is also “many are the NP.pl”, as in the American Civil War song Tenting on the Old Camp Ground:
    Many are the hearts that are weary tonight,
    Wishing for the war to cease;
    Many are the hearts looking for the right
    To see the dawn of peace.
    David: I said that Kam has 15 tones “by usual reckonings”, which count tones in stop-final syllables as tonemically different from tones in other syllables, even if tonetically the same. Otherwise we would have to say that Middle Chinese and Proto-Tai had three tones, only one of which (but which? there seems to be no non-arbitrary choice) appears in stop-final syllables. By the way, barytone strictly means ‘with an acute on the antepenult, or the first syllable if there is no antepenult’. I asked Nick Nicholas once how he would reform written Modern Greek accentuation, and he replied that he would leave off the written accent mark on barytones.
    Marie-Lucie: Puddleston’s point is that there is only one syntactic class of these words, which can govern a NP (“two years before the mast”), a complement clause (“he crawled before he walked”), or nothing (“I have seen it all before”). It is highly lexically specific which words can take which, and nothing is gained by calling the same words prepositions, subordinating conjunctions, or adverbs depending on their use in specific sentences. In defiance of etymology, Puddleston uses preposition for all, whether they are preposed or not (ago and galore are not) and whether they accept NPs or not. It is the same kind of simplification that is achieved by positing noun-noun compounds in English rather than saying, as pre-structuralist grammarians did, that every noun but the last is really a denominal adjective.

  29. J.W., I thought the same thing. He’s not refuting it at all. No matter how generous he is, he’s merely denying it.

  30. “US usage assumes that “refuting” the other side’s factual assertion is actually done via contrary evidence or something like that that tends to establish that the assertion is not, in fact, the the most likely account of what actually transpired historically, …”
    That’s the standard British use too.
    ” … whereas over there “refuting” can apparently be no more than a conclusory and self-serving denial”: that’s the new, dishonest use that has come sneaking in, usually deployed by politicians who don’t like to say “deny”. Why a QC would say it, though, is a conundrum. Sheer ignorance, perhaps.

  31. I’ve checked: J K-F was made a QC during the Blair years; no further explanation may be necessary.

  32. David Marjanović says

    Otherwise we would have to say that Middle Chinese and Proto-Tai had three tones, only one of which (but which? there seems to be no non-arbitrary choice) appears in stop-final syllables.

    Or, as in your transcription, we could consider them toneless.

    By the way, barytone strictly means ‘with an acute on the antepenult, or the first syllable if there is no antepenult’.

    Oops. My source (the same Wikipedia article?) was wrong, then; perhaps it only talked about words with two syllables?

    I asked Nick Nicholas once how he would reform written Modern Greek accentuation, and he replied that he would leave off the written accent mark on barytones.

    That’s exactly what the Soviet spelling reform of Greek did, right?

    It is the same kind of simplification that is achieved by positing noun-noun compounds in English rather than saying, as pre-structuralist grammarians did, that every noun but the last is really a denominal adjective.

    Incidentally, that’s the only way to account for yeast artificial chromosome. Throws me every time, because in German we’d compound “yeast” and “chromosome” and leave the adjective “artificial” in front of the compound: künstliches Hefechromosom.

  33. I’m sorry, are you trying to explain why yeast artificial chromosome is possible in English, or why it is used in preference to artificial yeast chromosome? Or both?
    By the way, to me (a native US speaker) the latter has a slight problem in that it wrongly suggests artificial yeast, whereas the former just sounds awkward, in the same way that Defense Advanced Research Projects Agency sounds awkward.

  34. marie-lucie says

    preposition
    Terms which were first created and defined in Latin in order to deal with Latin grammar, and later adapted to other languages structurally quite different from Latin, rarely fit all cases in the other languages. So it is with “preposition”. Without looking into “Puddleston”‘s mighty tome (ha, I see the pun now), I bet that the prepositions are (or at least could be) listed in three columns according what elements or structures they “govern”, and most of them only govern one or two (and depending on one’s definition, some of them also occur after a verb, “governing” nothing).
    This reminds me of a discussion some time ago where David M argued that there were no adjectives in German, only adverbs. He had a point, although the traditional interpretation was also defensible. Same thing with classifying the English “prepositions” separately or together depending on the degree of importance a grammar writer places on their functions.
    refutes in the most generous terms
    Obviously he did not just say NO! but added a “generous” amount of circumlocutions.
    refute ~ refuse ?
    I think that some people (not scientifically inclined) may be interpreting refute as a fancy form of refuse. In technical terms, refute rather than deny is the correct term used in the title of the book Conjectures and refutations which is about proposing and evaluating scientific hypotheses. For instance, creationists may think they are refuting the theory of evolution but they simply deny or refuse it, since they have no debatable arguments to support their convictions.

  35. J.W. Brewer says

    I don’t know about the specific QC here (although I’m vaguely acquainted with the political scandal this court case seems to relate to), but I became acquainted with what I’d called the British usage by hearing it from multiple barristers/solicitors I had occasion to deal with professionally circa the late 1990’s who generally seemed . . . perfectly respectable/honest by the standards of the profession. No doubt they were happy to seize on a euphemism if they thought they could get away with it, but my inference was and is that the new use dearieme deprecates had become widespread enough at least in legal circles in London that they reasonably could get away with it, whereas it would be a jarring enough usage in New York legal circles that a euphemism-seeking lawyer here would probably not use it because it would blow up in his face, rhetorically speaking.

  36. I wondered if this might be a case of ancient legal jargon rather than weasel talk, but apparently not: The OED (anyway, my tattered paper copy) doesn’t mention the “denial” sense of “refute”. Merriam-Webster (mw.com) does, however. Is this sense making inroads in the US now, too?

  37. It’s an odd one; we’re used to lawyers using unnecessary words (they charge by the word, don’t they?) but not to their using the wrong ones. But JWB has observed it so there we are. Do judges misuse it too?
    Afterthought: was it an age-related error, JWB?

  38. David Marjanović says

    I’m sorry, are you trying to explain why yeast artificial chromosome is possible in English, or why it is used in preference to artificial yeast chromosome? Or both?

    Probably both.

    the latter has a slight problem in that it wrongly suggests artificial yeast, whereas the former just sounds awkward

    Hmmm. You know, yeast artificial chromosome may not have been coined by native speakers of English at all.
    On the other hand, I should have mentioned that it’s based on the template provided by the bacterial artificial chromosome; there is no adjective to yeast.

    David M argued that there were no adjectives in German, only adverbs

    I didn’t; I argued that the dictionary form, which lacks a gender/number/case ending (so it looks like an English adjective), is the adverb, and that “be” goes with the adverb like all other verbs do, not with the adjective as everywhere else in IE (and indeed in the Highest Alemannic dialects). I was quickly convinced that the traditional interpretation accounts for a few cases that I couldn’t explain – though I forgot which ones those are. The traditional interpretation is a three-way partition between normal adjectives (forgot what they’re called), “predicative adjectives” and adverbs, where the latter two are identical in German while the former two are identical elsewhere.
    And yes, I had quite some trouble learning the adjective/adverb distinction of English and French.

  39. “That’s not a refutation. You’re just contradicting.”
    “No, I’m not.”

  40. J.W. Brewer says

    Someone can do some corpus investigation in appropriate BrEng sources over time if they want a better story or timeline than my half-remembered anecdotes – I can just say that it happened more than once, and I seem to recall other American colleagues being baffled by how the Brits were using the word, and it seemed and seems more likely to me that there was a Transatlantic difference in usage than that these London-based lawyers were just inexplicably misunderstanding the word.

  41. Yes, “decide” (for instance) allows a following “on” with a complement sentence, and “on” is lost unless the complement sentence is given nominal form. “He decided to leave/that he would leave/on leaving/on it.” That’s not the same as introducing a sentence with a preposition. It’s just a special kind of prepositional phrase — not a sentence.

  42. I’ve been wondering over the past couple of decades why I’m seeing what I think of as intrusive definite articles where I haven’t seen them during the previous thirty, forty years. They occur in geographical names, which was a specialty of mine. ‘The Hecate Strait’ instead of ‘Hecate Strait’; ‘the Saanich Inlet’ instead of ‘Saanich Inlet’. I’ve wondered if speakers of Slavic languages were miss-using them because they were unaccustomed, but mollymooly’s reference to the paper on Hibernian English grammar might be showing another influence.

  43. marie-lucie says

    iakon, I think that everyone says “the Atlantic Ocean” or “the Baltic Sea”, while maps will just show “Atlantic Ocean” or “Baltic Sea”, without the article. I agree that for “Hecate Strait”, etc, normal speech (at least in the region) would not use the article, but perhaps these names are being treated like “Atlantic Ocean”, etc, thus regularizing the naming pattern for bodies of water.

  44. Point, marie-lucie. My brain can be in non-function mode as well as slow.

  45. Yes, ‘bodies of water’ includes rivers, where the article has long been used.

  46. The OED3 gives sense 5 of refute as ‘reject as without foundation, repudiate’, with quotations going back to 1895.
    Arnold Zwicky has a set of links to articles on English names that are arthrous (with the) and anarthrous (without the). For geographical terms, if they have indefinite boundaries or are moving water, they are normally arthrous; if not, anarthrous.

  47. marie-lucie says

    I only encountered the word arthrous and its opposite anarthrous when I started reading Language Log (from which I was led here) a few years ago. I am still not comfortable with these medical-sounding terms (I see the root arthr and think of arthritis and arthritic). Can someone explain what the linguistic terms have in common with the medical ones?

  48. @marie-lucie: ἄρθρον árthron is Ancient Greek for both “joint; limb” and “connecting word; definite article”. (Latin articulus also has both senses; hence English “articulated”, for example, meaning roughly “jointed, having joints”.)

  49. David Marjanović says

    I’ve wondered if speakers of Slavic languages were miss-using them because they were unaccustomed, but mollymooly’s reference to the paper on Hibernian English grammar might be showing another influence.

    Another influence may be the many US universities that officially include “The” in their names, for example The University of Iowa. I still don’t understand why they do that.
    There’s also The Natural History Museum now. As you can guess, it’s the one on Cromwell Road in London, formerly the British Museum (Natural History).

    Arnold Zwicky has a set of links

    Your link leads straight back to this page.

  50. mollymooly says

    I guess John Cowan meant this Arthrousness page.

  51. “The University of X” is much more common than “University of X” in the US, I believe. In many cases it refers to the public university of the state X.
    On the other hand, “X University” is much more common than “The X University”. This is true regardless of whether X is a state name. The exceptions that come to mind are “The Ohio State University” and “The Johns Hopkins University”.

  52. J.W. Brewer says

    You can see the “X NOUN” v. “The NOUN OF X” pattern elsewhere, including in bodies of water. E.g., Hudson Bay v. the Bay of Fundy. (I think when Canadians say *the* Hudson Bay they typically mean the department store not the body of water, but I am open to correction on that.) On the other hand, it’s both the Gulf of Mexico and the Persian Gulf, so maybe gulfs are somehow different from bays, as (typically arthrous) rivers are systematically different from (typically anarthrous) lakes. In Scotland, it’s definitely the Firth of Forth/Tay etc., but quick googling seems to be getting inconsistent results on whether it’s the Solway Firth or just plain Solway Firth.

  53. I have the impression that adding a “The” in front of a secondary school name is a snob thing in the US and also a little bit in Britain. I’ve noticed lots of private schools doing it: The Dalton School, The Madeira School, as opposed to public schools like, say Stuyvesant High School.
    I never know whether to capitalize “The” in publication names: it’s the Guardian, but The Independent and The NY Times, if you go by their own conventions.

  54. It’s always possible that the most generous terms he was refuting in were boxes of files, but it doesn’t seem likely in the circs.

  55. I think that in logic, “refutability” refers to that logical principle which licenses one to assert the negation of a proposition which has been shown to allow the deduction of a falsehood. (I just looked on line, though, and failed to confirm this.)

  56. It’s “the Solway Firth” or, locally, “the Solway”.

  57. marie-lucie says

    Thanks for the explanations on the *arthric words! I had been thinking about bones more than joints, so the metaphor did not make sense.

  58. There are also the Gulf of Maine and the Gulf of St. Lawrence. Also the Gulf of California (a.k.a. the Sea of Cortez). And the Bay of Pigs, and the Bay of Biscay. That “of” really does seem to want a “the”, for the most part.
    San Francisco Bay, but on the other hand (and maybe this can be chalked up to poetic license) “I left my home in Georgia headed for the ‘Frisco Bay” (in the song “Sittin’ on the Dock of the Bay”).

  59. Empty, I think that would be “the Frisco bay”; i.e. “bay” is not part of the name here. But the other explanations may be sound; I am always prejudiced in favor of the speaker/writer being correct and it being the rest of us who fail to understand correctly.

  60. John, I’m sure you’re right about that example–also right to remove my apostrophe (maybe I should have written Fr’isco).
    the other explanations may be sound
    That reminds me: Puget Sound, Long Island Sound, Nantucket Sound, … but apparently the Mississippi Sound.

  61. marie-lucie says

    Ø: Puget Sound, Long Island Sound, Nantucket Sound, … but apparently the Mississippi Sound.
    Perhaps this is because Puget, Long Island, Nantucket are names in their own right, never needing the article, but references to the river rather than the Sound always use the article, as in Life on the Mississippi.

  62. J.W. Brewer says

    The arthrous Delaware River runs into anarthrous Delaware Bay, so there are presumably a lot of different sometimes competing factors at play here. (The state of Delaware is anarthrous, but it took its name from the river rather than vice versa.)

  63. Is the Mississippi Sound named after the Mississipi River or after the state of Mississippi?
    I thought the state of Delaware was named after a person.

  64. J.W. Brewer says

    The state was eventually (once it had fully separated from Pennsylvania) named after the bay/river, which in turn had been named (substantially earlier) for Lord De La Warr, one of the early governors of the Virginia colony.

  65. In one of Patrick O’Brian’s Aubrey/Maturin novels (in which he seems to aim for a high degree of historical accuracy) a character refers to the Connecticut.

  66. Surely in reference to the Connecticut River? That’s how we still refer to it today.

  67. My impression was that the character was referring to the place Connecticut as the Connecticut. Yes, here it is: The character is a Frenchman.
    http://books.google.com/books?id=0r6H-XT3cBgC&pg=PT116&lpg=PT116&ots=XjNNFm0kFl&dq=Patrick+O%27Brian+Fortune+of+War+%22the+Connecticut%22

  68. A Frenchman, moreover, whose grasp on the articles, it is none of the best:
    “You would not object on the national grounds?”
    “[…] if such is to be found under the Heaven” — and here he claims to be quoting an anglophone!

  69. I think we can rule that out as evidence for English usage.

  70. J.W. Brewer says

    I can match the “the Frisco Bay” with a somewhat earlier Chuck Berry lyric depicting a cross-country itinerary in the other direction (“from the coast of California to the shores of the Delaware Bay”). Chuck Berry is no Frenchman, but I suspect that definite article may have been poetic license, driven by rhythmic/prosodic concerns. In any event, the google n gram viewer lets one compare the rate of “of Delaware Bay” to “of the Delaware Bay” and the latter exists but is clearly the minority variant.
    I wonder, btw, if the particular itinerary in the Berry lyric is premised on the notion, in the pre-Interstate days, of getting behind the wheel, leaving San Francisco on old U.S. 50, and just taking it all the way to the end; the eastern terminus (in Ocean City, Md.) is about 25 miles too far south to be the “shores of the Delaware Bay”, but that might be within some sort of close-enough-for-rock-and-roll margin of error for geographical precision.

  71. How did it come about that in German there are two nouns See of different gender meaning (more or less) “sea” and “lake”?

  72. I googled “from the coast of California to the shores of the Delaware Bay” and the first 3 references were missing that “the” before “Delaware”. I’ll also note that it is not clear there really should be a “the” before Empty’s example “Ohio State University”. While it is true that OSU thinks it’s there, you only hear it there in official contexts. (I went to grad school at OSU.)

  73. Then there are the cases where the public university of state X has the word “State” in its name. Two examples:
    Official name: The Pennsylvania State University
    Everyday name: Penn State
    [There is also a “University of Pennsylvania”, but it is a private institution, nicknamed “Penn”]
    Official name: Michigan State University
    Everyday name: Michigan State
    [There is also a “University of Michigan, and it is a public institution!]

  74. marie-lucie says

    in some recorded accents of French a tonal distinction was found to exist between LAC and LAQUE
    I have been experimenting with pronouncing these two words, which I would normally consider homophones. Assuming that the speakers spoke a variety of Northern French (and therefore did not pronounce the final e in LAQUE), I wonder about the conditions of recording: were the words recorded within natural sentences, or by having people pronounce both words one after the other, in isolation? in the second case, I too would have a tonal difference, although it would not matter in what order they were said. The tonal difference would result from the intonation of the word sequence, not from inherent tones in each word. When recording dialect, persons unfamiliar with the dialect can misunderstand the conditions for some occurrences.

  75. David Marjanović says

    How did it come about that in German there are two nouns See of different gender meaning (more or less) “sea” and “lake”?

    No idea about the genders (sea female, lake male). But I suspect that one is Low and the other High German. First, there is no sea where High German is native; second, the Low German word for “lake” is… wait for it… Meer, just like in Dutch.
    Complicating factor: Meer has cognates meaning “sea” all over Indo-European; See is apparently limited to Germanic. But the cognates of Meer don’t specifically mean “sea” everywhere either.

  76. David Marjanović says

    cognates meaning “sea” all over Indo-European

    They even have the same neuter gender, at least in Slavic and Latin.
    One case comes to mind where the two “sea” words are used for disambiguation: Nordsee = North Sea; Nordmeer = a vague term that seems to include Europäisches Nordmeer (between Norway and Greenland, or so) and the Arctic Ocean, but never the North Sea. Which is west of most of England and all of Scotland, incidentally, implying that the name must be really old.

  77. marie-lucie says

    DM: the Low German word for “lake” is… Meer, just like in Dutch.
    This must be the origin of the suffix -mer (pronounced as if “mé”) in a few place names in the province of Lorraine, which is far from the sea but shares borders with Belgium, Luxembourg and Germany (local Germanic dialects are still associated with the area). One example is the name of Gérardmer, a small town on the shore of a lake.

  78. Fantastic, I love these little connections!

  79. Marie-Lucie: IF memory serves me right the LAC/LAQUE tone distinction was detected, sometime in the first half of the twentieth century, in the regional French of Burgundy or a nearby province by a Swedish or Norwegian researcher. And yes, this was a PHONOLOGICAL distinction, so both words were /lak/, segmentally.
    Said researcher, as the native speaker of a tone language, observed that similar such tone distinctions may exist or have existed in other parts of France without ever having been detected.
    David: André Martinet wrote somewhere (I think in his DES STEPPES AUX OCÉANS) that German DIE SEE and DER SEE match, in terms of grammatical gender, French LA MER and LE LAC, making him suspect that Romance-Germanic language contact within Charlemagne’s Empire might be the culprit.

  80. marie-lucie says

    Etienne: LA MER et LE LAC : Germanic influence makes sense, because the words for the sea are neuter in Latin and masculine in Italian, Spanish and Portuguese, and should normally have ended up as masculine in French also.

  81. Etienne- who is the culprit for which? Did the french rub off on the German or vice versa?
    David Maranovic – if I understand you right, today’s standard german took das See from Low German and die See from High German?

  82. Trond Engen says

    David M.: No idea about the genders (sea female, lake male). But I suspect that one is Low and the other High German. First, there is no sea where High German is native; second, the Low German word for “lake” is… wait for it… Meer, just like in Dutch.
    s/o: David Maranovic – if I understand you right, today’s standard german took das See from Low German and die See from High German?
    Neuter in Low German? And the semantic split is not noted in Beate Hennig’s Kleines Mittelhochdeutsches Wörterbuch:

    stMFswM [flekt. auch sêwe-], See; Meer.

    (Masculine or feminine with strong inflection, masculine only with weak, whatever that might mean.)
    In modern Mainland Scandinavian it’s s(j)ø m. after regularization of ON sjór/sær/sjár etc., all masculine. The multiple forms in the old languages are believed to reflect analogical substitutions in a messy paradigm with umlauts and shortenings, all stemming from a root *saiwi-. According to Bjorvand and Lindeman, this in turn could be from *saigWi-, related to ‘sieve’, but they’re not convinced by the semantics “what has sieved out (from swampy soil)”. I like it better. I think the semantics could be something like “sieving” -&gt “stream” (maybe -&gt “wave”) -&gt “sea”, essentially a poetic understatement or metonomy, similar to bøljan blå “the blue wave” = “the ocean”.

  83. marie-lucie says

    TE: sea/sieve
    When a wave is receding from the beach, the water at the front edge does not seem to flow back but instead sinks into the sand as if into a sieve. Could that be the idea that links the two words? So perhaps in regions familiar with both sea and lakes, the “sea” would be the water that ebbs back and partially sinks into the sand, while “meer” does not do so. But in inland regions, the difference was not understood and a local preference arose for one word or the other. Would that make sense?

  84. Well, it depends on the size of the lake. If it’s big enough to have wind-driven waves, the sieving effect will happen regardless.

  85. Trond Engen says

    Another Germanic word for ‘sea’ is hav etc. B&L see this as identical with the homonymous verbal noun from *habján-/*háfjan- “lift, raise”, i.e. something like “high sea”. It’s a little disturbing that the meaning in Dutch and LG is “shallow, protected sea”, though. It’s esoteric, but now I wonder if the words meaning “raise” and “sieve” could have been used to describe the movement of the vawes or the tide along the shore.

  86. marie-lucie says

    TE, is hav related to haven?
    Here too, when two or more words have very similar meanings, several things can happen to their evolution: one takes over, or becomes the default while the other is more specialized (or is forgotten), or they both develop more specialized meanings, especially if they are used by different human groups, especially if they live in different places or under different conditions. So if the same root ends up as “high sea” in one place or language and “shallow, protected sea” in another, it is not necessarily an argument for or against a common origin for the two. (This is a general comment, since I don’t know too many of the details of Germanic historical linguistics).

  87. Trond Engen says

    I meant to write “… the homonymous verbal noun ON hav n. “lifting, raising” from *habján-/háfjan- “lift, raise”, …”
    marie-lucie: The common origin of Scand. hav and D./L.G. haff is not in doubt. It’s the presumed original sense as an image parallel to “high sea” I found less convincing. Or at least weak enough to have me intrigued by the semantic pair “lift”/”sieve”.

  88. Trond Engen says

    ON haf n.

  89. Trond Engen says

    haven
    Oh, I forgot about that. Haven is thought to be loan from Scandinavian. ON höfn f. looks to me, superficial though I am, to be an inchoative to the verb hafa “hold, keep, etc.” &lt *kap-. But it isn’t that simple, since B&L says that hafa &lt *haBé:n- “hold” (Lat. capere) is a weak and durative formation from the strong verb hefja &lt *háfjan-/*haBján- “lift, raise” but also “cease” (Lat. capio). Moreover, höfn has lots of meanings, some apparently derived from one verb, some from the other, and some that can’t be safely attached to either of them. Of course, that may ultimately be irrelevant, since höfn was derived from the common root. B&L also says it has to be a very old derivation, since the ending -nó- was no longer attached directly to the stem of strong verbs in Germanic. OTOH, hafa was weak.
    Anyway, höfn used for seaports would apparently have come from a meaning “holding”, while haf would suggest “raising” or (older) “capturing”.

  90. Trond: In general the Germanic verbs with -j- in them are derived causatives which trigger umlaut, hence hefja < *hafja < hafa, not the other way about. In any case, capio and capere are just different citation forms (1sg present indicative active and present active infinitive respectively) of the same Latin verb meaning ‘hold, seize, take’ (English capture, captive), so something’s wrong with that story. It is one of a small group of irregular third-conjugation verbs that have some fourth-conjugation endings, so that the -i- comes and goes within the conjugation, appearing in some forms only. (Other verbs of this type include cupio, facio, fugio, iacio, specio). The usual Latin word for ‘cease’, though, is desino, desinere (English desist).

  91. marie-lucie says

    The large French seaport Le Havre on the Normandy coast, in the estuary of the Seine, is probably also from Scandinavian, one of the Viking remnants in French. Its old name was Le Havre de Grâce ‘the harbour of mercy’, with the extra words added to differentiate it from a number of other ports in the region, which would also have been referred to locally as le havre, while the word was unknown in the rest of the country except as the name of the well-known port city. This word is also used in Canadian French to mean ‘harbour’, which in “metropolitan” French is simply le port.

  92. Trond Engen says

    John: I know about the causatives, that’s one reason I found this hard to grasp. There’s something really strange going on in this paradigm. I wonder if part of it might be a very ancient case of deponency – maybe the old have/give duality itself.
    I lifted the Latin forms from B&L, assuming they were meant to represent different verbs, but I should have noticed that they don’t.

  93. marie-lucie says

    I forgot to say that un havre is also found as an archaizing word to mean ‘refuge, shelter’, which would agree with the Scandinavian words emphasizing ‘keep, hold, protect’. But according to the TLFI, the French word (first attested as havene) is a borrowing from Middle Dutch (perhaps through English, since the first French attestation is later than the Norman conquest). In any case, Middle Dutch is also a Germanic language, so the Proto-Germanic etymology would be the same as for the Scandinavian words.

  94. Trond Engen says

    Is that -n- &gt -r- regular in any way?

  95. marie-lucie says

    Is that -n- > -r- regular in any way?
    This change is quite common in the history of Romance languages, like the interchange between l and r discussed a while ago, so there is nothing particularly remarkable about it. The sequence -vn- in the adopted word would have been impossible in Old French, while there were (and still are) quite a number of instances of -vr- (eg livre ‘book’, chèvre ‘goat’, lèvre ‘lip’, oeuvre ‘(piece of) work’, poivre ‘pepper’, cuivre ‘copper’, suivre ‘to follow’, etc). Of course at the time the /r/ was the apical trill, not the uvular as nowadays, so its articulation was closer to that of /n/.

  96. David Marjanović says

    Etienne: LA MER et LE LAC : Germanic influence makes sense, because the words for the sea are neuter in Latin and masculine in Italian, Spanish and Portuguese, and should normally have ended up as masculine in French also.

    Oh, that finally explains la mer!
    Le lac doesn’t need an explanation, it’s straight from Latin lacus.

    if I understand you right, today’s standard german took das See from Low German and die See from High German?

    No. It took die See “sea” from Low German and der See “lake” from High German. In addition, there’s das Meer, meaning “sea” in the standard and “lake” in the north. Sorry for the confusion.

    When a wave is receding from the beach, the water at the front edge does not seem to flow back but instead sinks into the sand as if into a sieve. Could that be the idea that links the two words?

    Makes sense. The sieve itself, though, is neuter in German (das Sieb).

    Haven is thought to be loan from Scandinavian.

    It does show up in German, though: Low German Haven “port/harbor” (as in the place names Bremerhaven, Wilhelmshaven, Cuxhaven) was borrowed into Middle High German with /f/, modern Hafen, when the native form Haben had just become too confusing.

    It is one of a small group of irregular third-conjugation verbs that have some fourth-conjugation endings

    I was taught that as “mixed conjugation” (marked “M” in the dictionary) as opposed to “consonantic conjugation” (“3”) and “i conjugation” (“4”).

  97. Marie-Lucie, David: it’s a little more complicated.
    1-French “lac” is actually a learned borrowing from Latin (intervocalic /k/ followed by a back vowel (LACUS) should have been lost).
    2-LA MER: it is indeed feminine in French, unlike MAR in Spanish or MARE in Italian, but this needn’t imply that this was due to Germanic influence: former neuters often fluctuated between masculine and feminine gender for a long time, and indeed in Spanish this fluctuation lasted long enough that MAR is still feminine in some fixed expressions (“Hacerse a la mar”, “To set sail”, for example).
    A comparative diachronic history of SEE and MEER in Continental West Germanic, and of MARE and (popular reflexes of!) LACUS in neighboring Romance varieties, complete with the interaction of semantics and grammatical gender and language/dialect contact, would make a very nice little monograph or dissertation.

  98. David Marjanović says

    intervocalic /k/ followed by a back vowel (LACUS) should have been lost

    …oh. Yeah.

  99. Nice example I saw in an email signature today:

    La sal en el mar es más que en la sangre.
    Le sel dans la mer est plus que dans le sang.

    Note the preposition alternation as well as the gender alternations.

  100. Very nice indeed!

  101. North Sea. Which is west of most of England and all of Scotland, incidentally, implying that the name must be really old.

    Well, I don’t know: the Hudson River was formerly known in New York City as the North River, despite being west of the city. In this context North River surely means ‘river leading to the north’. (The overall direction of flow is south, but in NYC and its immediate environs it is strongly tidal, which is why the Mahican name literally means ‘river that flows both ways’.

  102. Turns out I’m wrong: the Dutch named it the North River in contradistinction to the South River, the Delaware. The English calqued the name when they arrived.

Speak Your Mind

*