Quid dicent?

A Guardian story by Alison Flood reports on some interesting material:

Ever been unsure about how to deal with a drunken family member returning from an orgy? A collection of newly translated textbooks aimed at Greek speakers learning Latin in the ancient world might hold the solution.

Professor Eleanor Dickey travelled around Europe to view the scraps of material that remain from ancient Latin school textbooks, or colloquia, which would have been used by young Greek speakers in the Roman empire learning Latin between the second and sixth centuries AD. The manuscripts, which Dickey has brought together and translated into English for the first time in her forthcoming book Learning Latin the Ancient Way: Latin Textbooks in the Ancient World, lay out everyday scenarios to help their readers get to grips with life in Latin. Subjects range from visiting the public baths to arriving at school late – and dealing with a sozzled close relative.

“Quis sic facit, domine, quomodo tu, ut tantum bibis? Quid dicent, qui te viderunt talem?” runs the scene from the latter, which Dickey translates as: “Who acts like this, sir, as you do, that you drink so much? What would they say, the people who saw you in such a condition? […]

The Latin learners are provided with examples of how to deal with visits to sick friends and preparations for dinner parties. They are also briefed on trips to the market to wrangle over prices (“How much is the cape?” “Two hundred denarii.” “You’re asking a lot; accept a hundred denarii”) and an excursion to the bank.

“We don’t know if they would have roleplayed the scenes with other students,” said Dickey, a professor of classics at the University of Reading. “But my hunch is that they did.”

Dickey said the texts were very commonly used. “We know this because they survive in lots of different medieval manuscript versions. At least six different versions were floating around Europe by 600 AD,” she said. “This is actually more common than many better-known ancient texts: there was only one copy of Catullus, and fewer than six of Caesar. Also, we have several papyrus fragments – since only a tiny fraction survive, when you have more than one papyrus fragment, for sure a text was popular in antiquity.” […]

There’s a phrasebook section on excuses (“You did what I told you?” “Not yet “Why?” “I (shall) do it soon, for I’m in a hurry to go out”), and a varied one on insults. “Maledicis me, malum caput? crucifigaris!” or “Do you revile me, villain? May you be crucified!” is one particularly vicious one, along with: “And does he revile (me), that animal-fighter? Let me go, and I shall shake out his teeth.”

“When we think of the Romans, it’s mainly of the rich and famous generals, emperors and statesmen,” Dickey told the Guardian. “But those people are clearly atypical: they’re famous precisely because they were remarkable. Historians try to correct this bias by telling us about the masses of ordinary Romans, but rarely do we have works written by or about these people. These colloquia give us real, contemporary stories about their lives and I hope my work gives a fairer and truer vision of ancient society.”

The book sounds like a lot of fun. Thanks, Bathrobe!

Comments

  1. Nothing about Lenin or celebration of the May Day? Also, why the archaicized translations? I can understand if translation is not idiomatic, but targeted to represent the structure of foreign phrase, but “I shall”?

  2. *mumbles something in Latin about materialismus dialecticus*

  3. SFReader says

    Vladimirus Eliae filius Lenin – revolutionarius Russicus, est conditor et dux Russicae factionis communisticae, dux revolutionis Octobris socialisticae (7 Novembris 1917), primus praefectus (praeses Concilii Comissariorum Popularium) Russicae Sovieticae Foederativae Socialisticae Rei Publicae et Unionis Sovieticae. Multa scripta politica et philosophica creavit.

    Fuit dux Revolutionis Octobris socialisticae, cuius gratia statuta est “dictatura proletariati”, quam Lenin divulgabat, factio Bolsevicorum (ab mense Martio 1918 “Russica Communistica Factio (bolsevicorum)” nominata) facta est regnatrix.

    A Secundo Omnirussico Convento Sovietorum (7–9 Novembris 1917) Vladimirus Lenin electus est praeses Concilii Comissariorum Popularium et Russiam sovieticam administrabat.

    Die 3 Martii 1918, Lenin suique confecerunt pacem cum Germania, sed Bellum Civile Russicum ad annum 1922 processit.

    Lenin mortuus est in Gorki die 21 Ianuarii 1924; cadaver mumificatus in mausoleum Moscuae positus est.

  4. George Gibbard says

    anyone: should it be “dictatura protaletariatūs”?

  5. SFReader says

    the word in nominative is “proletariatus”, I believe.

    It’s 2nd declension (-us/er), so genitive case should be “proletariati”.

  6. George Gibbard says

    I was thinking of senātus and magistrātus which are fourth declension, genitive singular senātūs and magistrātūs. I’m surprised if proletariatus (don’t know the vowel lengths) is second declension.

  7. George Gibbard says

    Apparatus, another fourth-declension noun, prescriptively has the plural apparatus in English.

  8. Athel Cornish-Bowden says

    Also, why the archaicized translations? I can understand if translation is not idiomatic, but targeted to represent the structure of foreign phrase, but “I shall”?

    There is nothing archaic about “I shall” and “we shall”: they are everyday usage in British English, though they may have disappeared from American English. Eleanor Dickey is British (at least, she works in a British university), so it’s perfectly natural she should write in the way she (probably) speaks.

    In her page at http://reading.academia.edu/EleanorDickey, she says: “I specialise [note spelling] in ancient languages and literatures and in scholarship and language teaching in antiquity, was educated at Bryn Mawr College and Oxford University, and now work at the University of Reading in England.” Bryn Mawr suggests an American origin, however.

  9. SFReader says

    “Diccionario auxiliar español-latino para el uso moderno del latín” by José Juan del Col, publisher -Instituto Superior Juan XXIII (2007)

    proletariado: proletariatus, us m. Sin: proletárii, orum mpl; classis proletária; proletária (vel operária, vel aerária) plebs (plebis f) ; páuperes; egentes. Uso: dictadura del -, dictatura proletariatūs.

    OK, the question is settled then.

    dictatura proletariatūs

  10. There is nothing archaic about “I shall” and “we shall”: they are everyday usage in British English, though they may have disappeared from American English. Eleanor Dickey is British (at least, she works in a British university), so it’s perfectly natural she should write in the way she (probably) speaks.

    Sure, sure, but – “Do you revile me, villain”? How about “You insulting me, jackass?” or whatever they say across the pond.

  11. Alon Lischinsky says

    @George Gibbard:

    anyone: should it be “dictatura protaletariatūs”?

    SFReader has come up with the short answer (yes).

    The general rule is that nouns ending in -us but representing an office, social position or collective body are of the fourth declension. Miller (2006. Latin Suffixal Derivatives in English and Their Indo-European Ancestry. Oxford: OUP, p. 51) thinks they were regularly formed as deverbals within Latin.

  12. (“How much is the cape?” “Two hundred denarii.” “You’re asking a lot; accept a hundred denarii”)

    “…accept a hundred demarii” seems strange. “will you take,” or “I’ll give you..” would be more usual.

  13. Why the strangely formal translations? Well, look at when these books were produced: “between the second and sixth centuries AD.” By that time, the Latin in these textbooks could have represented a particularly literary standard. Just like textbooks for foreigners today, these Greeks et al. would have first learned a “proper” way of speaking that had been slightly fossilized, and only later a more realistic register. And by the time these colloquia were used in the medieval era, their Latin had already become something far removed from any popular speech. My guess is that Dickey’s translations mirror that.

  14. I guess I can see that line of argument, but it still seems weird to render “Maledicis me, malum caput? crucifigaris!” as “Do you revile me, villain? May you be crucified!”

  15. Ah, if you look at the comments on the Guardian article, Prof. Dickey herself posted an explanation of the stodgy translation:

    “This is the author writing here. The reason I have translated the texts as I did is that the original translation, into ancient Greek, followed a very close line-for-line equivalence to the Latin. In my version I am trying to re-create the ancient reader’s experience, so I have kept the Latin and replaced the Greek with English that follows the same line-for-line equivalence; you can’t really see this in the Guardian, as they have removed the Latin and altered the layout. You can see how it was supposed to work in the extract posted here: https://reading.academia.edu/EleanorDickey. I am also publishing a more idiomatic English translation without the Latin, but you’ll have to wait until next year for that.”

  16. Ah, thanks very much for that! You’re a braver man than I; I religiously follow the commandment “Never read the comments.”

  17. Except here, which suggests that one should be selective. Comment is Free comments are fair-to-middling; YouTube ones are usually awful.

  18. Well, yes, I meant “on sites frequented by the many-headed hydra,” like newspapers.

  19. Lenin mortuus est in Gorki die 21 Ianuarii 1924; cadaver mumificatus in mausoleum Moscuae positus est.

    Cadaver is neuter, so mumificatum, positum. And in Gorki should really be a locative without in, but I’m not sure how you would do that.

  20. My experience reading CiF is that once a comment is bad, the entire subtree of comments below it is equally bad and can be skipped, so that it is one of the places where threaded posting makes sense. Here, where everyone seems to read everything, threaded posting would just make life more difficult for all of us.

    Even YouTube comments on obscurities can be decent: I’ve left a few myself.

  21. Gorki would pretty much be an indeclinable noun, I think; it just can’t be made to fit any Latin declension (unless you construe it as plurale tantum!) Similarly indeclinable city names are Nazareth, Hebron, Bethle(h)em. So in Gorki it must be.

    Latin has a few indeclinable common nouns too, notably nihil/nīl, which is ne + hilum ‘trifle’, the latter a first-declension neuter of unknown but possibly Semitic etymology.

  22. From a quick search of the Vulgate it looks like Latin speakers weren’t quite sure what to do with such foreign city names: there’s venerunt in Hebron, but also venit Nazareth, as an allatival accusative. But I haven’t found any locatival uses of the bare city name, so you’re probably right about in Gorki. (Part of me does want to make it Gorkis, on the model of Athenis.)

  23. @John Cowan: Eh, I’m not sure that Hebrew names are a good analogy here. With -ий/ая/ое we’re dealing with a highly productive nominal paradigm in another highly inflected IE language; to me it seems remiss not to adapt it to Latin morphology.

  24. Why not go all the way and make the name Amarus?

  25. Hat: Bah.

    Lazar: Wiktionary also lists the indeclinable (but feminine) name Figenoiama < fuji no yama!

  26. Hat: Bah.

    Surely you mean “Eheu.”

  27. In some of the other corners of the blogosphere that I frequent, the admonition not to read to comments at other sites is quoted as, “Don’t get out of the boat.” (No matter how good the mangoes look, it’s not worth it.)

  28. Maybe hīlum < fīlum ‘thread’? From Faliscan, perhaps?

  29. Oh, that’s an interesting idea. De Vaan just says “No etymology” for hīlum.

  30. Why not go all the way and make the name Amarus?

    Won’t do: it’s the diminutive of горы Го́рки Ле́нинские

  31. Oops, you’re right — I was thinking of the renamed Nizhny Novgorod, but of course it wasn’t renamed yet, and Lenin didn’t die there. (Now I’m creating my own Radio Erevan joke.)

  32. Yes, I had the same mental lapse that Hat did.

  33. fīlum wasn’t my original idea. It’s in Bakkum’s The Latin dialect of the Ager Faliscus, p. 83, fn. 43. The Faliscan entry in Wikipedia says some more about f/h alternations (late sound change, followed by h>f hypercorrections, to muddy the waters.)

  34. Trond Engen says

    I just realized that the Bakkum book is the same one that I’ve been reading since yesterday, Bakkum is very much in favour of Proto-italic, but that’s another thread.

  35. Is Latin nihil (and the shortened nil) indeclinable? Only if you do as some dictionaries do, and declare nihilum, nihili, and nihilo forms of a different word nihilum with exactly the same meaning. I would have thought it more plausible to say that nihil is sometimes indeclinable, sometimes declined as a 2nd-declension neuter, just as it is sometimes shortened to nil and nihilum to nilum. I’m no expert on morphology, but, so far as I’ve noticed, the forms with and without endings seem to be used just as interchangeably as the forms with and without H in the stem.

  36. Relatedly, I’ve noticed that guides to ecclesiastical Latin say to pronounce the intervocalic h in nihil and mihi as [k] (reflected in Italian annichilire and Spanish aniquilar). Was there any precedent for this in native Latin, or is it just a medieval invention?

  37. @Michael Hendry: I completely agree; in fact, that’s exactly what I tell students who are confused by textbooks’ and dictionaries’ listing nihil and nihilum as different words.

    They’re not totally interchangeable — nihil can’t, I believe, be used in other than nominative/accusative functions (so calling it “indeclinable” is rather misleading); and nihilum is somewhat preferred to nihil after prepositions that take the accusative (e.g. ad nihilum is more common than ad nihil, though the latter also occurs).

  38. @Lazar: Sounds like a medieval invention due to the fact that the Romance languages had lost [h], so the choice was between pronouncing those h‘s imprecisely or eliding them completely.

  39. Intervocalic /h/ is [k] in the ecclesiastical pronunciation of Latin, but I think it’s entirely artificial. All other /h/ is zero.

  40. I’ve seen nihil, mihi spelt nichil, michi in medieval texts; I’m not sure wether this is the source of the ecclesiastical pronunciation or just its reflection.

  41. My understanding, from having read a bit about Gregorian chant, is that the reason the “h” in “nihil” and “mihi” is pronounced [k] is because it’s easier to sing that way. I don’t know if that’s the original reason or the only reason for it but I’ve come across this explanation more than once.

  42. David Marjanović says

    What’s easier to sing about this?

  43. Kind of awkward to stretch /i/ over two or more syllables, I would think.

  44. David Marjanović says

    Oh, I thought “easier than [h]”.

  45. Essentially the same thing. Have you ever sung in a chorus? I assure you, [h] doesn’t make a good separator.

  46. What’s easier to sing about this?

    Like Languagehat said, it can be awkward to sing. With the “h” sound, it means releasing air from the throat at least twice on a single vowel. Not too awkward when spoken, but when someone’s trying to project with a singing voice in unison with others on a melisma…. With the silent “h”, it means sustaining the breath on the same vowel over at least a couple of syllables or beats while releasing air and, in a large room, trying to project. It ends up sounding like mi or nil instead of mihi or nihil, especially when sung on a single pitch.

    I also wonder if acoustics played a part. As worship moved from smaller spaces (homes and catacombs) to larger spaces (basilicas and cathedrals), reverberation would have started to make intelligibility a challenge. Not only for choirs but for priests and deacons as well. Maybe without the [k], mihi and nihil sounded more like syllables attached to the previous or following words.

    I once went to mass at a parish church built in the Romanesque style. I sat about 2/3-3/4 back in the nave. When it came time for everyone to recite the Nicene Creed together, this particular priest lowered his voice a bit. I was close enough to still hear him but also far enough away that I could hear the people behind me and pretty soon the priest in the front and the people in the back were no longer together. This is the sort of thing medieval speakers and singers of Latin would have encountered.

  47. Moving forward to the 1920’s, I’m reminded of Robert Benchley’s French for Americans: A Handy Compendium for Visitors to Paris, available in full in The Benchley Roundup

    Haven’t you got any griddle-cakes?
    N’avez-vous pas des griddle-cakes?

    What kind of a dump is this, anyhow?
    Quelle espèce de dump is this, anyhow?

    Two hundred francs? In your hat.
    Deux cent francs? Dans votre chapeau.

  48. does he revile (me), that animal-fighter?

    Is this referring to gladiators? Sounds awfully non-PC.

  49. I’ve always been a bit apprehensive about those ’s that pop up in the second declension.

  50. I really wish I’d kept a hold of “Zulu for the Household” (pub. Cape Town, 1909) which included such useful phrases for the Edwardian colonial woman as “Put the left-over rice pudding in the refrigerator” and “These shirts are inadequately starched”.

  51. David Marjanović says

    Stuart-Smith, p. 47–48, first footnote removed, the others included in the text, because I’m fed up with the linguistic practice of hiding both source citations and really important arguments in footnotes, forcing readers to go back and forth all the time, sometimes between several pages:

    The Latin phoneme /h/, represented by <H>, showed a defective distribution from the earliest period: /h/ occurred word-initially before vowels, but only rarely word-internally between vowels (after consonants only in a few compounds, e.g. adhibeō ‘I add to’). [Footnote: “There are a few instances of interjections which show word-final <H>, e.g. āh; the realization of /h/ in this position is not known {source}.”] That the loss of/h/ began before the earliest texts is shown by those words which are never spelt with <H>, but where /h/ is expected (from *gh). [Footnote: “It also took place before rhotacism, as is shown by diribeō ‘separate’ < *dis-hib- < *dis-hab- {source}.”] /h/ continued to be lost during the historical period, but to differing degrees depending on the position of the sound in the word, and on the variety of Latin. It is generally assumed that /h/ was lost earlier in rural dialects of Latin and in colloquial urban Latin, than in the language of the educated classes, though even for these speakers, using /h/ may have been largely an educated convention (Untermann p.c.). [Footnote: “Thus ānser without expected /h/ < *gh, is usually explained as of rural origin {sources} and {source} gives the lack of /h/ in meiō, and liēn ‘spleen’ as the due to the fact that these are ‘more or less vulgar words’. For the loss of /h/ in dialectal Latin, rural and urban, see e.g. {source}; see also {sources}, who argue that the non-standard urban dialects of Latin are probably based to a large extent on the regional dialects of immigrants from the surrounding countryside.”]

    In word-initial position, /h/ was lost relatively early in rural Latin (cf. e.g. Varro’s comments on rustic ēdus for haedus ‘goat’, LL 5.97), and probably also in colloquial urban Latin to judge from inscriptional evidence (e.g. ostiam for hostiam, GIL I² 193), and reports of the tendency to use /h/ where it was not expected (e.g. Arrius in Cat. 84) [that’s “in hinsidiis”, right?] and similarly hypercorrect variants of forms with /h/ where it was not etymologically justified (e.g. humerus beside umerus ‘shoulder’, cf. Skt aṃsa-). [Footnote: “For the implication that hypercorrect forms reflect socially-determined variation, in this case and in general in Latin, {source}. This variation led to some confusion about which words should be spelt with <H> and which without (e.g. (h)arena ‘sand’, (h)arundō ‘reed’ {source}.”] Educated speakers of Latin may have pronounced /h/ word-initially (hence the possibility of hypercorrection), but even in the literary standard, metrical rules indicate that /h/ was very weak in this position. [Footnote: “/h/ does not ‘make position’, e.g. adhibeō is scanned with the first vowel short, and words beginning with /hV-/ are elided as if they were /V-/ alone {source}.”]

    In word-internal position /h/ was lost in many words in most varieties of Latin, including the literary standard, at an early period (e.g. nēmō ‘noone’ < *ne-hemō). It is retained in the spelling of some words (e.g. uehō ‘I carry’, trahō ‘I drag’), but by the end of the Republic it seems that even literary Latin had lost /h/ in this position (cf. for mihi ‘to/for me’; nīl beside nihil ‘nothing’ in the same line, Cat. 17.21). Orthographic conservatism lead [sic] to the use of <H> in the spelling of forms where the sound was now lost, and from this <H> became common as an orthographic marker, to mark either vowel hiatus (e.g. ahēnus ‘bronze’ < *ayesnos, cf. Skt ayas) or vowel length (e.g. mehe for ‘by/from me’ Quint. 1.5.21; Leumann 1977:174). In later Latin attempts were made to reintroduce the sound intervocalically, where it appears spelt with <CH> or <C> (probably with the value [x] as contemporary Gk <χ>), in forms like michi, nichil (see e.g. Lofstedt 1961; Leumann 1977: 175).

    Both grammarians’ descriptions (e.g. Quint. 1.5.19: spīritus ‘breath’; Mart. Cap. 3.261: aspīrātiō ‘aspiration’), and the origins of Latin <H> (from Gk <h> [sic]), suggest that the main realization of Latin /h/ was as a voiceless glottal fricative [h]. Between vowels, however, it seems likely that /h/ was a voiced glottal fricative [ɦ], at least immediately before the loss of /h/ in this position; Allen (1958: 103–4); Allen (1978: 43, 45).

    The explanation for michi and nichil is as I thought: people tried to reintroduce [h], but couldn’t pronounce it; they opted for the closest model (which fits the grammarians’ descriptions), [x]; people who didn’t know Greek couldn’t pronounce that either, resulting in [k]. …And actually, the arguments for voicelessness in the 2nd-to-last sentence aren’t convincing at all.

    On singing, indeed I haven’t sung in a choir. It makes sense that [h] means you have less breath left to spend on the following vowel than after a stop, so you may not be able to hold it as long. But isn’t [h] still audible as a pause followed by a whole new voice onset? Or are you all talking about [ɦ], which is said to occur in some Englishes (neatly explaining e.g. “an historical”) but which I’m not used to from my German background?

  52. I don’t have [ɦ] as far as I’m aware (nor do I say “an historical”), but I’ve noticed that when /t/ and /d/ precede /h/ in my speech – typically across word boundaries – I use the flapped realizations of those phonemes, which otherwise occur only before vowels. For example, [ˌðæɾ ˈhæt̚] that hat.

    I’m actually curious about what would happen in these cases for someone on the Estuary English spectrum who had /t/-glottalization but not /h/-reduction. I find a realization like [ˌðæʔ ˈhæʔ] rather hard to spit out.

  53. I don’t think I’ve seen anything authoritative on how it’s supposed to work in italian-style ecclesiastical or choral pronunciation, but in classical Latin (e.g.) filii was either [fi:li:] (gen.sg) or [fi:lij:i] (nom.pl). If I read my VL right.

    That is, no word-internal hiatus. And hiatus was eliminated between words as well when scanning Latin poetry, but I don’t know if that applied to everyday (or solemn) speech.

  54. “Put the left-over rice pudding in the refrigerator”
    Wikipedia informs me that the first refrigerator for home use was invented in 1913…

  55. those iī’s that pop up in the second declension.

    Pronounced /ɪji/ in my tradition.

    really important arguments in footnotes

    Linguists are bad, but lawyers are infinitely worse: it is routine to have three lines at the top of a page and the rest all footnotes, the first of which is actually continued from the previous page and the last continued on the next page.

    But isn’t [h] still audible as a pause followed by a whole new voice onset?

    I think the problem is that people don’t (unless perhaps if specially trained) have precise control of their VOT when singing. Without precise control, the sound will be ragged as different singers transition into or out of the voiceless part of the hiatus (which is what intervocalic [h] really is). I note that I don’t aspirate initial voiceless stops, or only very lightly, in either English or German when singing them, though I certainly do when speaking/reciting them (and I don’t flap English /t/ in singing either).

    an historical

    Just a survival, I think, from the period when English lost all initial /h/; it exists today solely as a result of spelling pronunciations, which is why it is variable between accents in specific words. No need to introduce [ɦ] for that. If anywhere, I have it as a realization of /h/ between schwa and a stressed vowel, as in ahead [əɦɛd], but it is not extremely voiced, or extremely fricative either.

  56. Thanks for that long and extremely informative quote, David!

  57. January First-of-May says

    [initial /h/] exists today solely as a result of spelling pronunciations, which is why it is variable between accents in specific words

    …and almost universal (when expected by the spelling) in my own speech a few years ago. As in, I was saying (and writing) “a hour, a honor, a heir”. (I’ve recently, finally, started saying “an hour”, but still pronounce the /h/ in the other two words, because they’re much less common.)

    voiced glottal fricative [ɦ]

    I’m trying to figure out what a voiced version of English /h/ would be like, and end up with the Hebrew letter (ה)… or, at least, the way I was told to pronounce that letter when trying to learn Hebrew as a kid; apparently this is not its actual pronunciation.
    (The traditional Russian transliteration of Hebrew texts uses h for this Hebrew consonant, while all the other Hebrew sounds are transliterated with Russian letters… which, probably unintentionally, happens to make the stand out. I’ve seen a few modern transliterated texts that used ѓ instead – which probably isn’t any better.)

  58. The Hebrew sound is just a voiceless [h], much like the English one but pronounced more weakly (so that it’s very prone to elision, or for some speakers, to merger with glottal stop).

  59. 1/ 5/1: I’m trying to figure out what a voiced version of English /h/ would be like.
    Try making a prolonged h sound and start humming in the middle of it. I also find that when I say a very relaxed ɣ it can drift into an ɦ.

    TR: The Hebrew sound is just a voiceless [h], much like the English one but pronounced more weakly (so that it’s very prone to elision, or for some speakers, to merger with glottal stop).
    I wouldn’t analyze it that way. The glottal stop is not phonemic, in the sense of minimal pairs, I’d just call it an epenthetic sound before syllable-initial vowels. The realization of /h/ as [ʔ] is what I’d call a simple fortition, not a side-effect of lenition.
    (This is off the top of my head. Maybe you have a better analysis.)

  60. I think the Ukrainian /ɦ/ is the traditional way to pronounce any aitch in the Russian context, so our January First-of-May must be having the right ɦ already.

  61. David Marjanović says

    I find a realization like [ˌðæʔ ˈhæʔ] rather hard to spit out.

    …I find [ʔh] rather easy, but I’m probably cheating in what I mean by [h]… I think I only use actual [h] when I whisper; otherwise, I interpret the following voice onset – which is the exotic part of this page (click on the example for a soundfile) – as an indispensible part of /h/ (something that does not happen with any other voiceless consonant). Consequently, my native /h/ barely behaves like a fricative at all; e.g. it cannot be lengthened, which does happen in English singing (ma hhheart will / go-w-oooon…).

    an historical

    Just a survival, I think, from the period when English lost all initial /h/; it exists today solely as a result of spelling pronunciations, which is why it is variable between accents in specific words.

    To some extent, certainly, but that alone doesn’t explain why this phenomenon is only found today with words that are not stressed on the first syllable.

  62. Y: I totally agree about the non-phonemic status of glottal stop in Hebrew. What I meant (and didn’t express clearly) was that speakers who tend to insert a glottal stop between vowels where this is historically justified also often do so in cases where there’s historically an intervocalic /h/, while other speakers tend to elide both historical glottal stop and historical /h/. I don’t think the realization of /h/ as [ʔ] is fortition, but rather a result of loss of /h/ followed by automatic insertion of [ʔ] in hiatus environments.

  63. David Marjanović says

    I can’t believe I keep forgetting this, even though we’ve discussed it at least once before: whatever happened to mihi and nihil also happened to the name of the letter H, which was most likely never sung. Apparently it was /aha/ – the Western Romance (and English) names point at */aka/ instead. In German, BTW, it’s /haː/; in other words, H is treated like a plosive.

  64. why this phenomenon is only found today with words that are not stressed on the first syllable

    Because it’s very hard, in English, to pronounce [әhә-] without stress. As I pointed out some time ago on one of these blogs, my mother (b. 1920) used to say “an hibachi” for a word that certainly wasn’t part of her childhood vocabulary.

  65. Well, not very hard; I do it myself routinely, and I am a lazy man.

  66. never sung

    The Romans may have had an alphabet song, just like anglophones and germanophones (though probably not sung to “Ah! vou dirai-je, Maman”, which is not recorded before the 18C).

    treated as a plosive

    If it were, it would be /heː/. I suspect /haː/ echoes /kaː/, as in English /dʒeɪ/ echoes /keɪ/.

  67. Z, Y, X, W, V, and U,
    T and S and R and Q,
    P, O, N, M, L, K, J,
    I, H, G, F, E-D-C-B-A,
    Now I know my ZYXs,
    Next I’m going to march through Texas.

    It helps to pronounce W as “dubya”. Note the rhythmical similarities between “E-D-C-B-A” and “L-M-N-O-P”, and between “T and S and R and Q” and “W, X and Y and Z”, both devices to get the rhymes to align with the line ends.

  68. David Marjanović says

    Awesome. Too bad “L, K, J” can’t be replaced by “LBJ”.

  69. It should work in German to shift J to the fourth line and say “L und K / J-I-H, G, F, E-D-C-B-A”, provided you use the nonce name /y/ for Y, which doesn’t seem all that weird to me. (Note to non-germanophones: In the normal German alphabet song, W shifts up to the third line; of course “Weh” does not rhyme with “Zed”, but then “vee” doesn’t rhyme with “zed” either.)

    I leave the last two lines up to you: I don’t even know what the plural of X is, never mind how to rhyme to it.

  70. “Hey, I got an uncle lives in Taxes.”
    “No, I’m talking about taxes. Money, dollars.”
    “Dollars, that’s a-where my uncle lives! Dollars, Taxes.”

  71. January First-of-May says

    Now I know my ZYXs,
    Next I’m going to march through Texas.

    Reminds me of this old Times Like This strip (which I just spent thirty minutes finding in the archives).

    Incidentally, “going to march through Texas”, in quotes, gets exactly two Google hits, both of them involving actual marching in Texas. (Though apparently the song does exist, and some versions of it do reference Texas. Some do not. I liked the one with “me with sing you won’t time next” the most, personally.)

  72. The folk process at work, evidently. I knew the rhyme was “ZYXes / Texas”, but not the exact wording of the last line. Still, I like this version.

  73. David Marjanović says

    Zed

    -t, not -d; I can tell, because I don’t do final fortition. 🙂

    I don’t even know what the plural of X is

    I’d resort to not distinguishing it from the singular. The traditional rhyme for it is nix, dialectal-and-spreading for nichts.

    “Dollars, that’s a-where my uncle lives! Dollars, Taxes.”

    Are there non-rhotic Texans? 🙂

  74. January First-of-May says

    Are there non-rhotic Texans? 🙂

    There are certainly people with uncles living in Texas who don’t pronounce the R in “dollars”.
    I can be sure of it because I used to be one of them a few years ago (he’s in California now).

  75. Zet, is it? Okay, what happened to the final vowel of zeta? In French it became zède, with the usual intervocalic lenition, final /a/ > /ə/ (in Old French when all other final vowels were lost) and lastly the loss of final /ə/, which accounts for English zed (American zee is the product of analogy).

    Of course there are non-rhotic Texans: in East Texas (which is part of the coastal South linguistically) and African Americans everywhere. But in this case we are dealing with Chico, who wrecks the phonological and morphosyntactic levels of English as emphatically as Groucho wrecks the semantic level. (Harpo doesn’t do language at all.)

  76. I don’t think the realization of /h/ as [ʔ] is fortition, but rather a result of loss of /h/ followed by automatic insertion of [ʔ] in hiatus environments.

    You may be right. I am trying to think of a diagnostic example, and unfortunately I am not near Hebrew speakers at present. Would you say that /hahar/ ‘the mountain’ is pronounced by some [a.ar], by others [ʔaʔar]? Would some say [aʔar]? Does anyone pronounce /beit hasefer/ ‘the school’ always as [beit ʔasɛfɛr] instead of [beitasɛfɛr]?

  77. Would you say that /hahar/ ‘the mountain’ is pronounced by some [a.ar], by others [ʔaʔar]? Would some say [aʔar]?

    I’m not sure either what the exact pattern of variation is. I think the picture is complicated by things like whether an initial vowel is also utterance-initial: e.g. [ʔaʔar] might be possible at the beginning of an utterance but not likely otherwise. Even then I think [aʔar] would be more likely; that’s how my sister, who basically doesn’t have [h], says it. (Maybe the /h/ of the definite article is more easily elided than other /h/s?). But I don’t have any real data.

    Does anyone pronounce /beit hasefer/ ‘the school’ always as [beit ʔasɛfɛr] instead of [beitasɛfɛr]?

    I’d be extremely surprised if that was the case.

  78. I think Chico’s accent was broadly supposed to resemble that of an Italian immigrant in New York.

  79. Yes. Apparently Chico picked it up as a boy to ward off antisemitism. My grandson does the same thing, but it’s because he loves Mario and Luigi (warning: TVTropes!) so much. It’s very entertaining to hear a small black boy on a NYC playground speaking with an Italian accent. I wonder what, if anything, playmates and other parents think. He doesn’t do it all the time, just sometimes.

    His nickname was originally “Chicko” because he was a chick(en)-chaser. A typo in a vaudeville playbill changed all that, but the family kept the pronunciation.

    In Duck Soup, when he is disguised as Groucho, he retains his Chico accent. When Margaret Dumont (who did get the jokes) asks him what happened to his voice, he says he’s practicing up for going to Italy one day. She replies dryly, “Your dialect is perfect”.

    “The Foreign Hand Tie” by Randall Garrett (author of “Look Out! Duck!”, the Reviews in Verse, and countless other travesties) for those who love both science fiction and the Marx Brothers. In Soviet Russia, joke-ACC tells you-NOM!

  80. In later years, Chico himself was insistent about the correct pronunciation of this stage name. Groucho was not, and, in fact, he liked to have people from Chico, California on You Bet Your Life so he could make jokes about his brother.

  81. David Marjanović says

    Zet, is it?

    The one time I’ve seen it written, it was Zett. English is unusual in spelling out the names of letters so easily and often.

    …BTW, in Austria but not (much of?) Germany, the chemical endings -id and -it are distinguished not only by fortis/lenis, but also by vowel length: -id is long. (Both are stressed.) Lenes aren’t enough to really close a syllable, I guess.

    Okay, what happened to the final vowel of zeta?

    No idea. Analogy from the other monosyllabic letter names? (Even W, as you’ve mentioned, is just /veː/.) Perhaps from J in particular? (/jɔt/ is most widespread, and every time I use my native /jeː/ to explain the spelling of my name, people here in Berlin write down a G.)

  82. Actually, the names of the letters are extremely unusual for English in having no standardized spellings, only ad hoc ones, which means that they are particularly juicy examples of pure sound change at work uninfluenced by the written form.

  83. Well, except for zed. (Though Shakespeare did spell it with two d’s: “Thou whoreson Zedd, thou vnnecessarie letter.”)

  84. I’m familiar with the Zett spelling from Eszett for ß, though I’m not sure where I saw it written.

  85. David Marjanović says

    “Thou whoreson Zedd, thou vnnecessarie letter.”

    *pretends being able to raise one eyebrow*
    Fascinating.

  86. In the first Men in Black movie, one of the gags was that all the named Men in Black had names that were letters, but which could also be given names: J, K, L, and Z (pronounced Jay, Kay, Elle, and Zed). I assume this joke originated in the comics, but it was abandoned in the the later films and other media.

  87. David Marjanović says

    nix, dialectal-and-spreading for nichts

    It belatedly occurs to me that this may be an exact analog for how nihil could come from a Faliscan version of *ne filum as mentioned above.

  88. David Eddyshaw says

    From a quick search of the Vulgate it looks like Latin speakers weren’t quite sure what to do with such foreign city names: there’s venerunt in Hebron, but also venit Nazareth, as an allatival accusative

    @TR: Because it’s never too late to reply on LH, and it hasn’t even been four years:

    Kusaal actually has a very similar problem; Kusaal place names are intrinsically locative and don’t take the locative postposition n~ni which is required for other words used as place expressions:

    Ba bɛ zuoya la ni.
    “They are in the hills.”

    but

    Ba bɛ Bɔk.
    “They are in Bawku.”

    In conversation, most people treat foreign place names the same, but the Bible translations don’t; they usually resort to periphrases like

    Ba bɛ Jerusalem teŋin.
    “They are in Jerusalem-town.”

    where the -n is an enclitic allomorph of the locative postposition, despite the misleading orthography.

  89. That reminds me of Kazakh, where locative Almatyda is possible, but at least in the texts I’ve seen, Almaty qalasynda “in the city of Almaty” is much more frequent.

  90. David Marjanović says

    I wondered if that’s calqued from Russian, so I tried to find out if Almaty is indeclinable there.

    It was usually declinable in the 1850s, but other than that, the ru.WP article says the form Alma-Ata (declinable) is still used in Russia, so that’s the form it uses.

    This page does use it, as undeclinable, but without prefixing город.

Speak Your Mind

*