Precipitation Verbs.

Beth Levin of Stanford University has an interesting 2017 paper called Talking About the Weather: A Case Study of Precipitation Verbs which begins with a quote from Ronald Langacker:

In the eyes of linguists, such [=weather] expressions are nearly as problematic and ill-behaved as the weather itself: they not only have many special properties, but from one language to the next the same phenomenon is coded linguistically in ways that are lexically or grammatically quite distinct.

She proceeds to discuss “several challenging properties of weather events”:

Identifiability: Depending on the metereological phenomenon, it can be difficult to identify any participants in the event (e.g., becoming dusk). Perhaps it is possible to recognize a single participant (e.g., snow, rain).
Independence from the phenomenon: The participant, to the extent it is identifiable, is not independent from the phenomenon itself: snow and rain do not exist outside of the event of snowing or raining […].
Selectional restrictions: When a participant is expressed, weather verbs impose fairly strict selectional restrictions on it: outside of metaphorical uses, only snow can snow, only rain can rain. […]
Semantic role: It is hard to determine what semantic role to assign to this participant: when rain rains from the sky, is it acting or being affected?

She brings in Eriksen et al.’s suggestion that “weather event expressions fall into three major types according to which element in the sentence lexicalizes the ‘weather’ phenomenon”: the predicate type (It is raining), the argument type (Rain is falling), and the argument-predicate type (It is raining rain). She then goes on to her own proposed solution for encoding weather events; there is some Chomskyism, but not enough to set off my alarms and make me want to throw the computer against the wall. (We discussed precipitation expressions to some extent in last year’s word-order thread, starting here.) Thanks, Martin!


  1. David Eddyshaw says


    The Chomskyism is manifest in the characteristic tendency to attribute differences which are actually fundamentally semantic to invisible structural/syntactic differences. (For example, the fact that the command Rain! is cromulent* but Seem! is not seems to me to flow from the semantic properties of the actual verbs rather than the properties of any potentially preceding “it.” You can’t readily tell a person to “seem”, let alone the weather. Still, Levin doesn’t seem** too hardcore about it, to be fair. Hasn’t drunk the Kool-Aid …

    Interesting that temperature is different from precipitation, though not too surprising (for the very reason that she proposes, in fact.) Kusaal manages to use quite different constructions for “It’s hot” and “It’s cold”, let alone for rain …

    * Technical linguistic term
    ** Sorry

  2. Seem!
    Reminded me Russian покажись! (appear!), кажется “it seems”.
    For English speakers who have the verb, yes. For Russian speakers who have a noun it is less so…

    “flow” – why “follow” and “flow” are so similar? I suspect their synonymous application to facts has nothing to do with English…

  3. David Eddyshaw says


    Kusaal is pretty liberal in allowing all sort of verbs to be used in direct commands, even the always-imperfective sort that express predicative adjectival meanings: you can say Vʋe! “Be alive!” and you can, likewise, use the copula verb in direct commands: An baanlim! “Be quiet!” You can even use “exist” in a direct command: Bɛ kpɛla! “Exist (i.e. stay) there!”

    I don’t think there are any actual syntactic constraints, only pragmatic. I can’t think of many contexts where you might say Wa’am! “Be tall!” for example, but that could just be my lack of imagination. Perhaps when Harry Potter is translated into Kusaal … (though I suppose the spell would have to be in Bad Latin anyway …)

  4. David Marjanović says

    “flow” – why “follow” and “flow” are so similar?

    The German cognates are folgen and… well, fließen should be the cognate of float

  5. Or rather покажись is “become visible”. Actually явись!* (also “appear!”) apart of appearing in the sense of “becoming visible” and also in the sense of appearing somewhere or before someone is also used as a copula in the “logical” Russian: A appears B. Subjective and existential are already indistinguishable in appear-disappear.

    *I recently quoted this imperative. In the sense “my angel, with less** clothes on you will shine even brighter”

    ** or fewer?

  6. Giacomo Ponzetto says

    No idea what this might mean to linguists, but to my ear in Italian “Rain!” is not cromulent because atmospheric verbs are so strictly impersonal they don’t have a second person singular. I could bring myself to say something like “Rain, ye arrows!”, but that’s both a plural and not atmospheric. If I am to give a direct command to the weather, it has to be “Start raining!” or some such. Or of course “Let it rain!”, which also allows “Let it seem!”, at least with a predicative.

    By the way, isn’t the predicative the keey to Seem! in English too? Can’t Chomsky tell Levin: “If you cannot be more hardcore, at least seem more hardcore, dammit!” I would, but then I’m no native speaker.

  7. PlasticPaddy says

    Buon anno!
    For your sentence “try and seem” works better for me than “seem”. But I think I would say, “try and appear”.

  8. Well I can imagine “Rain!” in rainmaking. In Russian then it would be лейся! “pour [self]”. Our rains go or sometimes pour.
    seem more hardcore” – English “pretend” is commonly used in such contexts. I guess in Russian where we do not have “pretend” “at least seem/appear it if you aren’t” sounds more natural.
    Weirdly it is “seem [it] if you-not-appear [it]” (because “appear” is the copula) :-/

  9. Hmmm

    In Croatian, there is the verb ‘kišiti’ = to rain, but it’s more usual to say ‘kiša pada’ = rain is falling.

    I suppose you could say ‘kiša kiši’ = rain is raining, but much harder to find in the wild. This covers off on all 3 of Beth Levin’s types.

    As for the imperative, ‘padaj kišo’ would be the ordinary way of expressing it.

    There is also the archaic noun ‘dažd’ and verb ‘daždjeti’.

  10. David Eddyshaw says


    Yes, I was wrong. You can indeed use “seem” in a direct command. Again I wasn’t being imaginative enough.
    (I was just watching some of Martin Hilpert’s interesting Youtube videos about Construction Grammar: while he’s not nearly as bad as the Chomskyites in this regard by any means, he does have a tendency to flag sentences up as ungrammatical which in fact seem perfectly OK to me. The construction-grammar people at least don’t have any ideological hangups about that sort of thing, though. It’s just another Construction to write another paper about …)

    Kusaal paraphrases to avoid direct commands to things that can’t actually hear you: for example you’d have to say

    Kɛl ka saa ni.
    let.IMP and sky rain
    “Let it rain.”

    (it’s an interesting question just who the imperative “let” is actually addressed to here. The ka in constructions like this signals a change of subject, so it’s not the sky, anyhow …)

  11. J.W. Brewer says

    It’s maybe useful to think of “seem” as a copular/copulative/”linking” verb, which as such generally can’t be used w/o a complement. It’s definitely weird to just say “Seem!” But it’s just as weird (absent very specific context, I guess) to say in the imperative “Be!” or “Become!” or “Get!” But that’s not because there’s something special about the imperative because it’s odd to say “she seemed/was/became/got” and then stop the sentence there w/o a complement after the verb. Is obligatoriness of a complement for these verbs a pragmatic constraint or a syntactic one?

  12. David Marjanović says

    to my ear in Italian “Rain!” is not cromulent because atmospheric verbs are so strictly impersonal they don’t have a second person singular.

    Similar in German – it rains, not the weather or the rain or anything, just the dummy subject. Indeed, if the rain has to be said to do something, it doesn’t rain, it falls…

    (But the wind has a verb all of its own, wehen. Except it’s been replaced by gehen “walk” in dialects like mine, perhaps as a phonetic dissimilation of der Wind weht > der Wind geht…)

    The imperative (plural as it happens, but could just as easily be singular) crops up in a church song, but not of “raining” directly – there’s a separable prefix involved: Tauet, Himmel, den Gerechten / Wolken, regnet ihn herab “Dew, heavens, the Just/Righteous One / Clouds, rain him down”, and I think the whole thing is straight from a psalm anyway, so in addition to the poetic register there might be translation issues involved…

  13. J.W. Brewer says

    “Tauet, Himmel” et al. is apparently based on Isaiah 45:8 but is also apparently a Catholic-origin hymn, so presumably no one was looking at how Luthers Bibel handled the verse but instead went straight to “Rorate, caeli, desuper, et nubes pluant justum.” This was a text used liturgically before Vatican II with sufficient frequency during Advent that it gets its own wikipedia article:

    The most Vulgate-dependent English translation (Douay-Rheims) has that as the rather unidiomatic “Drop down dew, ye heavens, from above, and let the clouds rain the just.” The King James has “Drop down, ye heavens, from above, and let the skies pour down righteousness.” But one recent translation (the ESV) has “Shower, O heavens, from above, and let the clouds rain down righteousness.” Verbs found in other translations include “distil,” “drip,” and “sprinkle.” I don’t know if there’s something funky about the underlying Hebrew or whether the weirdness of weather-related verbs in English makes translators struggle. Perhaps David E. can tell us how this verse is rendered in Kusaal?

  14. @J.W. Brewer: I think it’s a syntactic constraint, but that is specific to that verb. For some other verbs, the situation will be different. I conclude this from the fact that: *He gets, sounds entirely wrong, whereas: He gives, is merely odd. (Futhermore, give can be coordinated in ways that get seemingly cannot, or cannot without difficulty. Compare U2’s*

    And you give
    And you give
    And you give yourself away

    which seems unproblematic, to the apparently parallel

    *And you get
    And you get
    And you get what you want

    which seems questionable, at best, to my ear.) It seems like give can drop its direct object entirely, given the right informational context, but get cannot.

    Of course, there are caveats. In these marginal cases, native speakers will not necessarily have identical** intuitions. I chose give for comparison to get, because both are ditransitive but they vary quite a bit in the acceptability of dropping objects. For get, the indirect object is essentially completely optional. Get the box, and, Get me the box are equally fine. It is much harder—although not quite impossible—to omit the indirect object with give, although it can be expressed prepositionally. (There are also ordering constraints that often trip up nonnative speakers, I have found:

    ?Give the book.
    Give me the book.
    Give the book to me.
    *Give to me the book.
    *Give the book me.

    The first one is bad, but it may only be pragmatically bad, not syntactically, rather like He gives.) There are other differences between the verbs too. Get can take a broad range of different complement types with ease, including ditransitively—for example,

    Get going.
    Get him going.

    Give can sometimes take other complements than direct and indirect objects, but it’s much finickier.

    * I really don’t like U2, but hearing “With or Without You” over and over was unavoidable for an American college student in the 1990s.***

    ** This reminded me that I have an anecdote and some thoughts about the word identical. In high school, our computer network had a word processor package on it called ProWrite (which is apparently sufficiently obscure that it does not even have a Wikipedia page). It was very simple, using only ANSI graphics, which meant that it could run on even our oldest and lousiest computers. Once we got rid of those models, I wanted to get rid of the program. (I was a student network administrator.) However, some of the teachers liked it, so we kept it. Among its other crude features, it featured a spelling and grammar checker. The spell check was functional, but the grammar checking was a joke. (I remember that it thought “demon” was an abbreviation for “demonstration.”)

    I did, however, occasionally learn interesting titbits from the grammar checker’s comments. In particular, I remember that it told me that using “identical to” was considered informal, and that the traditional locution was “identical with.” I am not sure I had ever encountered that usage before, but I thought about it and did some further research. I concluded that, under most circumstances, my idiolect definitely called for “identical to” for a statement that two entities were indistinguishable. However, I found that I started using “identical with” for the less-common situation when I needed to point out that two descriptions were actually of the same entity. (For example, “Although the Lie algebra su(2) is compact, and sl(2) is not, they are both simple, and the tensor product of su(2) with C is identical with the tensor product of sl(2) and C,” sounds fine.)

    *** Amidst all the other asterisks, did you notice the one in “U2’s*” was for a footnote?****

    **** I wasn’t sure where to put the triple asterisks for the third footnote, but having made a decision about that one, the positioning of the quadruple asterisks for this footnote was then determined by precedent.*****

    ***** The precedent only holds for this comment. I reserve the right to format my footnotes however I deem best in future communications.

  15. J.W. Brewer says

    Somewhat hilariously, the second verb in the LXX version of the verse is an inflected form (ῥανάτωσαν) of the verb ῥαίνω, which by what is presumably a remarkable coincidence looks in transliteration just like “rain-o” (okay, maybe “rhain-o” would be more consistent with traditional style). But the usual translation is “sprinkle” or “besprinkle.” Nonetheless, English translations from the LXX go with “rain” (Brenton) or “shower” (NETS), making it fit English weather-talk more smoothly.

  16. J.W. Brewer says

    @Brett: Maybe you would prefer this U2-free example that nicely contrasts the verbs (attributed by the internet to Angie Stone): “I give and give, even when I get nothing back – and that sets me up for disappointment.”

  17. J.W. Brewer says

    @Brett: That said, I think bare imperative “Give!” is pretty weird-sounding with the exception of fundraising pitches where the implied direct object is “money.” As in, e.g. “Please give generously.” I think that’s because a euphemistic avoidance of actually saying “money” out loud is locally overriding a default rule that would require an object. Maybe object-less constructions like “I give and give …” create a little more context from which the object can be, however vaguely, inferred than a bare infinitive does?

  18. @J.W. Brewer: That’s a good quote. I didn’t really know anything about Angie Stone, except that she was a hip-hop artist, so I just looked her up, and it turns out that she is originally from here in Columbia, South Carolina. Apparently, she was singing at a mega-church that’s right along my way to work back in the 1970s. Yet I’ve never heard anything about her.

  19. Someone’s (big, dangerous-looking) friend who did not know English at all when he was abroad for the first time impressed this someone by cornering a a fellow inhabitant of the hostel and saying “give mobile!”.

  20. a bare “get!” rings cromulent to me, but only because my lects allow dropping the “out” or “gone” that should follow it, especially when addressing animals and familiar people. which is to say: i think it works for semantic reasons rather than syntactical ones. and i think that’s also true for “seem”, in GP’s example (and both seem to me to support DE’s inaugural comment).

    on the weather side, though, marc blitzstein has the lovely line “the rain was raining rain, the snow was snowing snow” in one of the interstitial sections of The Purest Kind of a Guy*. but that’s a slightly odd phrase to my ears in english, and i suspect it’s calqued from a yiddish phrasing**: רעגן רעגנט זיך / regn regnt zikh, which is ‘it’s raining’, but with a particular valence that i’m finding it hard to articulate, but that’s definitely not the same as ס’רעגנט / s’regnt [‘it’s raining’ / ‘it rains’].

    der Wind weht > der Wind geht
    yiddish has this, too, with just the cognates you’d expect: a vint veyet (or vyeyet, depending on where you are), but i think vint can also take geyn, and regn often geyt.

    it’s an interesting question just who the imperative “let” is actually addressed to here
    and in english too! is the let-er in “let it snow!” the same as the let-er in “[hey! ho!] let’s go!”**?


    * from his sadly underappreciated No For An Answer; probably the best broadway song ever to meld class solidarity and cruising. robeson’s version is of course amazing, but doesn’t have this line (and the original cast album seems to have been scrubbed from youtube).

    ** which can be used with pretty much any verb.

    *** i’m not certain the “us” is who lets, here – my ear parses it with an external enabler (making it parallel to “[mamma mia] let me go”).

  21. vint veyet

    Вей, ветерок (Vey, vyetyerok) ‘Blow, little wind!”; originally a Livonian, and then a Latvian, folksong, and a movie (not in that order; no English versions are available):

    1, 2

  22. PlasticPaddy says

    Agents : sun, wind
    Objects: rain, snow, etc.
    Process: good weather, bad weather, windy weather, shower, storm, etc.
    The agents get to be subjects of verbs. The objects and processes are set in motion by unnamed agents. In Irish rain (fearthainn) is sent down (cuir) -tá [IS] sé [HE] ag cur [SENDING DOWN] fearthainne [RAIN (GEN.)]. You can also say tá sé ag cur baistí [BAPTISM (GEN.)].

  23. Agents : sun, wind
    Objects: rain, snow, etc.
    Process: good weather, bad weather, windy weather, shower, storm, etc.
    The agents get to be subjects of verbs. The objects and processes are set in motion by unnamed agents

    I don’t want to rain on your neat hypothesis, but rain can be a subject, too. Just google “the rain rains”. I think you’re right that weather is a process, but languages just arbitrarily assign sentence structures and constituent roles to it, partly based on syntactic restraints (e.g. English or German normally don’t allow subjectless sentences, so they need a dummy subject “it”), partly based on historical accident.

  24. let it seem.
    it films dogs and cats today in the cinema. It films Star Wars all night long.

    (in Russian both rain and TV “transmissions”/”program[me]s” go)

  25. PlasticPaddy says

    It is clear no schema will fit anything but a few languages. For “the rain rains” other things also rain (down): blows, manna, fire and brimstone, frogs, fish. There might even be two separate verbs: (1) imp. rain (2) trans. rain (down) with (2) formed either from the noun “fall like rain” or by repurposing the verb of (1). I agree this is more complicated than positing a single verb which then became specialised and defective in sense (1).

  26. “You get and you get” sounds perfectly cromulent to me. It parallels “you take and you take”. It has an implication of selfish behavior. I can easily imagine a fight in a relationship with one partner saying “you just get and get and never give”.

  27. I agree this is more complicated than positing a single verb which then became specialised and defective in sense (1).
    But my hunch is that this is what frequently happened historically. Where I know their etymology, rain verbs often originally had wider meanings like “flow”, and in those cases the noun is often deverbal. While in the languages that have constructions with nouns meaning “rain” and verbs like “go, fall” etc., the nouns themselves are either opaque or denominal. I don’t have the time now to check whether that hunch is correct.

  28. David Eddyshaw says

    The Kusaal copula aen, and indeed all transitive verbs of the imperfective-only minor conjugation that it belongs to, like mɔr “have” and wɛn “resemble”, must have a complement. The complication is that this complement can, as a matter of fact, be omitted: however, if it is omitted, the “gap” works like an anaphoric pronoun:

    Mani an na’ab, ka fʋn mɛ aen.
    “I am a chief and you, too, are [a chief].”

    This applies to all obligatorily transitive Kusaal verbs; to get rid of the implied anaphora, you have to supply an actual generic object: thus, “Thou shalt not kill” has to be

    Da kʋ nida.
    “Don’t kill a person.”

    Da kʋʋ is perfectly grammatical, but it has to mean “Don’t kill him/her/them.” However, zu “steal” is ambitransitive, so you can just say

    Da zuu.
    “Don’t steal.”

  29. January First-of-May says

    In Russian we can say ?дождит “(it) rains”, but it would sound weird and/or very colloquial. I can’t even think of an equivalent form for snow.
    Normally rain goes (идёт), and so does snow; snow can also fall (падает), and I think in principle so can rain, but it would also sound unusual.

    OTOH моросит “(it) drizzles” is perfectly idiomatic and any other way of saying it would be weird. Go figure.
    (…The least weird alternative option would probably the bare noun морось “drizzle”, as in на улице морось – an option that is also applicable to rain and snow.)

  30. My impression is that дождит means “(it) rains continually” rather than just “(it) rains,” but I’m no expert.

    And then there’s Kornei Chukovsky: “А и день за день будто дождь дождит…”

  31. I have heard, thought and may be used (I not sure here) дождит a number of times, but mostly as a gloss for English “rains”.*

    The thing is, such impersonal verbs are funny, and are constant source of language jokes. The usual translation for ‘Twas brillig is варкалось “’twas brilling”. We say меня тошнит “is sicks me” and as teenagers, when the worst of us start doing it often, we also joke “it vomits me”.

    * P.S. дождит does exist in literary Russian – and likely in a meaning slightly different from идёт дождь, or why else have two expressions?
    But people around me mostly use it for fun.

  32. I can’t think of a situation where “Rain!” (as a full phrase) could be interpreted by a listener or reader as definitely a verb and not a noun. If a person is saying it because they want rain to appear, the noun works too.

    I can, though, imagine someone saying “Rain, will you!”. And “Rain softly!”. Okay, I can’t imagine that in the real world, by I can picture a magician in a fantasy novel saying it.

  33. David Marjanović says

    “Rorate, caeli, desuper, et nubes pluant justum.”

    *facepalm* So that’s what rorate means.

    The switch from 2pl imperative (rorate) to 3pl present subjunctive (pluant) is interesting. But that can’t be rendered in German, where the 1/3pl is identical in the present indicative and the present subjunctive except for sind vs. seien.

    I think that’s because a euphemistic avoidance of actually saying “money” out loud is locally overriding a default rule that would require an object.

    I think give replaces donate rather mechanically here, and the direct object of donate is almost always money, so much easier to omit.

    רעגן רעגנט זיך / regn regnt zikh

    Interesting. Informal German in at least parts of Austria distinguishes spielen “play” from the nonstandard sich (mit etwas) spielen “play around” (with something). It’s been blamed on Czech without any details. (Blaming Czech is generally a good idea in Vienna, of course.)

    a vint

    The indefinite article with mass nouns to express “some/any” is a Bavarian feature; the way to say “it’s pretty windy right now” comes out as es geht recht ein Wind in my dialect, apart from phonetics.

    i’m not certain the “us” is who lets, here – my ear parses it with an external enabler (making it parallel to “[mamma mia] let me go”).

    In northern German, the lack of a 1pl imperative (or recognizable subjunctive for that matter) is compensated with “let’s” like in English, but whether “let” comes out as the singular or the plural imperative really depends on how many people you’re talking to: “let’s leave” comes out as lass uns gehen or lasst uns gehen.

    (In southern German, the problem is solved like in French, by taking a question and giving it another intonation: Gehen wir.)

    snow can also fall (падает), and I think in principle so can rain, but it would also sound unusual.

    Contrast Polish, where Pada. is a complete sentence and means “it’s raining”.

    (…I haven’t been to Poland in winter, so I don’t know if it can also mean “it’s snowing”.)

  34. The switch from 2pl imperative (rorate) to 3pl present subjunctive (pluant) is interesting

    It closely tracks the Hebrew (הַרְעִ֤יפוּ שָׁמַ֙יִם֙ מִמַּ֔עַל וּשְׁחָקִ֖ים יִזְּלוּ־צֶ֑דֶק), where the switch seems to have to do with the chiasmus. I don’t think Biblical Hebrew allows a structure like “and [vocative] [imperative]”; in fact I’m not sure it generally allows “[imperative] and [imperative]”, except with the meaning “do verb 1 and you will as a consequence necessarily do verb 2”, as in Do this and live.

  35. David Marjanović says

    That’s even more interesting!

  36. distinguishes spielen “play” from the nonstandard sich (mit etwas) spielen

    i didn’t think to say before (it was late) that the “[infinitive][conjugated verb] zikh” construction is an odd use of “zikh” that’s as much an impersonal-actor marker as a reflexive, and gets used with verbs that don’t usually take zikh.

    here‘s a lubavitsh hasidic song that shows it off in the first line:

    עסן עסט זיך, טרינקן ט‮רינגט זיך, װאָס זאָל מען טון אַז עס דאַװענט זיך נישט?
    esn est zikh, trinkn trinkt zikh, vos zol men tun az es davent zikh nisht?
    usually translated as “eating is easy, drinking is easy, what should a person do when praying isn’t easy?”

    (and here in a version by psoy korolenko that does his signature bouncing between yiddish and russian)

  37. David Eddyshaw says

    The Kusaal version of Isaiah 45:8 starts

    Yanam saŋgbana banɛ bɛ agɔllɛ, niimi
    “You heavens which are above, rain!”

    … which neatly shows that at least according to the translators’ Sprachgefühl there’s no problem with using the verb ni “rain” in a direct command in Kusaal. I suppose that’s not surprising if you can directly address the sky, as that’s the source of rain in Kusaal idiom (saa niid nɛ “the sky is raining.”)

    I don’t know how cromulent this would be for a Kusaasi who hadn’t encountered European conceptions of such things, though. There doesn’t seem to be any precedent for talking to inanimate objects in traditional culture (trees don’t count. They’re animate.)

    Anyhow, it seems to confirm my feeling that there are no syntactic restrictions on using verbs in direct commands in Kusaal, only pragmatic ones.

  38. I see no regional difference between “Lasst uns gehen” and “Gehen wir”. The difference for me is that “Lasst uns gehen” sounds a bit formal and old-fashioned (as kids we were told to use this construction to translate Latin subjunctives like gaudeamus). “Gehen wir”, on the other hand, for me is perfectly normal and idiomatic spoken and written German. Without checking, I am certain that you will find this construction in the works of North German writers as different as Thomas Mann and Arno Schmidt. I suspect the “lasst uns” construction is one of the reasons Arno Schmidt claimed philologists as translators from Latin or Greek suffered from “stilistische Phimose”.

  39. David Eddyshaw says

    The Welsh versions of Isaiah 45:8 (both 1588 and Y Beibl Cymraeg Newydd) tell the heavens to defnynnu “drip” before shifting into third-person imperatives (which are a thing in Literary Welsh.)

  40. David Marjanović says

    The difference for me is that “Lasst uns gehen” sounds a bit formal and old-fashioned

    I’ve heard people actually talk like that in informal situations. I think they were all from Berlin or north of there. To me it’s alien enough that I wouldn’t use it even in formal writings or translations.

    (I wasn’t taught to use it to translate Latin either*, even though I was taught other oddities. Gleichsam as the mechanical replacement for quasi comes to mind; in real life, quasi has been borrowed as is, though in many situations praktisch or the perfectly native sozusagen are more idiomatic, and gleichsam is hardly in anyone’s active vocabulary.)

    Admittedly, none of this means that the gehen wir construction doesn’t occur anne Waterkant; I wouldn’t find it noticeable, so I could easily have been overlooking it.

    * Gaudeamus igitur, iuvenes dum sumusFreuen wir uns also, solange wir jung sind.

  41. David Marjanović says

    Perceptions of formality are funny things in German anyway. In northern Germany (and this is documented for a change; I think it’s in the dtv Deutscher Sprachatlas), Apfelsine is colloquial (compare sinaasappel next door in Dutch), and Orange is formal; in southern Germany, it’s the other way around! In Austria, finally, Apfelsine is practically unknown and never used, so it’s perceived as neither formal nor informal but as exotic.

    (…though orange juice seems to be a variably pronounced Orangensaft everywhere. …Or, in northern Germany, O-Saft.)

  42. The Septuagint has third-person imperatives for both verbs, but the first is εὐφρανθήτω “let it (the sky) rejoice”, which makes no sense as a translation of הַרְעִ֤יפוּ. I wonder if this reflects a different source text than the Masoretic version.

  43. J.W. Brewer says

    @TR: I tend to think “were looking at different Hebrew than the MT has” is usually a better bet for that sort of LXX variation than “were just making stuff up for no apparent reason,” when you have a passage where “probably trying to translate the MT but in a way that shows they just didn’t understand it very well” isn’t an available option. Obviously if the MT is sufficiently difficult/obscure at a given point, the “no apparent reason” qualification may drop out and you get the more plausible “guessing in the dark about what might make sense in the larger context of the difficult bit, given that they can’t actually figure out the Hebrew.”

  44. This is odd because the meaning of רעף r‘p is not obscure. It occurs in four other contexts in the Old Testament, and the parallel in this verse is straightforward as well.

  45. J.W. Brewer says

    @Y, so that’s evidence making it seem more likely that the LXX translator(s), working several centuries before the Masoretes (largely) succeeded in eliminating manuscript variation in Hebrew, was working from a Hebrew text of Isaiah that didn’t say what the MT says.

  46. There seems to be quite a lot of such variation in Isaiah:

  47. Again, it’s very odd. Parallel couplets are pretty much universal in biblical poetry, including Isaiah. The explanation would be that LXX had access to only a corrupted version of this text, which broke the parallelism, and didn’t try to fix it.

    An interesting question (which I don’t know if anyone has the answer to), is: given that the LXX translators were probably just as aware of scribal errors as modern scholars, and given that they must have had access to many more variants of the texts, did they practice a form of textual criticism, and try to correct apparent scribal errors in the texts which were available to them?

  48. J.W. Brewer says

    @Y: Ah, but you are perhaps presuming that the Masoretes necessarily got it right and the variant Hebrew reading the LXX translators may have been looking at instead must have been corrupt because inconsistent with the MT? The parallelism in the MT is certainly commonplace in OT texts, but is the lack of it in the LXX reading (and, by hypothesis, in its Hebrew Vorlage) so unusual as to be inherently suspicious?

    What I find fascinating about the LXX is that it’s the only translation of Scripture which can be confidently asserted (based on the New Testament, at least) to have Divine blessing, or at least Divine acquiescence, but in all sorts of ways it’s a really lousy translation, according to the theories of translation generally extant among modern translators of the Scriptures. So do the moderns know something the LXX translators didn’t, or is it the other way around?

  49. I very much think the lack of parallelism (and consistency with the rest of the verse) are so unusual as to be inherently suspicious.

    TR, how would you translate the whole verse from Greek into English?

  50. The Septuagint has:

    εὐφρανθήτω ὁ οὐρανὸς ἄνωθεν καὶ αἱ νεφέλαι ῥανάτωσαν δικαιοσύνην ἀνατειλάτω ἡ γῆ ἔλεος καὶ δικαιοσύνην ἀνατειλάτω ἅμα ἐγώ εἰμι κύριος ὁ κτίσας σε

    “Let the sky rejoice from above and let the clouds sprinkle justice, let the earth bring forth mercy and let it bring forth justice together/at the same time, I am the Lord who created you”

    The Masoretic text:

    הַרְעִ֤יפוּ שָׁמַ֙יִם֙ מִמַּ֔עַל וּשְׁחָקִ֖ים יִזְּלוּ־צֶ֑דֶק תִּפְתַּח־אֶ֣רֶץ וְיִפְרוּ־יֶ֗שַׁע וּצְדָקָ֤ה תַצְמִ֙יחַ֙ יַ֔חַד אֲנִ֥י יְהוָ֖ה בְּרָאתִֽיו׃

    So there are several differences — not just raining vs. rejoicing, but the Greek leaves out the earth “opening”, and “created you” vs. “created it” at the end. (What is “it”? I actually don’t fully understand the syntax of the Hebrew, which is why I’ve left it untranslated — what’s the subject of תַצְמִ֙יחַ֙ “cause to grow”?)

  51. I tried to post a comment twice but it’s vanished both times — Hat, can it be restored?

  52. Done!

  53. The syntax looks indeed garbled, but the hif’il in תַּצְמִיחַ probably carries the intransitive meaning, ‘will be brought forth (as a plant)’.

    The Great Isaiah Scroll of Qumran reads, as far as I can tell:
    הריעו שמים ממעלה ושחקים ייזל צדק האמר לארץ ויפרח ישע וצדקה תצמיח

    “Cry out, skies from above, and let clouds (pl.) sprinkle (sg.) justice; he who says to the earth, and let salvation bloom, and let justice be brought forth”, or something like that.

    This is garbled in yet a third different way from both the Masoretic and the LXX version. Maybe the LXX source also had the verb רוע rw‘ ‘to cry out, to cheer’, and they did with it what they could.

    It looks to me like several poor versions, riddled with scribal errors and lacunae, were cobbled together in different ways to form this verse.

  54. I very much think the lack of parallelism (and consistency with the rest of the verse) are so unusual as to be inherently suspicious.
    What I remember from my reading about New Testament textual criticism for my master’s thesis all those years ago is that, in Gospel textual criticism, for situations where some variants have parallelism (inside the specific gospel or with other gospels) and others don’t, the preferred explanation is the other way round – the non-parallel variants are seen as older, because scribes tended to introduce parallelism where there wasn’t, because they expected it or because they had memorized the parallel wording of the other parts of the text and that overrode the text they actually were supposed to copy.
    Basically, if one variant makes more sense or is more consistent than another, chances are that this is due to emendation 🙂

  55. David Eddyshaw says

    Proclivi scriptioni praestat ardua.

  56. @January 1 V, re Snowing:

    In Croatian, we have ‘sniježiti’ for ‘snowing’.

    Like rain, snow can also fall: ‘pada snijeg’, or it can snow snow ‘sniježi snijeg’ or it can simply snow ‘sniježi’.

    I’m surprised that Russian wouldn’t have a snowing verb. After all Rusdians experience it more than Croatians.

    In other precipitation verbs, in Croatian rain is also associated with the verb ‘pljuštiti’ eg. ‘kiša pljušti’ = it’s pouring. Though unlike English, the verb ‘pljuštiti’ can only be used for precipitation in the literal sense.

    Other verbs are ‘lijevati’ (pour) and ‘rositi’ (sprinkle or drizzle).

  57. @zyxt sniježiti (in Russian it would be snéžit’ or snežít’) is something I would be able to form immediately if I need a snowing verb. I am confident that you will find it not only in dialects but in the literary corpus, even though I am not sure if I ever read or heard it. And we use zasnežennyj “covered with snow (about fields, plains, forests: spaces rather than objects)”, literally “snowed over”, a deverbal adjective (or a passive participle) implying *zasnežit’ “to cover [space] with snow”.

    But in modern literary Russian we usually say “rain goes”, “snow goes”, “snow falls”, also we have nouns for snowfall (one for snowfall, many for snowfall with wind).
    As I said, this “go” is also applied to shows, films, TV programmes, etc. Even to “work”. Maybe snowing verbs weere simply consumed by this system of describing processes that we observe with “go”.

    Yet we can say “it pours” (about rain), it is expressive and usually followed by adverbial “…as if from a bucket” or “for Nth hour in a row already”.
    For snow it would be “it sweeps” I think, if it is violent enough.

  58. For the Masoretic text הַרְעִיפוּ שָׁמַיִם מִמַּעַל “Drop down, ye heavens, from above” with harʿîp̄û vs. Qumran הריעו שמים ממעלה “Shout, ye heavens, from above” and Septuagint εὐφρανθήτω, both with the equivalent of a Masoretic hārîʿû, it is interesting to compare the passage Isaiah 44:23, for which the Masoretic text has הָרִיעוּ תַּחְתִּיּוֹת אָרֶץ hārîʿû taḥtiyyôṯ āreṣ “shout, ye lower parts of the earth!” (Qumran הריעו תחתיות הארץ as far as I can read it, so essentially the same):

    רָנּוּ שָׁמַיִם כִּי-עָשָׂה יְהוָה, הָרִיעוּ תַּחְתִּיּוֹת אָרֶץ, פִּצְחוּ הָרִים רִנָּה, יַעַר וְכָל-עֵץ בּוֹ: כִּי-גָאַל יְהוָה יַעֲקֹב, וּבְיִשְׂרָאֵל יִתְפָּאָר

    εὐφράνθητε, οὐρανοί, ὅτι ἠλέησεν ὁ Θεός τὸν Ἰσραήλ· σαλπίσατε, τὰ θεμέλια τῆς γῆς, βοήσατε, ὄρη, εὐφροσύνην, οἱ βουνοὶ καὶ πάντα τὰ ξύλα τὰ ἐν αὐτοῖς, ὅτι ἐλυτρώσατο ὁ Θεὸς τὸν Ἰακώβ, καὶ Ἰσραὴλ δοξασθήσεται.

    “Sing, O ye heavens; for the Lord hath done it: shout ye lower parts of the earth: break forth into singing ye mountains, O forest and every tree therein: for the Lord hath redeemed Jacob, and glorified himself in Israel.”

    Here, in 44:23, the Septuagint uses εὐφραίνεσθαι to render a form of Hebrew רָנַן rānan “cry out for joy” (KJV “sing”) and falls back on σαλπίζειν “to trumpet” to render a form of חֵרׅיעַ hērîᵃʿ “shout in triumph or joy” (if such were the readings of the Hebrew text that the translators had before them). Maybe interference from this passage is what disturbed an original parallelism of rain and fecundation imagery in the antecedent text of Septuagint 45:8.

  59. I am confident that you will find it not only in dialects but in the literary corpus

    Yes, снежи́ть is uncommon but definitely exists. Some literary examples:

    ― Снежит, Григорий Евлампыч, ― сказал он, почтительно кланяясь швейцару.
    –Aleksandr Ertel, Гарденины (1889)

    Он рехнулся. «Ходит и думает, что снежит.»
    –Andrei Bely, Кубок метелей (1907)

    Воскресенье. Ветер, небольшой мороз, снежит немного.
    5-ое февраля. -5°, пасмурно, чуть снежит.
    Ходим и мы, со страхом, с тоской. Второй день снежит и снежит.
    –Aleksandr Boldyrev, Осадная запись (блокадный дневник) (1941-1948)

    Потом снежить стало посильнее, липко и густо
    –Sergei Zalygin, Комиссия (1976)

  60. In Japanese, precipitation 降る /furu/, that is, it falls/descends:

    雨・雪・霙・霰・雹がふる /ame/yuki/mizore/arare/hyō ga furu/ ‘rain/snow/sleet/hail falls’
    Rain can also 降り注ぐ /furisosogu/ ‘pour hard’.
    All can 不利募る /furitsunoru/ ‘fall with increasing intensity’ or 降り頻る /furishikiru/ ‘fall incessantly’.

  61. I had a look at an interesting article: Arie van der Kooij, Isaiah in the Septuagint, in C. C. Broyles & C. A. Evans, Writing and Reading the Scroll of Isaiah: Studies of an Interpretive Tradition, Brill, 1997, vol. 2, pp. 513–529. The subject of the LXX translation of Isaiah had been treated by two older, pre-Qumran monographs. Relevant points he makes are: Isaiah is unique within the LXX, as a case of relatively free translation, motivated at places by the aim of producing good Koine Greek, more so than an accurate translation; by an imperfect knowledge of Hebrew; and, where prophecies occur, by “actualizing” them, that is by rephrasing them to refer to events and situations of the translator’s time.

    The article opens some rabbit holes to explore, if I wanted to form a better idea of interpreting 45:8, but I’ll stop and say that all three versions show evidence of corruption, some accidental (misreadings), some intentional (homogeneization). Both Hebrew versions have several syntactic blemishes, which cannot be fixed without creating others.

    Getting back to rain, I’ll say that whoever wrote the Masoretic version used the verb r‘p ‘flow’ as parallel with nzl ‘drip’, referring to metaphorical rains bringing up metaphorical plants. Whether that was in the original lost text is beside the point.

  62. 不利募る should be 降り募る.

  63. Trond Engen says

    What bothers me in the quoted passages from Isaiah is the repetition of the exact same word for “justice” with different roles in the metaphor. “Let justice rain from heaven, so that salvation/mercy/grace will blossom and justice grow on earth.”

  64. They are different derivations from the same root: צֶדֶק ṣedeq and צְדָקָה ṣᵊdāqā. The two meanings overlap in BH, and I can’t easily set them apart,and I wouldn’t be surprised if there is diachronic variation in the meanings (they are distinct in later Hebrew, where ṣᵊdāqā is used for ‘charity’). That indeed could suggest a text cobbled from two sources.

  65. Trond Engen says

    Distinctly different senses/types of justice in the original would reduce my botheredness significantly, and it could even make it a figura etymologica. This is what I actually was expecting to find, but I obviously didn’t look close enough in my attempt to spell my way through the Hebraic script.

    [“Spell” is more than charitable. “Count and compare” is more like it.]

Speak Your Mind