"may" from the OE mæg?

Joelline

Senior Member
American English
Hi,

What is the origin of usage of the modern English word "may," used to express a wish or hope or used in blessings or curses or in expressions such as "may he rest in peace" I thought it was from the Old English "mæg," but I've been told that it is more modern when used in this way.

Thanks.
 
  • modus.irrealis

    Senior Member
    English, Canada
    According to my Old English grammar, mæg was used as an auxiliary with meanings of "to be able to" or "to be permitted to", so it does look like the wish-meaning is more recent. The earliest quote in the OED with this meaning and form seems to be from 1521.

    Historically, it's a subjunctive mood.
    By it, do you mean the use that Joelline is asking about, i.e. that it goes back to the subjunctive? mæg itself is a past-indicative used as a present, with subjunctive mæge.
     

    Joannes

    Senior Member
    Belgian Dutch
    Sorry about the confusion. The use of mæg (infinitive magan I guess?) conveying wish is historically a subjunctive. (I had no idea what the form would be; mæge certainly makes sense.)

    So may on its own didn’t have a wish meaning, this was attributed by the mood. Same for these examples:
    God save the Queen!
    Long live WordReference!
    God bless you!
    ... and most other subjunctives in main clauses (although I’m not sure if people still analyse them as subjunctive forms :confused:)

    I think the subjunctive was well established when modal may evolved. When it did, it could quickly be used in the subjunctive to express a wish.

    (The development of the usage of what are now English modals is pretty well described. Some googling should get you to more detailed information.)
     

    vince

    Senior Member
    English
    According to my Old English grammar, mæg was used as an auxiliary with meanings of "to be able to" or "to be permitted to", so it does look like the wish-meaning is more recent. The earliest quote in the OED with this meaning and form seems to be from 1521.
    1521? That seems pretty late, basically in the Early Modern English period. What did they use before then?
     

    modus.irrealis

    Senior Member
    English, Canada
    Sorry about the confusion. The use of mæg (infinitive magan I guess?) conveying wish is historically a subjunctive. (I had no idea what the form would be; mæge certainly makes sense.)
    Ah, thanks. (And it is magan.)

    1521? That seems pretty late, basically in the Early Modern English period. What did they use before then?
    They would (or at least could) have just used the present subjunctive, like in the examples Joannes gave. For an Old English example, there's abreoðe his angin "may his enterprise fail."
     

    Forero

    Senior Member
    According to my Old English grammar, mæg was used as an auxiliary with meanings of "to be able to" or "to be permitted to", so it does look like the wish-meaning is more recent. The earliest quote in the OED with this meaning and form seems to be from 1521.


    By it, do you mean the use that Joelline is asking about, i.e. that it goes back to the subjunctive? mæg itself is a past-indicative used as a present, with subjunctive mæge.
    I have heard this before about the present being the old past, but I am curious what the present indicative used to be and where the other past tenses came from (mighte, moghte).
     

    modus.irrealis

    Senior Member
    English, Canada
    I have heard this before about the present being the old past, but I am curious what the present indicative used to be and where the other past tenses came from (mighte, moghte).
    I don't know about the original present forms, but it's even possible that they never had them in the first place (at least with the same meaning). One of these verbs, wat "know", is cognate with ancient Greek οιδα which is a verb that is formally in the perfect tense but has a present meaning, also "know." Plus many of these verbs seem to occur throughout the Germanic languages so you'd have to go pretty far back to see how they came about, but I don't know any of the details.

    The new past forms are weak pasts built off of the new present forms, but they don't seem to all that regular. At least I don't see any pattern in examples like (where it's 3rd singular present, plural present, 3rd singular past):

    ah agon ahte "possess"
    cann cunnon cuðe "know how to"
    ðearf ðurfon ðorfte "need"
    sceal sculon sceolde "be obliged"
     

    Joannes

    Senior Member
    Belgian Dutch
    The new past forms are weak pasts built off of the new present forms, but they don't seem to all that regular. At least I don't see any pattern in examples like (where it's 3rd singular present, plural present, 3rd singular past):

    ah agon ahte "possess"
    cann cunnon cuðe "know how to"
    ðearf ðurfon ðorfte "need"
    sceal sculon sceolde "be obliged"
    I would guess ahte is regularly suffixed as a past tense. The others are probably strong verbs requiring vowel change for the past tense. The last two seem to be of the same class, the second one of another (compare begin-began/sing-sang and drive-drove). Cann is probably a real irregular third person form; compare Dutch kan - kunnen - kon (same forms) of kunnen 'to be able to'.
     

    Whodunit

    Senior Member
    Deutschland ~ Deutsch/Sächsisch
    I'd say that mæge would be the correct subjunctive form for sentences like these:

    May you find a good girl!
    May all your dreams come true!
    ...

    In German, we still use the subjunctive (Konjunktiv I) form of mögen to express this idea:

    Mögest du eine gute Magd finden! (I'm trying to make it sound archaic, hence it doesn't very good ;))
    Mögen all deine Träume wahr werden!

    The indicative mæg would be expressed as mag in German, which is wrong for such wishing sentences:

    Magst du eine gute Magd finden! :cross:
    I can't use the second example here, because it would be "mögen", too, but in the indicative

    ______


    By the way, the word kennen (to know) has become semi-regular (there must be a better term for verbs like kennen, brennen, nennen, rennen etc.) in German, because it gets the typical regular forms -te and ge-t, but changes the stem vowel:

    kennen - kannte - gekannt
     

    modus.irrealis

    Senior Member
    English, Canada
    I would guess ahte is regularly suffixed as a past tense. The others are probably strong verbs requiring vowel change for the past tense. The last two seem to be of the same class, the second one of another (compare begin-began/sing-sang and drive-drove). Cann is probably a real irregular third person form; compare Dutch kan - kunnen - kon (same forms) of kunnen 'to be able to'.
    Yeah, the present forms are all old strong forms and my grammar even lists them according to their ablaut class (although it look like in some of the verbs like ah the ablaut has been erased), but I was thinking that the new past forms, which I understand are all weak pasts, would be regular because the weak past is normally so regular, but I can't see the pattern, especially with cuðe. I mean, why does this have a fricative instead of a t or d?
     

    Joannes

    Senior Member
    Belgian Dutch
    Yeah, the present forms are all old strong forms and my grammar even lists them according to their ablaut class (although it look like in some of the verbs like ah the ablaut has been erased), but I was thinking that the new past forms, which I understand are all weak pasts, would be regular because the weak past is normally so regular, but I can't see the pattern, especially with cuðe. I mean, why does this have a fricative instead of a t or d?
    Yes, that's strange. Perhaps it has something to do with the loss of the /n/?

    (By the way, the ending was regularised in the 14th c., apparently.)
     

    modus.irrealis

    Senior Member
    English, Canada
    Yes, that's strange. Perhaps it has something to do with the loss of the /n/?
    I think you're right -- at least, I know about English-Dutch pairs like goose/gans where the /n/ dropped out before a fricative, but that would mean that it was a fricative ð from the get-go which is still strange to me. I don't know anything about the history of Dutch but do you know if was originally a fricative there in Dutch, or at least in the plural past forms which I saw were konden. And same question for the German kannte that Whodunit mentioned.

    (By the way, the ending was regularised in the 14th c., apparently.)
    Thanks -- and a l added by analogy. I wonder if that was purely orthographical or did people pronounce could with an l?
     

    Forero

    Senior Member
    I don't understand either about the nt becoming eth and then ld, but I remember this quote from Chaucer:

    "To ferne halwes, couthe in sondry londes" (To far-off shrines, known in various lands.)

    In Middle English, couthe meant familiar or known; uncouth was the opposite.

    This is from the same Old English word as "could". Also related are kith, know, and ken, as well as can (originally "to know how to").

    One of the Greek words for "know" was oida, which also meant "to have seen". (Explaining the use of a perfect as a present tense.) I can't help but notice "gotta" as an auxiliary, which is the past tense of "get" being used as a present tense of "have".

    But how do we know that maeg (related to mighty and O.E. magan "to be strong") was a past tense rather than a present subjunctive?
     

    modus.irrealis

    Senior Member
    English, Canada
    I don't understand either about the nt becoming eth and then ld, but I remember this quote from Chaucer:

    "To ferne halwes, couthe in sondry londes" (To far-off shrines, known in various lands.)

    In Middle English, couthe meant familiar or known; uncouth was the opposite.
    That's interesting -- uncouth has come a pretty far way from its original.

    But how do we know that maeg (related to mighty and O.E. magan "to be strong") was a past tense rather than a present subjunctive?
    It can't be the latter since it doesn't end in -e, which is the ending of all sing. OE subjunctives. But more positively, it has to be the former because the 1st and 3rd pers. sing. are the same and the plural magon has the -on ending of the strong verb past. Although, I'm looking at its conjugation and there are many odd things about it, but it doesn't seem like there's anything else it can be.

    And I'm sure that mæg is a form of magan and not just related to it, but "to be strong" or "to prevail" was the meaning of magan as an independent verb but this got weakened to "to be able to" and "to be permitted to" as it became used more and more as an auxiliary verb.
     

    Joannes

    Senior Member
    Belgian Dutch
    Thanks -- and a l added by analogy. I wonder if that was purely orthographical or did people pronounce could with an l?
    Such an orthographical pronunciation would only be likely if would and should were still pronounced with /l/ at that time. I don’t think that was the case; pronunciation for both was /ud/ -- which caused to give could an analogue spelling in the first place. By the way, the phrase “where it is historic” seems to imply that it isn’t (and hasn’t become) for could.

    I think you're right -- at least, I know about English-Dutch pairs like goose/gans where the /n/ dropped out before a fricative, but that would mean that it was a fricative ð from the get-go which is still strange to me. I don't know anything about the history of Dutch but do you know if was originally a fricative there in Dutch, or at least in the plural past forms which I saw were konden. And same question for the German kannte that Whodunit mentioned.
    I can’t tell. Dutch and German lost þ (> d) between mid 8th c. and 11th c. There are only few pieces of text from that time that are considered Dutch. Could be interesting to note, though, that a past form with /s/ is sometimes found in Antwerp: kan - kunnen - kost (pl. kosten). I have no idea how that came to be.

    Maybe I should also add the forms of kennen (kent - kennen - kende) ‘to know, to be acquainted with’, which are regular as you can see. Kennen is originally the causative of kunnen -- I’m pretty sure it’s the same for Ger. kennen (caus. of können), by the way.
     

    modus.irrealis

    Senior Member
    English, Canada
    Such an orthographical pronunciation would only be likely if would and should were still pronounced with /l/ at that time. I don’t think that was the case; pronunciation for both was /ud/ -- which caused to give could an analogue spelling in the first place. By the way, the phrase “where it is historic” seems to imply that it isn’t (and hasn’t become) for could.
    Good point. It does look then like it was merely orthographical.

    I can’t tell. Dutch and German lost þ (> d) between mid 8th c. and 11th c. There are only few pieces of text from that time that are considered Dutch. Could be interesting to note, though, that a past form with /s/ is sometimes found in Antwerp: kan - kunnen - kost (pl. kosten). I have no idea how that came to be.
    Interesting and makes things more confusing.

    Maybe I should also add the forms of kennen (kent - kennen - kende) ‘to know, to be acquainted with’, which are regular as you can see. Kennen is originally the causative of kunnen -- I’m pretty sure it’s the same for Ger. kennen (caus. of können), by the way.
    Yeah, now I see that it's the causative. And I guess we can add English/Scots ken, which based on form and meaning, looks like the same formation.
     

    tom_in_bahia

    Senior Member
    South Florida/Phoenix-Tucson/the Adirondacks. Native of North American English
    ah agon ahte "possess"
    cann cunnon cuðe "know how to"
    ðearf ðurfon ðorfte "need"
    sceal sculon sceolde "be obliged"
    I see the following:

    ahte - ought
    cann - can
    cuðe - couthe --> could???
    sceolde - should

    But, what is the modern descent from the third row:
    ðearf ðurfon ðorfte

    This idea of "know how to" being related to can is very interesting to me. I researched the etymology of can versus know and they come from the same original PIE word. Interesting to me is the proximity (at least at first glance) of cunnon with konnen (the Haitian Creole adaptation of the French verb.
     

    modus.irrealis

    Senior Member
    English, Canada
    ahte - ought
    cann - can
    cuðe - couthe --> could???
    sceolde - should
    That's right, and there's also ah > oweand sceal > shall.

    But, what is the modern descent from the third row:
    ðearf ðurfon ðorfte
    This word doesn't seem to have survived in English, but there is a verb tharf/thar in the OED that's labelled obsolete. The word does seem to have survived in Scots, however, where it gave thar.

    Interesting to me is the proximity (at least at first glance) of cunnon with konnen (the Haitian Creole adaptation of the French verb.
    Is konnen from connaître? If so then it's probably a coincidence since the French word involves the prefix con-.
     

    HistofEng

    Senior Member
    USA Eng, Haitian-Creole
    ^^ Yes it is.

    I think Tom was saying it was interesting because the Old English words used for "know how to" (cunnon) and "know" (cnawan) came from the same Proto Indo-European root word, and now in one of the PIE daughter languages (Haitian Creole) "know how to" and "know" are the expressed by the same word with the same roots as "cunnon" and "cnawan"

    one PIE root --> (word diverged into 2 with distinct meanings) cunnon 'know how to' + cnawan 'know' --> (the two meanings, still distinct, converge on a single word etymologically connected to the original root, although in a diff. lineage) konn 'know how to' + konn 'know'


    Sorry, it's a little confusing, I'm not sure I've made much sense.
     

    Forero

    Senior Member
    But curiously the common root in connaître is in the nn, not the co(n). English can, ken, know; Greek gignosco; Latin cognosco, French connaître

    The co(n) prefix in Latin and French is cognate to the Germanic prefix for past participles (e.g. gekonnt).
     

    tom_in_bahia

    Senior Member
    South Florida/Phoenix-Tucson/the Adirondacks. Native of North American English
    Interesting that before seeing this etymology path (which makes a lot of sense), I had always just assumed that however "know" got to its modern state, the vowel that used to be between /k/ and /n/ got sucked out along the way.

    Strange that Spanish, Portuguese and Italian all have a related word for "to know a person/place". However, the "co-" in Portuguese and Spanish seems to almost not exist in other words. The Portuguese word: conhecer has a palatal nasal (-nh-) which would correspond, normally, to a "gn" in Italian (and sometimes to the ñ in Spanish: banho--baño--bagno, for example). So, either the -nh- in Portuguese is coincidence or it preserved the original phonetic idea of the -gn- in cognosc* while Spanish and Italian did not.

    My French is rusty, would the double n in connaître be pronounced as a geminate? That would also show a related phonetic descent, wouldn't it?

    (*I put a star after cognosc because I don't know the infinitive in Latin.)

    My comment about Haitian Creole was purely about the sound of the word and the way "connaître", probably in one of its morphemes, like "connait" was reduced to a nasal "konnen" in Creole, which sounds like the words we were discussing for "to know how".

    And with this consonant plus nasal combo that seems to embody the idea of "know" in Indo-Euro languages, would the zna- words from Slavic languages be related...I assume so. I can see a z and a g having some path to one another through sound shifts and I vaguely remember a declension rule from Polish that a final g would go to z before adding the declension, or something to that effect.

    Is palatal nasal the correct term for ñ in Spanish/gn in Italian and French/ nh in Portuguese??? Correct me, please.
     

    Forero

    Senior Member
    Interesting that before seeing this etymology path (which makes a lot of sense), I had always just assumed that however "know" got to its modern state, the vowel that used to be between /k/ and /n/ got sucked out along the way.
    Sometimes the vowel was missing in PIE - whatever "zero grade" means.
    Strange that Spanish, Portuguese and Italian all have a related word for "to know a person/place". However, the "co-" in Portuguese and Spanish seems to almost not exist in other words.
    The prefix "co-" (Latin "com-"/"con-"/"co-" depending on next sound) is very common in all the Latinate languages.
    The Portuguese word: conhecer has a palatal nasal (-nh-) which would correspond, normally, to a "gn" in Italian (and sometimes to the ñ in Spanish: banho--baño--bagno, for example). So, either the -nh- in Portuguese is coincidence or it preserved the original phonetic idea of the -gn- in cognosc* while Spanish and Italian did not.
    I think different speakers of Latin must have had different ways to handle "co(n)-"+"gno", some keeping (or adding) "n" as part of "co(n)-" and dropping "g" (or assimilating "gn" to "n") and others using "co-" and keeping their version of "gn".
    My French is rusty, would the double n in connaître be pronounced as a geminate?
    It was (in OF), but no longer is, a geminate sound in French, but it may have an effect on the "o".
    That would also show a related phonetic descent, wouldn't it?

    (*I put a star after cognosc because I don't know the infinitive in Latin.)
    cognoscere with the same issues as the inchoate verbs in -escere (palatalized "c" right after "s" - keep "sk"?, drop "s"?, drop "c"?)
    My comment about Haitian Creole was purely about the sound of the word and the way "connaître", probably in one of its morphemes, like "connait" was reduced to a nasal "konnen" in Creole, which sounds like the words we were discussing for "to know how".
    Is the "-en" ending common in Creole?
    And with this consonant plus nasal combo that seems to embody the idea of "know" in Indo-Euro languages, would the zna- words from Slavic languages be related...I assume so. I can see a z and a g having some path to one another through sound shifts and I vaguely remember a declension rule from Polish that a final g would go to z before adding the declension, or something to that effect.
    I've read somewhere that "zn-" is the normal Slavic reflex of PIE "gn-".
    Is palatal nasal the correct term for ñ in Spanish/gn in Italian and French/ nh in Portuguese???
    Yes.
     

    modus.irrealis

    Senior Member
    English, Canada
    I completely missed out on the actual link there that Forero pointed out. Very interesting.

    So, either the -nh- in Portuguese is coincidence or it preserved the original phonetic idea of the -gn- in cognosc* while Spanish and Italian did not.
    Just to add here, the pronunciation of -gn- in Latin was in Classical times like the ngn of wrongness.
     

    tom_in_bahia

    Senior Member
    South Florida/Phoenix-Tucson/the Adirondacks. Native of North American English
    I completely missed out on the actual link there that Forero pointed out. Very interesting.


    Just to add here, the pronunciation of -gn- in Latin was in Classical times like the ngn of wrongness.
    I find that really fascinating. So it was a velar nasal? Doesn't Greek do something like that with a double g? The velar nasal is not very common in modern Romance languages, is it? I mean, I know it exists in -nc-(hard)/-nqu-/-ng- (hard) clusters in Spanish, but can't really exist in Portuguese because an n before a velar consonant would nasalize the preceding vowel and not be affected by the following consonant (Port. pancada /pã-'ka-da/ vs. Span. tengo /'teNG-go/). And, thinking about the example with 'tengo': Portuguese makes it tenho (see also vengo --> venho, pongo --> ponho)...maybe Portuguese had a trend of changing the velar nasal to a palatal nasal.

    I derailed the thread, didn't I???
     

    modus.irrealis

    Senior Member
    English, Canada
    I've heard the same about the -gn- in Classical Greek gignosco.
    And also about -gm- as well, although it seems there's much more evidence in favour of a nasal pronunciation of g in -gm- than in -gn-. But in Latin, the situation is the opposite where's there's more evidence of it in -gn- than in -gm-.

    I find that really fascinating. So it was a velar nasal? Doesn't Greek do something like that with a double g? The velar nasal is not very common in modern Romance languages, is it?
    A velar nasal, yes -- although it was rare in Latin as well (although you're right that it's gotten even rarer since), and the sound only occurred before other specific sounds and wasn't really "independent."
     

    HistofEng

    Senior Member
    USA Eng, Haitian-Creole
    Is the "-en" ending common in Creole?
    Yes, it is a common ending (nasalized vowel) in Haitian-Creole (Kreyól Ayisyen), but I don't think there's a pattern as to the type of words that have it. Nouns, verb, adjectives, etc, sort of like the "er" ending in English.

    "To know" can either be said as "konnen" or "konn" depending on the phonetic environment (and, secondarily, the preference of the speaker).
     
    < Previous | Next >
    Top