From Online Etymology Dictionary’s FB post:
The ides of March approach. What’s the past tense of “beware?”
Century Dictionary explains, “Like be gone, now begone, be ware came to be written as one word, beware, and then was classed by some authors with the numerous verbs in be-, and inflected accordingly; hence the erroneous forms bewares in Ben Jonson, and bewared in Dryden.”
But calling any usage “erroneous” consigns you to history’s dustbin. Better if Caesar had waswared the people who had told him he should have bewent.
Fave comment (from Helen Pollock): “Bejaysus! ????” (Of course there’s a commenter who has to spell out that “‘Beware’ is an imperative form. […] It wouldn’t have a past tense.” Thanks, Captain Obvious!)
Bewore.
Beenware?
There’s a kids’ alphabet song by Yusuf Islam (the former Cat Stevens) with the line:
“Ta is for Taqwa, bewaring of Allah”
Always felt a bit grammatically awkward, but how to fix it?
For at least a year my son would consistently respond to my “Be careful” with “I’m becarefuling!” (Not a witticism, just how he thought it was conjugated.)
Being an imperative form doesn’t necessarily mean that you have no associated past tense – you can say ‘Stand!’, and report that the person stood. The odd thing about beware is that it seems to have *only* an imperative form.
be warned
Kusaal, despite not going in much for defective verbs in general, has mit, which is always imperative. The great majority of instances use the verb impersonally along with a subordinate clause complement, to mean “see that it doesn’t happen that”, as in
Mit ka ya niŋi alaa.
NEG.IMP.let and 2pl do thus:NEG.
“See that you (pl) don’t do that.”
Cf (simply using the negative imperative of a main verb):
Da niŋini alaa
NEG.IMP do:2pl thus:NEG.
“Don’t (pl) do that!”
Must less often, mit is used with a 2nd person subject and a NP complement, and then it means “beware (of)”:
Miti ziri nɔdi’esidib banɛ kenni ya sa’an na la.
Beware.IMP:2pl lie prophet:PL DEM.PL:NZ come:IPFV 2pl among hither ART.
“Beware of the false prophets who come among you.”
Bewore.
I would probably have produced “bewared” if it ever came up, but I agree that “bewore” sounds better.
Dryden’s use of the supposed error, viz. “Once warned is well bewared,” certainly seems well within the bounds of poetic license.
pupaware / larvaware / eggware.
I like the “Bewore” suggestion since it lends itself easily to an obvious extension in these times of far too many “bewares” to count – we can become beworn out.
The odd thing about beware is that it seems to have *only* an imperative form.
Because it isn’t a verb,it’s a spelling mistake.
„Was ware“ is the past tense. „Is ware“, „will be ware“ etc.
Lameen, one of my professors told us that when he admonished his little son, “Behave!” the boy protested, “I’m being have!” (rhymes with “brave”).
So, you think you and other readers think that that commenter’s statement is true, and obviously true?
“Beware” being defective makes it odd, but, to me, what makes it even odder is that the set of parts it has is determined by the [i]form[/i] of parts of “to be”: where the latter is “be”, we have “beware”. So “beware” has a present infinitive, as in “I told them to beware of the dog” and “They should beware of the dog”.
Please disregard my earlier comment: it got posted in corrupt form, and the web page did not give me the expected 15-minute grace period in which to make corrections.
— Of course there’s a commenter who has to spell out that “‘Beware’ is an imperative form. […] It wouldn’t have a past tense.” Thanks, Captain Obvious!
So, you think you and other readers think that that commenter’s statement is true, and obviously true?
“Beware” being defective makes it odd, but, to me, what makes it even odder is that the set of parts it has is determined by the form of parts of “to be”: where the latter is “be”, we have “beware”. So “beware” has a present infinitive, as in “I told them to beware of the dog” and “They should beware of the dog”.
Rosie,
As a native speaker I find any attempt to treat “beware” as an actual verb wrong. It is not defective, it is simply “to be” combined with the adjective “ware”. If someone says “hey, beware!” My automatic response is “Yes, I am ware”, not ” Yes I am bewaring”. The same with “begone!” – “Dude, I am so gone”. The confusion arises because “ware” is an archaic form of “aware” so “I am ware” does sound a bit odd.
The Welsh verb “to know”, gwybod, has a very irregular mainly-literary stative synthetic present gwn gwyddost gŵyr etc, but the majority of its forms are made up of the gwy- component followed by the relevant part of the verb bod “to be”, thus etymologically “be aware.”
https://en.wikipedia.org/wiki/Literary_Welsh_morphology#Bod_and_compounds
@ Vanya
So you would parse Rosie’s examples as e.g., “I told them to BE WARE of the dog”. I am not able to do this unconsciously, not only because, as you say, “ware” in that sense no longer exists (for me BE WARE would appear on the walls of the dressing room at a slave market????), but also because for me beware is inseparable in the way of fossilised expressions like o’clock.
@Robin:
Just so.
Caesar really should have beworn the Ides of March. (There’s just no telling some people.)
“So you would parse Rosie’s examples as e.g., “I told them to BE WARE of the dog”.
Yes. It is interesting. My very first unconscious response to the question “what is the past tense” was also, like David Eddyshaw’s, “bewore”, which also immediately sounds wrong. But thinking of it as an imperative, I realized my immediate unconscious response to a command “beware”! would be to parse it as verb+adjective.
@de
In Irish you have also a split between knowing, having “book/training” knowledge of and being acquainted with. So
Tá (a) fhios agam sin = I know that
Tà eolas agam ar/faoi sin = I know about that (from book/training)
Tá aithne agam air = I am acquainted with him/it (masc.). You can see the construction is SUBJ + TO HAVE + VERBAL NOUN + OBJ. The alternate construction SUBJ + VERB +OBJ is available for aithne but
Aithním é = I recognise him
I think this construction is available for fios with the synthetic verb form fionnaim but for the more usual (and historic) feadar the construction is only available in the negative, e.g.,
Ní fheadar = I have no clue
Of ESL dictionaries, Macmillan says “USUALLY IN IMPERATIVE” and Cambridge says “only in infinitive and imperative”. The infinitive form is the same as the imperative form, but is the infinitive the same as the imperative?
OED1 “The origin of this is involved” with interplay of OE v. bewarian & adj. wær & v. warian; “After 1600, the verbal aspect so far prevailed that the inflexions bewares, bewared, bewaring, were used by good writers; but these have again been discarded, and beware is now used only where be ware would be a possible construction, viz. in the imper. (chiefly), the infin., and pres. subj. (rarely).”
Tá (a) fhios agam sin = I know that
That brings back memories of my Modern Irish class (with Mícheál Ó Siadhail at the Dublin Institute for Advanced Studies in 1975), where we were drilled in the exchange (using Connemara pronunciation) /ˈwilisad/? ‘You know?’ /ˈtɔsam/ ‘I know.’
So, you think you and other readers think that that commenter’s statement is true, and obviously true?
Sorry if I gave the impression I thought the commenter was correct in every respect; I just meant he was explaining an obvious joke, even if his explanation was a bit off.
“Beworn”.
I think it actually works: “He was beworn, but he went ahead and did it anyway.”
“The learner will readily perceive what an awkward thought would be produced in a listener by hearing a person say, I had bewared, thou hadst bewared, he had bewared; or I had quothed, thou hadst quothed, he had quothed. Neither of which expressions is allowable in grammar.” Oliver Beale Peirce’s 1839 “The Grammar of the English Language,” giving two examples of what he calls a “defective asserter.”
(Much like a late 20th century academic linguist, Peirce decided that the traditional categories and labels of the parts of speech were unsuitable and made up his own system, in which the parts of speech, for English at least, were “Names, Substitutes, Asserters, Adnames, Modifiers, Relatives, Connectives, Interrogatives, Repliers, and Exclamations.”)
So I was completely misled by bewahren “keep ~ preserve ~ conserve”.
Defective “quoths” seem to be quite common cross-linguistically: Latin has inquit, Middle Welsh has heb(yr) and med [mɛð], Old Irish has ol …
Indeed, “quoths” aren’t always even verbs; Guldemann has a nice study of the variation in what he calls “quotative indexes” (https://books.google.co.uk/books/about/Quotative_Indexes_in_African_Languages.html?id=XEy8hYOYGs4C)
Sounds like an interesting book!
Russian has мол, де, дескать, and maybe others I’m forgetting.
Even within Germanic languages, you can call defective “Beware!” (w/o explicit object) an imperative verb if you want, but it’s doing exactly the same sort of thing as German “Achtung!,” which is not any sort of verb and, in that “imperative” context, is a rather funny sort of noun. An English example of a noun used in that “imperative” fashion would be “Attention!” (in eye dialect: atten-SHUN!) as used by a drill sergeant.
Sounds like an interesting book!
True; alas, not £122 worth of interesting.
Kusaal uses ye, which looks like the root of yɛl “say” (which drops its -l in flexion), but I think that is an illusion: it’s probably just the clause-linking particle ye “that”, with ellipsis of the preceding verb.
Comparative evidence suggests that the tempting idea that these two etyma are themselves related is false, despite its obvious plausiblity*: ye “that” seems pretty clearly to come from *ɲɪ, but the initial of yɛl does not go back to *ɲ, which is preserved unchanged in some of the other Western Oti-Volta languages, and should have caused the vowel of yɛl to be nasalised in Kusaal, which it’s not. [The e/ɛ difference is a quirk of the 2018 orthographic reform, and doesn’t bear on the question or relatedness.]
*Lots of languages thereabouts do have a clause-linker “that” which is derived historically from a verb “say”, not least the English-lexifier creoles.
it’s doing exactly the same sort of thing as German “Achtung!”
Not exactly: it can take an object, which a noun can’t, and as rosie pointed out, it has an infinitive too (which is more than you can say for “quoth”, even though that one goes back to a perfectly cromulent verb historically.) Moreover, even without an object, you can have
They said to him “Beware!”, but he did not.
You cannot, however, say
*They said to him “Attention!”, but he did not.
German “Achtung!,” which is not any sort of verb and, in that “imperative” context, is a rather funny sort of noun.
Not so much. It is a modern alternative to gib Acht! The French have attention! = fait attention!.
Surely I must not point out that gib Acht! does not mean “high eight!”. That would be gib’ mir acht!, an ordinal cousin of gib’ mir fünf = “high five!”.
an ordinal cousin
Surely you mean “an octopodal cousin.”
You cannot, however, say
*They said to him “Attention!”, but he did not.
You can say
# They said to him “Attention!”, but he did not attend to what they said. #
Surely you mean “an octopodal cousin.”
Maybe we can agree on “octodigital”.
Thumbs are in opposition to (the other) fingers. Are they opposed to being called “digits” ?
https://www.youtube.com/watch?v=mFSRCG4DrmI
Stretching the limits of relevance far beyond breaking point. But it’s Seun Kuti! And Cheick Tidiane! Doing a Fela Kuti song! On behalf of the Mapuche!
(And it’s got “opposite” in the title. So that’s fine.)
That you can restate a one-word utterance as a longer sentence that is more syntactically normal doesn’t necessarily mean that the “surface structure” you see/hear is mystically derived from the “deep structure” you don’t via some sort of process even Chomskyans may no longer believe in. Some online sources classify that sort of use of “Achtung!” (or, for example, the AmEng “Heads up!”) as “interjections.” Which may be a sloppy category of convenience, or may be a way of indicating that these are idiomatic expressions used in an imperative sort of way where a more complicated analysis of a longer sentence they might be considered a clipped remnant of is not actually useful.
“ware” is an archaic form of “aware”
Or wary.
Such ruminations can provide cribs or Faustregeln useful in learning (to understand) a language. If one is not interested in that, what’s the effing point ?
No, “Beware!” is a verb, specifically, because it can be the antecedent of do used as a proform (CGEL p1524.) “Attention!” can’t. Only verbs can be antecedents of “do” used in this way.
This is nothing to do with Chomskyan deep structures, or any other kind. It’s all on the surface.
Acceptable: They said to him “Beware!”, but he did not.
Not acceptable: They said to him “Attention!”, but he did not.
Verbs which lack some normal flexional forms are not rare cross-linguistically. There’s nothing far-fetched about analysing “beware” in this way, especially as (as rosie points out) it’s also got a perfectly good “infinitive”: “We shouted at him to beware.” (But not *”We shouted at him to attention.”)
Fowler, in addition to the infinitive and imperative, also accepts Unless they beware, which he calls subjunctive. I’m not sure about that description — Unless he bewares sounds just about OK to me, but forty-odd comments in I’ve lost whatever weak intuition I might have had about these forms.
(ETA: bewares is worth 13 points in Scrabble, which seems to settle it.)
In Victorian English, beware takes a direct object (“beware the Jabberwock”). In modern AmE it uses an of-complement (“beware of the Blob”). Is the direct object still used anywhere?
Stack Exchange discussion.
Which is an interesting extension given the adjectival origin — you can’t “be careful the Jabberwock”. The OED has some notes on the origin of the transitive construction, among other things:
@dm,TR
There is a verb biwaren in Middle English which has to do with goods or money, but it means to spend, not to save (sparen is used in contrast to biwaren). The etymology is via an Old Norse verja meaning to wrap/enclose or dress but also to spend money. The ON word is apparently related to the German Wesen, which would then be (etymologically) the external aspect or dress, rather than something intrinsic.
I would suppose that this sense of waren/bewaren might have displaced a homonymous verb with a sense “to defend”.
Fowler’s Unless they beware falls neatly within the rule given in the OED for more recent use, viz that beware is possible iff be is. Unless they be is grammatical in just the same hifalutin register as Unless they beware.
Interesting about “beware” having got reirregularised at some time since 1600. Language is weird.
The inflected beware must have died off by the time of Johnson’s Dictionary, since he gives the same rule: “It is observable, that it is only used in such forms of speech as admit the word be : thus we say, he may beware, let him beware, he will beware ; but not, he did beware, or he has been ware.” (It wasn’t quite 100% dead: OED has an example from Emerson.)
What CGEL is saying (if I understand correctly) is that even though you can’t say “he did beware”, you *can* say “he did”, as long as “did” replaces the entire verb.
I made up the specific examples with do-as-proform and “beware”; CGEL is not to blame for them. I referenced CGEL only in case people were unfamiliar with the idea of pro-verbs. (The issue is not quite straightforward: do as pro-verb may actually only be an unequivocal thing for British English, with the US English uses potentially explicable just as special cases of auxiliary “do.” But I don’t think that affects the contrast between “beware” and “attention.”)
The examples thus reflect my own acceptability judgments; I would be interested if other L1 speakers disagree, especially if they find my acceptable sentence unacceptable. (It would be especially neat if Americans found it unacceptable, and UK speakers thought it was OK. But don’t let me put my thumb on the scales …)
CGEL only references “beware” as such twice, once as a verb (sic) taking “of” + NP as a complement, and once in a fairly technical discussion of the grammatical status of “infinitive” to, where it calls “beware” a “highly anomalous verb.”
And ait for good measure.
Ah, sich wehren “to defend oneself”; wehr(e) dich “defend yourself”, “fight back”; Bundeswehr “Germany’s military”.
Also un-reflexive, but oddly with a dative, in 18th-century literary German: wehret den Knaben “keepeth the boys under control”. That should be the same dative as the one the OED mentions (but fails to illustrate).
But how did it escape umlaut?
Or has Wehr “weir” confused things on the German side?
Beware signs (+often cute dogs, for those that like dogs)
https://www.boredpanda.com/cute-guard-dogs/
30 total
Beware of dog: 14 (includes “of attack dog”)
Beware of the dog: 9
Cuidado con el perro: 1
Attention au chien: 1
Ovdje čuvam ja: 1
[Thai/Laotian (?)]: 1
[No beware sign visible]: 3
I note that there are no “Beware the dog” signs.
Supposedly Germans on ships say “Wahrschau!” in contexts where landlubbers say “Achtung!”, but I don’t think that “wahr” has anything to do with the various “wehr”‘s under discussion.
猛犬注意：
https://livedoor.blogimg.jp/laba_q/imgs/0/8/081ffa08.jpg
David E.: your “acceptable” sentence grates a bit against my (AmEng) native speaker ear. The subtly-different “They told him to beware, and he did” seems a bit better, however. OTOH, that seems to lead inexorablly to “They told him to beware, and he bewared,” innit? Is there another instance in which the “pro-verb” use of “do” can’t be turned back into the verb for which do is supposedly pro-ing? You can sidestep the problem with something like “They told him to beware and he complied/obeyed,” but I don’t think the pro-verb “did” covers that sort of side-stepping.
On the third or perhaps fourth hand, “They ‘bewared’ him, and he did,” sounds perfectly cromulent. Verbing weirds language, they say, and I think maybe you can even verb a verb, with “‘beware” (maybe “beware-prime” or something?) meaning “to say ‘beware’ to.” “She come-hithered him, and he did,” would be a similar construction, innit?
That this beware-prime verb is *not* defective but can be fully conjugated, even unto participles and gerunds and whatnot, is … well I don’t know if it should be surprising or not. I think probably not.
“They ‘bewared’ him
Well, if we’re bewareding now, all bets are off. But me no buts.
that seems to lead inexorably to “They told him to beware, and he bewared,”
I don’t think so: after all “The sheep stood on the burning deck, thereby proving that they were very stupid” doesn’t imply that there has to be a plural “sheeps.” A proform can render explicitly distinctions which were only implicit in its antecedent.
“We put it down there, or at least I think we do.”
“We put it down there, or at least I think we did.”
I don’t see a Transatlantic* difference, since I think all my acceptability judgements are the same as David Eddyshaw’s.
* My phone’s autocomplete suggested both capitalized “Transatlantic” and downcase “transatlantic,” just as I was wondering whether to capitalize a proper adjective that does not begin with the “proper” part.
“Sheep” does not lack a plural form (it’s a count noun, not a mass noun); it just happens to have an irregular plural form identical to the singular form. But the usual claim for “defective” beware is that it lacks a past tense form altogether, not that it’s a weird irregular verb like “put” where the past-tense form happens to be the same as the bare infinitive (and the past participle ditto).
But on what may be the fifth or sixth hand, try this fill-in-the-blank exercise:
You said they told you to beware?
Yes.
And did you?
Yes I did.
Sorry, what was it you did?
I ________, just like they told me to.
I did beware, just like they told me to.
(The plot thickens. Though I think this still complies with the OED rule.)
But I don’t think this is relevant to your original thesis, which was that “Beware!” is doing exactly the same sort of thing as German “Achtung!”
“Beware” is without dispute a very peculiar verb, but it still has verblike properties that “Attention!” completely lacks. You could declare that it’s not a verb, on account of lacking any inflected forms; nobody can stop you. But if you do, it will be at the price of gratuitously complicating your description of English grammar, because of all the places elsewhere where you will need to amend “verb” to “a verb, or the unique-category word ‘beware.'”
My original thesis might be that the exclamation/command “Beware!” may be doing something different than the verb in “they told him/you to beware,” just as the exclamation/command “Attention!” may be doing something different than the noun in “He ordered them to stand at attention.”
Admittedly a more complex exclamation/command like “Beware the Jabberwock!” may make things more complicated, but going back to German, you can say “Vorsicht!” (a noun, at least in other contexts) just like you say “Achtung!” and you can expand it with a complement as in e.g. “Vorsicht vor dem Hund!” (=”Beware of the dog!”).
There’s an analogy with the Kusaal mit “see that it doesn’t happen that”/”beware of.”
Morphologically, mit is absolutely unique, and clearly doesn’t fall into either of the Kusaal conjugations. Moreover, unlike every other verb in the language (to beg the question) it can’t take any preverbal tense/aspect/mood or polarity particles at all. Kusaal doesn’t have anything like proform-do, so no test along those lines is any help. It is also the only verb (if it is a verb) which can be used impersonally in the imperative (if that is indeed what is going on.)
So why call it a “verb” at all?
Well, Kusaal does have a handful of undoubted verbs with fixed mood or polarity, notably zi’ “not know”, which can only be indicative; other moods simply use mi’ “know” with the appropriate negative particles. So there is a precedent within the language itself.
More to the point, mit always takes a following NP complement or a clause complement introduced by ka “and/that.” The only other words that can take following NP complements in Kusaal are verbs or prepositions, and the only other words that can take (non-nominalised) clauses as complements are all unequivocal verbs.
Furthermore, mit can be followed by the enclitic 2pl subject pronoun y(a)/ni; all other words without exception that can appear in this context are unequivocal verbs. The long and short of it is that it’s a whole lot simpler to categorise mit as a verb and explain in the appropriate place about its unique perversities than to regard it as the sole representative of its own part of speech, which will then need to get an honorary mention along with verbs every time you talk about objects or purpose clauses.
Whether this is “correct” in some ultimate transcendental-linguistics sense is a question I find hard to take much interest in, on account of not believing that there is such a thing anyway.
With regard to “Beware!” versus “Attention!”; one of these things is not like the others at all; one of the remaining three is not altogether like the other two:
Sing!
Beware!
Attention!
Jump!
Supposedly Germans on ships say “Wahrschau!” in contexts where landlubbers say “Achtung!”,
That’s because it can be said faster than Warszawa. In a storm, even one unnecessary syllable can mean the difference between life and the other thing.
The WiPe claims, on the other hand:
# Wahrschau
1. Warnruf: „Achtung!“, „Vorsicht!“ (aus dem Niederdeutschen: Warnung,[13] aus mittelniederdeutsch warschuwinge = Warnung,[14] vgl. niederländisch waarschuwen = warnen). Davon abgeleitet wahrschauen:[15] warnen, instruieren, benachrichtigen #
And here is a Warschau am Mittelrhein.
Typo correction: that’s a Wahrschau am Mittelrhein.
Ben Jonson’s “bewares” was noted by Noah Webster, in 1828:
Webster stops short of calling Jonson “erroneous”; only the Century Dictionary dares to go that far (and also to declare that it’s “properly” written as two words!). Jonson’s use might be translationese: he’s translating Horace’s Art of Poetry
as
cavet is an ordinary present indicative verb in Latin, so there’s the temptation to use a present indicative in English.
Beware my Love John Bonham Version
https://www.youtube.com/watch?v=nIRyQs88Wpo
@de
Maybe I am missing something, but look and watch are completely normal verbs. Yet there is a rule against past in the context of the bare command “look/watch out!” the same as for beware, i.e. “I did watch out”, but not “I watched out”. Of course “I watched out” is perfectly OK in other contexts (even commands which are not the bare two words), i.e., “I watched out for a red Volvo”. I would explain this as that certain commands ( i.e., beware and watch out, but not climb down) are treated grammatically like infinitives with a missing preamble as in (you have to) beware/watch out.
Compare can/could, where one automatically suppletes the missing form, e.g. “I can’t do that yet, but I’ll be able to eventually.” None of the synonyms or paraphrases for “beware” is preeminent enough for analogous automaticity.
@Plastic:
Good find: I was casting around in vain for another English verb which doesn’t do inflected forms, and couldn’t think of any; but I hadn’t thought of phrasal verbs. I think “watch out” does indeed meet the JWB challenge “is there another instance in which the ‘pro-verb’ use of ‘do’ can’t be turned back into the verb for which do is supposedly pro-ing?”
“They told him to watch out, and he did.” but not “He watched out.”
However, as you say, it is not as thoroughly defective as “beware”:
Sing!
Beware!
Attention!
Jump!
Watch out!
Answers:
No, I won’t sing!
No, I won’t beware!
*No, I won’t attention!
No, I won’t jump!
No, I won’t watch out!
I’m already singing!
*I’m already bewaring!
*I’m aleady attentioning!
I’m already jumping!
I’m already watching out!
I see what you are driving at with “infinitives with missing preambles”, but it seems to me that you could actually analyse any command with an imperative verb that way in English (the “infinitive” and “imperative” forms are identical with absolutely all verbs), so analysing only the exceptional cases like “beware” in that fashion doesn’t actually add anything to the simple statement that these verbs behave in the odd way that they do: it’s just a more roundabout way of saying the same thing.
In the case of “beware”, much the simplest statement is just that it lacks all inflected forms.* The “infinitive” and “imperative” are not really “different” at all, but different uses of the same form: in the ideology of CGEL the supposed formal distinction is just a hangover from the traditional grammar which tried to fit English into Latin categories.
*As JWB rightly points out, this is not the same as possessing the forms but not distinguishing them in the surface shape, like “put/put” or “sheep/sheep”: the irregularity is not the indistinguishability of some forms, but their outright absence. My examples with “put” and “sheep” didn’t answer his question, as he quite reasonably complained. The odd behaviour of “beware” is not that it conflates forms but that it lacks some; however, because this behaviour is a formal rather than semantic irregularity, in all cases where “beware” is not actually made to display inflected forms, it can be treated just as if it didn’t conceal a dark secret and it can go drinking with its verb buddies and proform hangers-on on a Saturday night and they never suspect a thing.
Oops. Needs revision. You also can’t say: “I beware”, “we beware” etc, except in Fowler’s subjunctive sense. So it’s not enought to say that “beware” is irregular solely in that it only has the single form “beware”, some uses of that form are also verboten.
The accurate formulation is that it lacks all forms which are ever inflected in any English verb: as “be” marks more distinctions than any other verb, you end up with the OED rule.
However, just because there is (one) English verb which makes enough formal distinctions, you can still describe the behaviour of “beware” as a pure formal irregularity rather than bringing in semantics: the price is accepting that all English verbs other than “be” have formally conflated (say) 1st and pl forms, rather than that English verbs don’t have that distinction in the first place. But I think you probably want to do that anyway.
What I’m saying here, is, that OED, it’s pretty good. Shows promise.
If you look for natural examples of “watched out,” many of them are not the phrasal verb, but in some other construction like “watched out of her window.” But I don’t see anything ungrammatical or odd-sounding about, e.g., “The young hunters watched out for the reappearance of Jed Sanborn, Snap and Shep going to Firefly Lake for that purpose.” And that seems like the phrasal verb – the young hunters could well have been warned or commanded “Watch out for Jed Sanborn!” (The example is from Ralph Bonehill’s “Out with Gun and Camera: or the Boy Hunters in the Mountains,” published 1910; you can also easily find examples of “watching out” for the same phrasal-verb construction) I guess PlasticPaddy’s claim is that you can say “watched out for X” but not bare “watched out”, which seems like an odd place for the line to be drawn. If anything, that suggests that the bare command “Watch out!” is actually somehow a different usage than a command like “Watch out for that tree!” rather than the two examples being variants of the same thing.
Maybe I don’t have the right corpus-searching shortcuts to look for a counterexample in the limited time I have while finishing my coffee this morning. “I didn’t say I was watching out four hours, no, I said I was watching out on an average on and off” seems close to a “bare” usage of phrasal “watch out” in an inflected form, but in the context of the discourse it refers to watching out of a window.
OED rule … that OED, it’s pretty good
Credit where credit’s due, Samuel Johnson gave the same rule, and rosie also came up with it without looking in any references.
Inflected forms of beware got into print a few times in the 20th century, and still do today. Some can be attributed to translationese or non-native English—a lot of the hits for “bewares” in Google News come from India and Pakistan, so maybe it’s accepted in Indian English. But others look like regularization within English. A few examples:
These are rare, and maybe most copy-editors would have struck them out, but they’re more evidence that beware acts like a verb.
The interesting thing is how “beware” went through an earlier phase of being regularised and then became anomalous again.
I wonder if this could be accounted for by the word becoming rarer and more associated with archaising and/or high registers, so that people were mostly acquiring it in stereotyped contexts, with the result that the process of turning it into a well-behaved verb had to be restarted from scratch, as it were.
Interestingly, of those three twentieth-century examples, I would have unhesitatingly rejected the first two, but would have happily passed over the third without noticing anything odd if it had read “the wise trader bewares Greeks bearing gifts.”
(I also find de la Mare’s example definitely less odd than Emily Murphy’s.)
I wonder if this could be accounted for by the word becoming rarer and more associated with archaising and/or high registers, so that people were mostly acquiring it in stereotyped contexts, with the result that the process of turning it into a well-behaved verb had to be restarted from scratch, as it were.
This is certainly true for me as an English L2 speaker. I’d previously encountered it only in the stereotyped context of the exhortation “beware of”, which I mentally filed as a fixed literary expression.
Before reading this discussion I had assumed that it was the imperative of an otherwise obsolete verb “to beware”.
@ktschwarz: I find the use of a finite form much worst in the Emily Murphy quote than in the two others. I think Edwin Lefevre’s passes muster because it is directly adapting a well-known expression. On the other hand, I don’t know whiy Walter de la Mare’s phrasing also works for me; somehow, it just does. Of course, I am known to be unusually partial to de la Mare’s wordsmithing; he is my all-time favorite poet. So it may just be a matter of my personal, idiosyncratic appreciation for this writing style. I do think, however, that de la Mare must have used the peculiar finite form “bewaring” intentionally; he could not possibly have been unaware that the word had an unusual character.
@David Eddyshaw: When I see evidence for a process of linguistic regularization, followed by regression to the older, less regular form, my first inclination is to suspect that the regularization was actually only occurring in certain dialects. The reappearance of the irregular form would not, in this scenario, be an atavism. Rather, the irregular form probably persisted without modification in some regional, ethnic, or socioeconomic dialects, and those dialects eventually won out over the regularizing ones.
I find the use of a finite form much worst in the Emily Murphy quote than in the two others. I think Edwin Lefevre’s passes muster because it is directly adapting a well-known expression. On the other hand, I don’t know whiy Walter de la Mare’s phrasing also works for me; somehow, it just does.
I agree that quot 3 > quot 2 >> quot 1 in terms of acceptability; if anything, at first glance the Murphy quote sounded like a malapropism to me, though now that I’ve looked at it again I think I can see the intended meaning.
(It’s probably the passive that does it – offhand I can’t think of any forms of a theoretical *to be bewared that would be in any way acceptable, though IIRC some had been found earlier in this thread.)
“Ta is for Taqwa, bewaring of Allah”
Wuddaya expect from the guy who sings the line God’s recreation of a new day using DRESS in “Morning Has Broken”, when it’s obviously God’s re-creation using FLEECE.
For me, ?They told me to beware, and I did beware is a lot more cromulent than *They told me to be wary, and I did be, so beware can appear where be cannot.
As I could could never live with myself if I failed to correct my error and led Hatters astray over Kusaal grammar, I am impelled to say that my statement above that mit “beware” is used impersonally in constructions like
Mit ka ya niŋi alaa.
Beware and 2pl do thus:NEG.
“See that you (pl) don’t do that.”
is incorrect. The verb acquires a preceding subject pronoun in exactly the same circumstance as direct commands do, such as reported speech:
O daa yɛli ba ye, Ba mit ka ba niŋi alaa.
3AN TENSE say 3PL that, 3PL beware and 3PL do thus:NEG.
“She told them to see that they didn’t do that.”
I don’t know why mit doesn’t take a postposed 2pl subject in this usage like other commands addressed to several people, but it can’t be because it’s impersonal. Perhaps its a “generalised ‘you'”, as in proverbs like Bʋŋ ya’a bɔɔd ye o lubif, fʋ pʋ nyɛti o tʋbaa. “When a donkey wants to throw you off, you can’t find his ears.”
Apologies for any confusion. Carry on.
“When a donkey wants to throw you off, you can’t find his ears.”
That is wonderful. Much more vivid and outgoing than the self-improvement maxim “determination is the key to success”. All this 24/7 incurvatio in se gets on my tits.
Here’s a related German saying in modo negativo, unfortunately still of intent hortative: einmal gebockt ist nicht gelammt. It’s one of my favorites, but most people don’t know it.