User talk:AtterCleanser44

Hi! Was wondering, if you had to give pairs/a.s.o. like take/nim, call/cleep, worth/become, ken/kithe/make known, wit/know, and say/queath differences in connotation, what would you give them? TimeMaster (talk) 23:26, 13 March 2021 (UTC)


 * Interesting question. I'll go through each pair one by one.


 * Take and nim - Nim was the main word for take in Old English and so was pretty broad in its uses. Take is simply the Norse word for it and later came to replace it. If I had to include both words, I'd have nim be the more informal equivalent of take.
 * Call and cleep - Cleep was the native word for call, and call is from Norse. If I had to include both words, I'd say that cleep would be a more formal or literary word for call and would be limited to mean name. It's like how yclept is a literary word for called (as in named).
 * Worth and become - If worth had survived alongside become, I'd rather that worth were the normal word for become. Become wouldn't be uncommon, but might be a bit more formal to use.
 * Ken and kithe - pretty arbitrary, since both words aren't used in standard speech anymore. If it were up to me, I'd use kithe as the normal word for make known. Ken would be a more formal equivalent.
 * Wit and know - Know would be the normal word. Wit would be a literary equivalent for it.
 * Say and queath - the latter would be a more formal and literary equivalent of say. It's like how quoth (historically the past tense of queath) is used as a literary synonym for said, and bequeath is already a somewhat formal word. --AtterCleanser44 (talk) 02:08, 14 March 2021 (UTC)

Interesting. I'd been thinking about trying to carve out a niche for them that a French word currently holds in connotation. Maybe "seize" for nim, "invoke" for cleep, "declare" for "queath", and the like. But since they were so general in Old/Middle English, it's very hard to say what they might have survived as if they had done so at all. TimeMaster (talk) 05:40, 14 March 2021 (UTC)

Do you disagree with trying something like the above? I'm generally not a big fan of formality/literary distinctions myself. Btw, when you write "mot", are you saying it should have irregularly become a short vowel, notwithstanding the original long "moot" (< mótan)? TimeMaster (talk) 19:53, 17 March 2021 (UTC)

Also, what are your thoughts on putting back old infinitive forms for modals like "cun", "shull", "mow" (may), and "dur" (dare)? Will and moot seem to share their forms with their present tense. And your thoughts on putting back "thurve" (present "thar", past "thurft"/"thurst"), and some older preterite present verbs like dow (deigh, dought, meaning "serve to" or "be fit to"), unn (ann, outh, meaning grant/bestow/wish/regard with favor/ordain/permit/allow), and mun (man, munt, meaning "remember") (and maybe now/neigh/nought, but that's confusing and "is enough" serves the purpose just fine)? Although I think these may have naturally regularized. Still, particularly "dow" seems like it could be a nice modal. TimeMaster (talk) 20:16, 17 March 2021 (UTC)


 * It's certainly not a bad idea to give some words specialized meaning. For example, beseech had a few different meanings in Middle English, but now, it's only used to mean implore, so beseech is a pretty useful word to replace the likes of implore. I think it'd be useful to give cleep the specialized meaning of invoke and queath (which I personally would spell queathe to show the voiced consonant) the meaning of state.


 * As for the modals, I based the mot variant on the entry in the Anglish wordbook. It seems that in late Middle English, a variant in which the vowel was shortened began to appear. Admittedly, I'm not sure whether it would've become the common form if it had stayed in actual use, so I guess it's better to go with the expected variant and change it to moot.


 * I think that the infinitives would have likeliest conformed to the first- and third-person singular present. There would now be no difference between the infinitive and the present tense for will, owe, and moot, so analogy could have played a role and caused speakers to discard the formally distinct infinitives of the other modal verbs. In fact, there are a few rare Early Modern English examples of can being used as an infinitive, e.g., Francis Bacon wrote, "In evil, the best condition is not to will; the second, not to can".


 * As for the obsolete preterite-present verbs:


 * Thurve - I wouldn't mind bringing this back. From what I can tell of the ME forms, the present tense would be thar, and the past tense thurt. It seems that thar can be used to show necessity or obligation, and in the negative, it shows a lack thereof, so it's a good word to use alongside moot. In other words, I moot not means I must not, and I thar not means I need not (ironically enough, it's the other way around in German).
 * Dow - a useful verb for saying that something is worthwhile. It seems that the form from the infinitive has been extended, so one would say I dow. In other words, it's a regular verb, except that the past tense is dought.
 * Un - this seems like a good word to use for grant. I'd use this as a regular verb.
 * Mun - useful as a regular verb meaning remember.
 * Now - not needed, since we can make do with be enough. --AtterCleanser44 (talk) 23:14, 17 March 2021 (UTC)

You're definitely right about the analogy with will/moot/owe. And yes, I had been reading more and it seems "thurt" ended up mostly winning out over "thurst" and other forms. As for the infinitives, I guess that makes sense. We'd end up with sentences like "I like canning walk!", "Canning get food is needed by all life", and "I want to can play the piano". Also, do you think reviving some old meanings for "can", as in "be familiar with", would be a good or bad idea? I suppose "know" going from "recognize" to a wider meaning isn't necessarily bad, but still.

It is rather irksome how moot (after it shifted meaning), thar, and owe all have such similar meanings. I was thinking about trying to make some connotation sheds. The old "be allowed to" isn't that useful, imo, and if a speaker REALLY wants that meaning, they could use "be let to". Maybe moot would be for deductions, owe for moral obligations/debts/and so on, and thar for requirements? I don't know.

I do think "un" would be spelled "unn" to follow the three letter rule (hence inn, eye, a.s.o.). It seems unn and mun and dow and wit were used as non-auxiliaries (if auxiliaries at all for some) in OE too, but unn had some secondary meanings aside from the "grant" thing that I think had to do with being used in phrases with other verbs (but I'm not totally sure). TimeMaster (talk) 00:01, 18 March 2021 (UTC)


 * From what I can tell, can as a non-auxiliary simply meant know. For example, I can French meant I know French, and he can much about magic means he knows much about magic. If we bring it back, perhaps we can separate can from know by having can be more related to ability; hence, one can say I can German, but not I can this feeling (so know is broader, and can has a more objective tone).


 * I don't think we need to limit moot for inferences only. It wouldn't be a problem to have a few different ways to show obligation, since we're all right with using both must and have to. I'd also say that the main worth of thar is that it's a handy way of showing lack of obligation, so thar not and not have to mean the same thing. Maybe the verbs would have been differentiated over time in terms of register, so moot and thar are likelier to be used in rules and laws than have to. As for owe, perhaps the difference is that it is more related to moral duty, so it doesn't sound as authoritative as moot and thar.


 * I'm not sure that the verb would now be spelled unn. Though the three-letter rule is a thing, it doesn't seem natural to have double ns at the end of a word, inn being an exception. And there are verbs that are spelled with only two letters such as be, do, and go, so I don't think people would have felt strongly compelled to spell the verb with nn. --AtterCleanser44 (talk) 02:35, 18 March 2021 (UTC)

Yes, the "be familiar with" know. I think "know" took up that meaning (earlier being just "recognize") after "can" started being only a helper. So "I can German" would be like "I'm familiar with German", in other words "I know [how to speak] German". It's true it wouldn't be an issue to keep moot in its current meaning. German shifted things around a bit with dürfen and müssen (which kind of swapped meanings) in similar ways to English. I guess it's just frustrating having them merge (except in the negative) for no apparent reason. Well, the exception to the three letter words are for grammatical/extremely common words like be/do/go. I don't think unn would qualify, so it would join "inn" in being an exception. The only other possible exceptions would be ann (possible present tense of unn to disambiguate with an), enn, and onn. But enn/onn/ann aren't (non-grammatical) words. TimeMaster (talk) 03:22, 18 March 2021 (UTC)

Maybe "deduction" wasn't the right word. Just the general epistemic (very) "likely to" senses of it would be predominant, and thar would take over the "have to" sense. It also just occurred to me that these verbs could have their past tenses used in the present, like we do today with could and might. Any thoughts on that as it applies to these verbs? And also on any participle forms of these helpers? Kind of like how we say a "mustread book" (although this is in the deontic sense I dislike), maybe we could get other helpers as prefixes like that. But I'm not sure the best way to disambiguate passive vs active. Maybe a past participle in either the helper or the non-helper. TimeMaster (talk) 05:52, 18 March 2021 (UTC)


 * I suppose that you can have the main sense of moot be epistemic if that's what you want. It's understandable why you'd want thar to be the main word to show obligation. As for me, I'm not too bothered by how we have a few different modal verbs for obligation. At this point, how the revived modal verbs ought to be differentiated is up to one's personal taste.


 * As for the past tense, yes, we often use the past subjunctives of the modal verbs in contexts referring to the present. One reason we do this is that it sounds softer. For example, would you pass me the salt? sounds softer and thus politer than will you pass me the salt?. This isn't a thing only in English, since the same thing is done with the modals in German. Now that I've thought about it some more, if we bring back moot, owe, and thar, then we can differentiate the use of their past subjunctives thus:


 * When we talk about the present, owe can be used alongside the past subjunctive ought, which sounds softer than owe and thus means should in meaning. So we owe to leave has a stronger sense of moral obligation than we ought to leave (which sounds more like a recommendation).
 * Besides being the past tense of moot, must can be limited to being used in hypothetical situations. So if must isn't used to refer to the past, it refers to a hypothetical situation, e.g., I must work if I needed money means I would have to work.
 * Thar is like moot, so if thurt is used to refer to the present, then it's a past subjunctive referring to a hypothetical situation, e.g., in even if I were given more work, I thurt not worry, the main clause means I would not have to worry.


 * The past participles of the modal verbs would have later been based on the past tense forms. For example, the past participle of can would have become could (in fact, there are a few rare Early Modern English examples of past participle could). However, I don't think that people would think to use the past participles for adjectives. Phrases such as must-read and must-see are not part of a productive process; instead, they're simply based on the expressions themselves and come from how people will recommend something by saying something like "you must see this!". It's like how the expression hail fellow, well-met became an adjective of sorts, but it doesn't follow that any long expression can be turned into an adjective like that. So I doubt that people would be using the modal verbs as prefixes or anything. --AtterCleanser44 (talk) 09:08, 18 March 2021 (UTC)

That all makes a lot of sense. Perhaps must/thurt would end up with some separate non-conditional non-past meanings too, but it'd be too hard to foretell, lol. I do think some participles would be useful if someone doesn't want to say "that can [verb]" or particularly "that can be [verb]ed" with relative clauses all the time. Hence why -able became so popular. But I personally find -enly and -bere pretty tenuous, so something like "canread" (literate) and "couldread" (legible) or some variant thereof could be really nice, but I admit the "mustread" is a tenuous thing to base that all off of, especially when trying put in a passive-active distinction. Also, I remembered another helper-like verb (but keeping a "to", along the lines of want): wone (pronounced wun, past tense wont, pronounced wunt). It adds habitualness, so "I wont (or: was wont) to speak" is "I used to speak" or "I habitually spoke". "I wone (or: am wont) to speak" is "I habitually speak" (not that much difference from the plain "I speak", since "perfective present" isn't really a thing compared to our perfective past (preterite) and future tenses, so the simple present became habitually, as well as stative, although some verbs keep a continuous sense anyway, like "smell" and "feel"). And "I will wone (or: be wont) to speak" is "I will habitually speak". Or perhaps the wone/be wont will end up like the "used to"/"be used to" distinction we have today. What do you think of this? TimeMaster (talk) 18:07, 18 March 2021 (UTC)


 * For past habits, we don't need to turn to wont. We can already show former habits with would, e.g., when I was a child, I would walk to the beach every day. Of course, would is also used as a past subjunctive, so if we need to make it clear that we're talking about the past, we can use an adverb such as formerly. And in sentences about past states, e.g., I used to be a singer and we used to live in France, we can say I was once/formerly a singer and we once/formerly lived in France.


 * For the meaning of be used to, though, I think that be wont is a good substitute therefor. For example, I am used to this kind of treatment is now I am wont to this kind of treatment. In other words, won (it's better that we spell it this way, if you ask me) is a verb meaning accustom, and wont, the past participle, is used to mean accustomed.


 * I ought to accustom myself to my new home. > I ought to won myself to my new home.
 * I am accustomed to being underestimated. > I am wont to being underestimated. --AtterCleanser44 (talk) 21:04, 18 March 2021 (UTC)

That's true, I'd put those on the Everyday Anglish Words/English-Anglish Wordbook pages. As well as "would once" to avoid confusion with conditional senses. There's still history in English for using wone (or won, yeah) like this though, it seems, although your idea for having it mean "accustom" (/in the habit of) seems excellent if we have "dwell" and "live" (and maybe others) for the old "inhabit" meanings. What would you give the meaning of "I won(e) something", "I won(e) to something", "I wont something", and "I wont to something"? "I'm used to something", "I'm used to [doing] something", "I was used to something", and "I was used to [doing] something"? It could also keep some old meanings of "live", "continue to be", and "be present (in)". TimeMaster (talk) 22:18, 18 March 2021 (UTC)


 * I'd rather that won were used only for accustom. We already have live, dwell, and abide for inhabit, so it's not as if there were a stark lack of words for that. Hence:


 * I won something - meaningless without an accompanying to, e.g., I will won the visitor to our way of life.
 * I won to something - incorrect, since won needs an object.
 * I wont something - past tense of the first sentence, e.g., yesterday, I finally wont myself to my new schedule.
 * I wont to something - like the second one, incorrect.
 * I am wont to (doing) something - matches I am used to (doing) something. --AtterCleanser44 (talk) 23:02, 18 March 2021 (UTC)

May I ask what your thoughts are on whether we should revive the modal infinitives "cun", "shull", "mow" (may) and "moot" as they could be handy in some cases such as "to cun" instead of "to be able to" as well as to revive "moot" back to the present tense and "must" can be the past subjunctive again. Eversix86 (talk) 21:28, 21 March 2021 (UTC)

We discussed this earlier, they do support bringing back the other present tense moot, but not infinitives, as they think the present tense form would supplant it, as seen in a few rare Early Modern English examples. TimeMaster (talk) 21:48, 21 March 2021 (UTC)

Thanks for the clarification on my first statement and I can see your point but at the same time if we can only see it in a few rare cases in Early Modern English examples then could that really mean that it would happen especially if they're from a few writers? Eversix86 (talk) 17:38, 22/03/21


 * It's hard to say, since the use of the modal verbs as infinitives had become rare in Early Modern English. In terms of attested examples, cun had died off by Early Modern English, whereas can saw marginal use as an infinitive. However, perhaps can would not have arisen as an infinitive form if the use of the infinitive had not fallen earlier on. Perhaps then the irregular forms would have been kept (despite the possibility of analogy). In other words, we might have kept both cun and will as infinitives. It might have been like how German uses both können and sollen, even though there is no change in vowel in the finite forms for the latter. --AtterCleanser44 (talk) 21:06, 22 March 2021 (UTC)

Thank you for that information but what are your thoughts on whether we should revive them and if so how as my opinion is that one could say "to cun" instead of "to be able to" (and the infinitive would work better and may be more logical than "to can" in my opinion) and there may be more possibilities with the other infinitives as well which is why I've thought it's not a bad idea.

Eversix86 (talk) 20:56, 28/03/21


 * Hello, I've sent a message to an account of the same name as yours on Reddit. I've found that talking here is tiring, and I wish to continue our conversation elsewhere, if you don't mind. --AtterCleanser44 (talk) 20:10, 28 March 2021 (UTC)

Contact
Are you on Discord or Reddit? Communicating on wikis is a pain. Hurlebatte (talk) 21:02, 23 March 2021 (UTC)


 * Yes, I've just created a Reddit account. Here it is. --AtterCleanser44 (talk) 21:30, 23 March 2021 (UTC)