User talk:Msh210/Archive/Hebrew roots

This page is an archive of old discussion. Please don't edit this page. If you wish to communicate with me (msh210), you can do so at User talk:Msh210. Thanks!

Well then,

How does this look, then? http://en.wiktionary.org/w/index.php?title=%D7%A9%D7%9E%D7%A9&diff=3202159&oldid=3047252

Thanks in advance, --Connel MacKenzie 22:55, 20 October 2007 (UTC)

Thanks. Only thing is that the "make use" sense, while its root is שמש, is not conjugated as that three-letter word (in any person, tense, etc.). So while there should be some mention of the word on the page שמש, it shouldn't be as the word שמש (not that you put it there). What Ruakh and I have sort of decided at talk:שמש is that there should be a page for the root, possibly ש-מ-ש, which should list all of the terms derived from the root, but without (full) definitions; the page שמש would then (as in English) list all (and only) words spelled שמש. So the "make use" sense would be at השתמש rather than שמש, but would be linked to from ש-מ-ש. And שמש would have "See ש-מ-ש" atop it. That's some of Ruakh's ideas and some of mine I've listed; neither of us agrees with all of them, but it seems a good compromise. The only question I haven't addressed is: as you know from the page שמש, the same word in different constructions (binyanim; e.g., hitpael and piel) has different meanings; and each each construction has a full conjugation table full of words. Each of those words can certainly have an entry ("past-tense, second-person, feminine, plural form of...") but the question is, "form of" what? In other words, where do we put the canonical form of each root-cum-construction? Ruakh says to put it at the past-tense third-person masc., and I have to admit that's traditional, but it seems so arbitrary. It makes more sense to me to put it at the infinitive (not the "to-" infinitive but the bare infinitive), except that certain constructions don't have an infinitive at all (which helps to explain the traditional system of using the past-tense form). So that question is slightly open, although I suppose we'll wind up using Ruakh's suggestion. This is probably a wordier response than you sought, but that's the risk you take in asking me something.  :-)  But the short answer is "good and thanks".—msh210℠ 02:27, 21 October 2007 (UTC)

That seems to be a recurring problem. We don't treat English words that way, though. That is, we could use traditional dictionary-style "lemmas" with all inflections simply listed as "forms." But we don't. For abstract language elements, we do need a consistent way of saying "this doesn't mean anything on its own; it is only used to construct derived terms." In English, we call those prefixes and suffixes, then define what they (partially) mean, anyhow. For CJKV, we have "Particle" which fills a similar role (but some might argue, poorly.) I don't think Ruakh is likely to be rational in any such discussion though...certainly not if I broach the topic again. I don't think he views his cemented-in-place notion (that the traditional way is the only way) as even being inflexible. The Wiktionary way to address these things has been by spelling then by =Etymology=. As much as I dislike the multiple etymology structure, it does prevent the impossible situation of having two (or more) =Verb= heading sections arranged incorrectly as he suggests. If we re-think the "multiple etymology" approach for English (perhaps restricting it to homophones, or inverting the hierarchy of the structure) then it would make sense to follow along in other languages (as he inadvertently suggests on that talk page.) But being inconsistent, for the sake of being inconsistent, just means that bots and English readers will be lost whenever encountering a Hebrew root. In summary: I don't know where to go from here, on this topic. But thank you for reviewing my potentially controversial (can-o-worms) edit. --Connel MacKenzie 03:24, 21 October 2007 (UTC)

I think I'm always rational, but you may disagree. ;-)   I'm O.K. with your edit there; more work is certainly needed, but the existing format had serious problems, and your version is no worse, so if it makes you happy, that's just peachy-dandy. I think I'm fairly flexible about various possibilities discussed sanely, but neither excluding Hebrew roots, or calling them something like "prefixes" or "suffixes" or "particles", is really an option; when you insist on these, we're outside the realm of "discussed sanely". (BTW, when you insist that ELE cannot expand to support headers needed for other languages, and further, when you claim — as you have before — that someone proposing a new header must be doing so in an attempt to destroy Wiktionary — do you not consider yourself inflexible?) —Ruakh_TALK 18:04, 21 October 2007 (UTC)

I agree with your system, Connel, of identifying the root and binyan (construction) only an etymology section: that's consistent with English words' entries. Like Ruakh, though, I'll have to insist on having entries for roots. Not because they're used per se, but because (a) it's traditional to mention them when discussing words, (b) people will look for them, (c) it's a way to relate words like 'make use" and "serve" that are related in form (same root) and often in meaning, and (d) we have, as you mention, such things in other languages (things that meannothing on their own but are used to construct words). (Note that roots are still being used to construct words: e.g., there's a new word (in a new binyan (construction)!) used nowadays — hitputar ("be asked to resign") — which is formed from an existing root.) Like Ruakh, though, I it makes no sense at all to call them "prefixes" or "suffixes": only "root" makes sense as a header. (Well, "triliteral root" also does, but some have four letters, so we'd need another header.)—msh210℠ 16:42, 22 October 2007 (UTC)

Please keep personal comments on BP or somewhere, both of you. This talk page is not for "I don't think Ruakh is likely to be rational in any such discussion" or "when you insist on these, we're outside the realm of 'discussed sanely'" or "I don't think he views his cemented-in-place notion... as even being inflexible" or "do you not consider yourself inflexible?". Thanks.—msh210℠ 16:42, 22 October 2007 (UTC)

Sorry. I'll behave. —Ruakh_TALK 19:51, 22 October 2007 (UTC)

It's true that the masculine third-person singular past tense is an arbitrary tradition, but so is the choice of the infinitive. The infinitive seems a more natural lemma to you because you're a well-educated English speaker; but less-well-educated English speakers frequently add -ing entries here as lemmata (apparently because it's the most noun-like form in normal speech; consider something like "Fooing is like barring, but with a baz instead of a bip"). If the argument is that the shortest/least-affixed form should be used, then the masculine third-person singular past tense makes exactly as much sense as the bare infinitive in most cases, and more sense (as you've noted) in the cases where there is no bare infinitive. I've heard it said that the really least-marked form is something called the "infinitive absolute", like הָלוֹךְ (halókh) (which you might recognize from Shir ha-Ma'alot) and מוֹת (mot), but these forms are incredibly rare for all but a few verbs, so I don't think we'd serve anyone by lemmatizing them. —Ruakh_TALK 18:26, 21 October 2007 (UTC)

Just a brief comment: What you, Ruakh, are calling the "'infinitive absolute', like הָלוֹךְ (halókh)" is what I meant by "the infinitive (not the "to-" infinitive but the bare infinitive)". That form exists, I think, for all binyanim for which the "to +" (l'-) infinitive exists.—msh210℠ 05:34, 22 October 2007 (UTC)

I'm sorry, but I think you're confused about something … the bare infinitive (also called "infinitival construct" and "gerund", but "bare infinitive" is IMHO the best and clearest name) in this case is לֶכֶת (lékhet). (The to-infinitive is לָלֶכֶת (lalékhet).) As I said, the bare infinitive is exactly as short and un-affixed as the masculine third-person singular past tense (never shorter or less affixed), with the additional problem that it doesn't exist for two binyanim. (That said, this isn't necessarily a fatal deficiency, as no one's forcing us to follow the traditional binyan distinction — for example, we could call m'fu'élet the "feminine singular passive present tense and passive present participle of pa'él". So I'm really not arguing that we can't use the bare infinitive; I just don't think the benefits are that great, and it has all the major drawbacks that come with bucking tradition — violation of user expectations, inconsistency with the other Wiktionaries, internal inconsistency due to editors unfamiliar with our standard, etc.) —Ruakh_TALK 15:56, 22 October 2007 (UTC)

You're absolutely right: I am confused about that, primarily because haloch is such an odd word. For most words (at the risk of being wrong again), the haloch form and the lechet form coincide: shamor, bakesh, hikanes, hitpalel. (I guess "ayin-vav" words in kal, like mot and shir and ba, are exceptions also.)—msh210℠ 16:26, 22 October 2007 (UTC)

Both forms are quite rare in Modern Hebrew (except insofar as the bare infinitive exists inside the to-infinitive), so I'm not actually terribly familiar with them. הָלוֹךְ (halókh) and לֶכֶת (lékhet) are my stock examples because they are actually used (the former in phrases like "round trip", the latter in phrases like "planet"). I think they usually differ for pa'al (kal) verbs, where you have (e.g.) shamor vs. sh'mor; but in pi'el at least, they are the same. (I really don't know about the other binyanim. And even with pa'al, I'm not completely clear on the whole thing, as one version of the Fourth Commandment uses shamor as an imperative, as in "shamor 'et haShShabbath l'qad'sho", but in Modern Hebrew we use sh'mor, as in "shmor merkhak", and I don't know what the story is here, or whether it's related at all.) Incidentally, we've been saying the to-infinitive contains the bare infinitive, which is mostly true, but slightly false in that the sh'va na' in a normal pa'al bare infinitive becomes a sh'va nakh in the to-infinitive, which then induces a dagesh kal in a following begedkefet consonant, as in m'khor "sell" → limkor "to sell". —Ruakh_TALK 19:51, 22 October 2007 (UTC)