Disputatio Vicipaediae:De nominibus propriis/en

Agencies named after people or places

Could I get a clarification in the Place Names section? I would like to see a rule for how to translate honorific names that never had an attested Latin name. E.g., the Johns Hopkins Institute, the Francis Drake Memorial Bridge, Martin Luther King Jr. Boulevard. If I follow the existing rules, I would get:

• Johns Hopkins Institute (Institutum Ioannis Hopkins)
• Francis Drake Memorial Bridge (Pons Memorialis Francisci Drake)
• Martin Luther King Jr. Boulevard (Via Martini Lutheri King Jr.)

Are these correct? Would the lemma for the article in all cases be the native language version? -- Robert.Baruch 14:41, 20 Ianuarii 2010 (UTC)

Would the existing rules really give you Institutum Ioannis Hopkins in preference to Institutum Ioannes Hopkins? It isn't Johns Hopkins's Institute. Of course the syntax may be uncertain even in the original! IacobusAmor 15:36, 20 Ianuarii 2010 (UTC)

Would that apply to all the others, then? It's not Francis Drake's Bridge, nor is it MLK Jr's Boulevard. The objects were named after, but not owned by, the person. So what would the forms be? -- Robert.Baruch 15:57, 20 Ianuarii 2010 (UTC)

Also, names of things that are named after their locations when those locations never had an attested Latin name. E.g., Podunk University, Ceciltucky Building. Also locations which have a translatable Latin name. E.g., Rosetown University.

• Podunk University (Universitas Podunk)
• Ceciltucky Building (Aedificium Ceciltucky)
• Rosetown University (Universitas Rosae Oppidi)

Are these correct? Again, would the lemma for the article in all cases be the native language version? -- Robert.Baruch 14:41, 20 Ianuarii 2010 (UTC)

I should think so. But if you can find an attested Latin form of Rosetown (or of Hopkins above, or of any such name) anywhere in the world, in reference to any old Rosetown (or any old Hopkins, etc.), it should be applicable to this one. IacobusAmor 15:36, 20 Ianuarii 2010 (UTC)

At least with the names of persons, the International Code of Botanical Nomenclature, adapted from the International Code of Zoological Nomenclature, attests a process that should be applicable:

If the name ends in a vowel, add i; but if the name ends in a, add e.

If the name ends in a consonant, add ii, except when the name ends in er (when you add only i).

If an adjective is wanted, use the suffix -(i)ana.

Ergo Francis Drake's Bridge = Pons Francisci Drakei vel fortasse Pons Drakeianus, et MLK Boulevard = Via Martini Lutheri Kingii vel fortasse Via Kingiana. Fortasse etiam Rosetown University = Universitas Rosetowniana. An exceedingly well-attested example of the last pattern is Universitas Harvardiana, in which the adjective = Harvard + -iana. IacobusAmor 14:05, 16 Martii 2010 (UTC)

I added

I added a section on names of cultural institutions: we have usually been translating these, and I assumed we want to go on doing so, though the rule was not clear. Thus:

1. Nomina institutionum culturalium (scholarum, academiarum, museorum, societatum eruditarum etc.) vertenda sunt si verti possunt. N.B.: saepe nomina Latina officialia iam habent (vide e.g. VP:UNI); talia nomina officialia praeferimus.

If anyone disagrees, please say so! Also, if anyone could translate this into German, I'd be grateful. Andrew Dalby (disputatio) 13:49, 16 Martii 2010 (UTC)

Hmm, it does seem kind of contra the VP:NF thrust of "if the name is already translated, use a translated form; otherwise, don't translate" that almost every other item on this page expresses. If we do decide to add this rule we might want to clearly outline the reasons that this category of names is a special case, to prevent opening the floodgates to people bringing up their own categories of things they feel should be translated, citing this rule as precedent... ¶ Also we might want to clearly delineate where translatable "institutiones" end and non-translatable "societates" begin — is a local art gallery closer to a business (and thus not to be translated unless already translated) or to a museum (and thus to be translated if at all possible)? Does it depend on some factor that might not be readily apparent to the outside observer? The division seems kind of arbitrary. —Mucius Tever 04:21, 17 Martii 2010 (UTC)

In justification, (a) many such institutions have already been given Latin names, especially if they are long-established; (b) such names are usually translatable, and are in fact commonly translated among modern languages (whereas names of commercial companies usually aren't). Compare our rule about place names (higher on the page): "it is not to be translated unless it uses ordinary or translatable words in its own language, in which case we translate them". The rule I have proposed actually follows the same pattern: "... if translation is possible". With a name like "National Gallery" or "British Library" it is possible. Similarly in English we can translate my nearest university, and call it "University of Poitiers" or "Poitiers University". It would be odd in an English sentence to use its official name and say "I called in at the Université de Poitiers today". I would argue that we are following common international and everyday practice if we translate such names. Andrew Dalby (disputatio) 10:56, 17 Martii 2010 (UTC)

Andrew is quite right with respect to the common practice of English: we ordinarily translate a wide variety of social institutions: we speak of the University of Chile, not the Universidad de Chile; the King of the Belgians, not the Koning der Belgen, the Roi des Belges, or the König der Belgier; the Brazilian Institute of Geography and Statistics, not the Instituto Brasileiro de Geografia e Estatística; the Moscow State Academy of Choreography, not the Московская государственная академия хореографии; the National University of Mongolia, not the Монгол Улсын Их Сургууль; the National Diet of Japan, not the 国会; the Mexican national football team, not La selección de fútbol de México; the Thailand national football team, not the ทีมชาติไทย; and so on. However, common practice sometimes retains phrases from other languages: we ordinarily speak of the Autobahn, not the Federal Motorway; the Ballets Russes, not the Russian Ballets; the Althing or Althingi (though even here a bit of anglicization occurs, since the Icelandic spelling is Alþingi). No "white line" may be discernible, but the (overwhelming?) preference is for translating. IacobusAmor 11:53, 17 Martii 2010 (UTC)

And sometimes English is of two minds: we speak of the International Federation of Association Football, but then we often shorten that to the acronym FIFA, not IFAF. Thank you, France! IacobusAmor 12:08, 17 Martii 2010 (UTC)

Mind you, English is almost certainly a special case because of its worldwide use (an entity is likely to have a name for itself both in English and its own language), and when foreign scripts are involved it's another matter altogether. (Also, are we really saying the King of the Belgians is an institution that would fall under this rule?) —Mucius Tever 00:02, 18 Martii 2010 (UTC)

Well, if not the king, then the Belgian Kingdom (Koninkrijk België, Royaume de Belgique, Königreich Belgien) might be considered an institution. IacobusAmor 04:16, 19 Martii 2010 (UTC)

I say it's not necessarily so — I suppose the bit that gets me is that we say we must translate it if possible. The existing rule, about companies, says to translate if it's normally translated by other languages, but sometimes this is done and sometimes it isn't — if I watch Spanish television out here, commercials for companies and institutions usually don't translate their names (indeed, they often pronounce them with an American English accent). Thus, for example, on the British Library's website, there is a page for international visitors with introductions to the library in various languages — and they start with, e.g. "La British Library es la biblioteca nacional del Reino Unido," "Die British Library ist die Nationalbibliothek des Vereinigten Königreichs," "La British Library est la bibliothèque nationale du Royaume-Uni," "British Library は、英国の国立図書館です," etc. and using the English name of the institution throughout. (Note also that the wiki articles of several European languages leave the name untranslated as well.) —Mucius Tever 11:32, 17 Martii 2010 (UTC)

I should think that trademarks must be retained in their registered forms. What Andrew seems to be talking about is a more generic process, especially in regard to institutions whose names fit into any of numerous patterns: the University of X, the National Institute of X, the President of X, the X (city, state, national) Library, the X Museum, the X Gallery of Art, the X (city, state, national) Orchestra, the X Navy, the X national Y team, and so on. IacobusAmor 12:03, 17 Martii 2010 (UTC)

Yes, you're right, that's exactly the kind of example I have in mind. But I agree that the borders are hazy.

One also notes an increasing marketing-led tendency among institutions (not only commercial ones) to insist on a corporate identity with a fixed name, so I'm not at all surprised that the British Library's website does this. However, wikipedias tend to use "the commonest name" for their titles, not necessarily "the full official name" or "the precise registered name", and I think wikipedias are right. Our particular problem, of course, is that no names are common in modern Latin ... Andrew Dalby (disputatio) 12:43, 17 Martii 2010 (UTC)

I think part of the issue is that a lot of the things being put forward as things that should be translated are actually things that do not have their own names — they're called by descriptors instead of names. The British Library has apparently decided it is not just the British Library, but that "British Library" is its name. Names, of course, are the chief focus of the VP:TNP policy, but there's still the trouble with translating descriptors that there is not necessarily a fixed way to do it: even you gave "University of Poitiers" and "Poitiers University" as alternatives, and this ambiguity is precisely why VP:UNI is useful—to collect official name renderings—and why VP:NF is necessary—to prevent people from arbitrarily preferring one possible form over an equally valid one when we, as an encyclopedia, have no actual authority to cite or to exercise in favor of either; cf. the "Mirbellus" discussion at Talk:Octavius Mirbeau, where the fact that "Mirbellius" would have been equally likely was almost entirely ignored by the promoter of "Mirbellus". ¶ And of course, the issue of 'no names being common in modern Latin' is another reason we have VP:TNP: to discourage people from creating a translation and using the strength of Wikipedia's name to promote it—when it's rather our job to find information, not create it. ¶ Also I'm thinking the above-mentioned rule about place names of "it is not to be translated unless it uses ordinary or translatable words in its own language, in which case we translate them" may well be a poor rule itself (my mistake for putting it in). It's certainly not a common practice — for example, given the interwiki links at en:Colorado Springs, Colorado we're the only one who translates 'springs' (and in the singular, for whatever reason; we have Fons Coloratensis, no source cited). —Mucius Tever 00:02, 18 Martii 2010 (UTC)

On the less-travelled roads that cross the boundary into Colorado Springs, the signboards read "Colo Spgs". Or at least they did in 1968. Translate that!

I agree with every word, Mucius [except where you say "a lot of the things being put forward as things that should be translated are actually things that do not have their own names": I raised the issue of cultural institutions only, and the point about them is that they do have names, but the names are often translated. If Iacobus has put forward other things that raise different issues, please let's deal with them separately!]

So, let me now choose another neighbouring institution, more recently founded, whose Latin name may not as yet exist (at least, I can't currently verify it). Since English (and other language) speakers are indeed free to invent one or several translations of Université de Nantes -- a name that is made up of words even more easily translatable into Latin than into English -- are you arguing that we should not call it "Universitas Namnetensis" (using individually verifiable Latin words, but a phrase that I can't verify); whereas we can call its neighbour Universitas Pictaviensis because that can be verified?

The thing is that any writer of modern Latin (e.g. in a botanical diagnosis, or in the Latin preface to an edited classical text) would feel free to create this term and may do so any moment now, just as a writer of English would feel free to create "University of Nantes". I feel doubtful that we should prevent ourselves from doing likewise.

[Added later:] There's a linked issue, that of categories. We have about three hundred categories for teachers and alumni of universities/colleges (nearly all created by me, Hendricus and Schulz-Hameln). These use verifiable Latin names wherever such can be found (and I have added many verified names to the page VP:UNI); where not, they use newly-minted Latin terms (and they are occasionally renamed when an existing Latin form turns up). Would you argue that we shouldn't do this, and instead of Categoria:Alumni Universitatis Algeriensis we should have Categoria:Alumni جــامــــــعة الجـــــــزائر - بن يوسف بن خـدة (I don't currently know how to find a transliteration of this official name, and the French and English pages don't help)? Or, what should we do? Andrew Dalby (disputatio) 13:15, 18 Martii 2010 (UTC)

Well, I suppose the point I mean to make is that there are differences between us and other writers of modern Latin. The person who writes a botanical diagnosis or prefaces a classical text is in the business of producing knowledge about the world in Latin—but as an encyclopedia (at least, as an encyclopedia run on Wikipedian principles) it's our job to collect knowledge about the world in Latin, not produce it. Sometimes it just happens that there is no knowledge about a thing in Latin. Certainly we might write 'Universitas Namnetensis', but what authority do we have for it? It may sit with its [citation needed] tag indefinitely—or worse, be posted without one—giving the appearance that we as a project don't care about verifying our information. This is partly what we have the doctrine of "don't make things up" for, and I suppose my feeling (at the moment) is that it should apply to names of things when they happen to be phrases just as much as when they happen to be single words. We might talk about an "universitas Namnetensis" indefinitely but it's not up to us to suggest that "Universitas Namnetensis" is its name—the university, the botanist, or the classicist might just as easily have used 'Nannetensis' or 'Nantensis' or, confusingly, 'Nanceiensis' (as the Catholic Encyclopedia has it). ¶ As for the Arabic—which is "Jamiat al-Jazair Benyoucef Benkhedda," or something like "Ǧāmiʿaẗ al-Ǧazāʾir – bin Yūsuf bin H̱idaẗ" in strict transliteration (though I'm not sure about the vocalization of the man's name)—and for similar examples, we already have rules about transliteration... on VP:TNP itself, actually. I do agree that Categoria:Alumni Gamiat al-Gazair would be unusual, but the principle of transliterating but not translating when there is no established usage in the language is the English WP's rule and, to borrow language from another naming policy page, does allow us to sidestep the sometimes-difficult question of what a thing should be called, instead focusing on what it is called. (WP:UE suggests "If there is no established English-language treatment for a name, translate it if this can be done without loss of accuracy and with greater understanding for the English-speaking reader" which, though still too generous for my taste here, might even be a little better than the current phrasing which wants a translation whenever possible.) Mind you, Universitas Algeriensis is ambiguous (I looked for 'University of Algeria' first; I suppose 'Universitas Icosiensis' would be too archaizing?) —Mucius Tever 23:51, 18 Martii 2010 (UTC)

(P.S. — it does appear to be Universitas Nannetensis. —Mucius Tever 00:07, 19 Martii 2010 (UTC))

(Heck, while we're at it, Universitas Algeriensis is attested as well. —Mucius Tever 00:10, 19 Martii 2010 (UTC))

OK, OK, but (as you surely understand) I am using these merely as examples. We both admit that there are cases where institutions of this kind have as yet no attested Latin name. Those are the cases under discussion. By all means let's choose better examples :)

A side issue: your interesting citation of Catholic Encyclopedia shows that any source, even on what appears to be its home ground, can be unreliable. See the discussion at Disputatio:Colcata#Caliquit for another such case. Andrew Dalby (disputatio) 10:14, 19 Martii 2010 (UTC)

You can ignore the appended attestations for the purposes of my argument; I didn't know of them till after I had written the longer comment. ¶ Yes, the Catholic Encyclopedia's Nanceiensis is surely an error (Nanceiensis et Tullensis for "Nancy and Toul" is on the same page). ¶ As per my points below, though, when given names, we can often say if one is most correct or if one appears to be an error, but when creating at a name, even with informed guesses, we can't know what is correct—in part because what is correct depends greatly on what is used, and if we knew that we wouldn't feel the need to create the name in the first place—and even if our creation might be right, until we have an authority to back it up, we can't suggest that it is right—in contrast to the native name, which, even if unwieldy when used in or transliterated into another language, certainly is correct. —Mucius Tever 22:04, 20 Martii 2010 (UTC)

I take your points (and the need for accuracy and citations) very seriously, but your words "when creating ... a name, even with informed guesses, we can't know what is correct" seem to me to take a step towards a different world. In the world I live in, there seem to be many more things than in yours! There are names that lack any authority certifying them as correct; there are names with conflicting authorities; there are authoritative names so unwieldy that Wikipedia would never use them in a pagename; and there are things named in some foreign language for which, if I'm to speak of them at all, I have to find a way of doing so in a language I can manage. And with all this I haven't begun to suggest that any particular way of speaking or writing "is right". Nor does Wikipedia, so far as I remember. Andrew Dalby (disputatio) 22:26, 20 Martii 2010 (UTC)

"There are names that lack any authority certifying them as correct/there are authoritative names so unwieldy that Wikipedia would never use them in a pagename" — I apologize for being unclear. In general when I have been mentioning authority I have been referring to the authority of usage, not names imposed by an authoritative figure. Indeed, I think my whole feeling about this point is because we are an authoritative figure and we need to be careful with the words we use—if we put a word in a lemma, people will think of it as authoritative, however misguided (or not) it may be. (Along the same lines, you've already seen my dislike for Vicipaedia's custom of putting Formula:Fontes desiderati in footnotes when across the rest of the Wikipedian universe footnotes are recognized as giving a statement more authority—in effect, we are suggesting indeed that it "is right".) Now, Cicero once said, to a similar issue, "usum loquendi populo concessi, scientiam mihi reservavi," and I think that rule applies to us — just because we "know" what the name ought to be, it's up to the world of Latin that we are observing to come up with it, and if they want to spell it Nannetensis, so be it, and if they haven't said it at all, we can't predict what they will say. ¶ "there are names with conflicting authorities" — Even with your usage of 'authority' I do know about that—hence my stating the need to choose the best (and several criteria along which bestness can be judged), when we have multiple options. ¶ "there are things named in some foreign language for which, if I'm to speak of them at all, I have to find a way of doing so in a language I can manage" — That may be so if you're writing elsewhere, but if you're writing for Vicipaedia, why should it be a language you can manage? Your userpage suggests that the languages you're familiar with are different from the ones I am, and probably every Vicipaedian has their own list—it's likely we have users who could find "Gamiat al-Gazair" as ordinary in Latin text as we would find "the École des Beaux-Arts" ordinary in English text. —Mucius Tever 00:42, 25 Martii 2010 (UTC)

You're right, I didn't understand your use of the term "authority" and therefore lost your argument completely. Likewise, it's evident that you've lost my argument, to judge by these words of yours: "if you're writing for Vicipaedia, why should it be a language you can manage?". I was trying to make a general point, but I should have said "I have to find a way of doing so in a language I and my interlocutors/readers can manage". All should then be clear: in the case of Vicipaedia, the language in question is Latin. That's why we're here. Andrew Dalby (disputatio) 22:54, 25 Martii 2010 (UTC)

Still, I think, the point holds: "you and your interlocutors/readers", in the case of Vicipaedia, is potentially the entire world. We can't make assumptions about what would be unfamiliar or difficult for them—especially given the assumption that people who have interest in a thing are likely to be closer to it to begin with. But making up something new would certainly alienate them. If I were talking about the N|u language in English I would have to pronounce it with the click sound, un-English as it may be—if I simplify it "so that it's in a language I can manage", I can't be sure of being understood, because whatever name I make up won't be the usual name for it—in fact, by trying to Englishify it, I am making it so that is actually not English at all. The same would hold for Latin. We can learn to recognize a foreign name; but if we have to invent a Latin name, we are doing so because the name is not Latin—the form we create would just be our idiolect. —Mucius Tever 11:34, 26 Martii 2010 (UTC)

I still have the impression we live in different worlds! As it happens I have never talked to anyone about /Nu, but I have had conversations about Shona and Ndebele. I know, therefore, that while local bilinguals and experts have a particular pronunciation of these names, using vowels and consonant groups that are unfamiliar in English generally, the great majority of English speakers might have difficulty at first even in recognising these names when they are pronounced in the expert way, and certainly do not pronounce them in that way. In conversations about language names (these and many others) in which non-experts participate, phonological compromises actually are made. That much I know from lifelong observation, and although I've never tried to do it I strongly suspect that the resulting forms are not just [anyone's] idiolect but could be described and predicted in a (rather specialised) grammar. If I were to talk about /Nu to anyone, I would have to make such a compromise myself because I have never learnt to pronounce the alveolar click. Well, if I'm applying your analogy correctly, I would say that the majority of readers of any general encyclopedia are not local experts (in other words, I don't accept your assumption that people who have interest in a thing are likely to be closer to it to begin with as being a rule-of-thumb we can work with: at the very least, we're also writing for people who click on "random article"). Writers of encyclopedias definitely need to make assumptions about what would be unfamiliar or difficult for readers. Even the decision about what to link in a piece of wikipedia text depends on the making of such assumptions. Our ability to select words to link, in itself, disproves your assertion that we can't make such assumptions: we can and do, every day.

I am pessimistic about this discussion, Mucius. Aren't you and I wasting one another's time really? Should we just leave it? Andrew Dalby (disputatio) 01:19, 27 Martii 2010 (UTC)

My mindset is generally very un/non-bureaucratically tuned, so I may be asking wrong questions. (I generally do). Consider the sentence "... but as an encyclopedia (at least, as an encyclopedia run on Wikipedian principles) it's our job to collect knowledge about the world in Latin, not produce it. Sometimes it just happens that there is no knowledge about a thing in Latin." Is this supposed to mean that we are expected to "favere linguis", if nobody extra Vicipaediam has treated an issue (such as memetics, postmodernism, cognitive linguistics, evolutionary economics, emergentism, etc) in Latin? I guess what I'm opposing most, is the possible implication or role setting that as Vicipaedians, we are ex definitione inferior Latinists wrt those blogistanians or tongue-in-cheek self-asserters who happen to produce googleable vestiges in Latin(isque). As a case in point, let me remind of the "authority" who elaborated te idea that blog is "blogis" in Latin, etc. Sometimes it just may be the case that it's us who have to produce a piece of knowledge in Latin, because others don't care or fail to do it satisfactorily. I'm sure that I've misunderstood the whole issue (this is my general experience in situations like this), but the irresponsible homunculus in me fooled me into this, again. :-) --Neander 02:50, 19 Martii 2010 (UTC)

Well, certainly it's not a matter of us being inferior. But let's take the 'blog' example. We already have several guidelines in place about names of things (which we mostly see when people propose strange ones). First off we'd see, is it an attested term, or is it something made up for Vicipaedia? If the latter, we throw it out per VP:NF. If then there are multiple options, we'd then look at attested terms and find the most common terms. ('Blogis' would be thrown out at this step as it was a one-off creation not in use anywhere.) If then there are multiple equally common (or equally rare) forms, we'd prefer the most accurate or correct names. ('Blogis' would most likely be thrown out at this stage too, for several reasons.) If then there are multiple equally accurate names, we'd prefer the most classical. And if then there are multiple equally classical names, perhaps we'd prefer the most familiar to modern eyes; etc. Take the 'university of X' example. It's no trouble to translate if we mean that 'universitas X' generically means 'a university in X', not the name of the university, but if we want to propose that the name of the university actually is 'Universitas X' then we have trouble with all these rules: first off, we are making it up, so it fails VP:NF; the existing (native) name would win. Second off, since we need to make it up, it can't be common; the native name would win here too. Third off, since we need to make it up, we have no way at all to confirm it would be correct—we might be right (as with Universitas Algeriensis) or we might be wrong (as with Universitas Namnetensis instead of Nannetensis), but either way, we are guessing; the native name wins here too. Fourth off, since we need to make it up, it can't be a classical term. (Admittedly, this round would be a tie.) And even at the fifth stage, choosing the most familiar name, the native name is very likely to win here too—unless, for whatever reason, translated versions of the name are more familiar than the native one and the Latin translation can use visibly cognate words (and I think at this point I would support the rule to translate if it had both of those conditions). Someone outside of Vicipaedia, composing Latin, need not care about all these details. But we're a reference that should try to stick to cited sources and for us those details are more important. —Mucius Tever 01:21, 20 Martii 2010 (UTC)

Neander is right. A reference work doesn't necessarily exclude "new knowledge." Even a new arrangement of old facts—such as all the wikis in effect make—can be the equivalent of "new knowledge" (and can in some nonwiki realms be copyrighted). Encyclopedias may & do publish "new knowledge" for the first time. That's not a speculation, but a fact. Ergo, Andrew's point is still well taken. ¶ And that doesn't mean that Vicipaedia has to accept attested absurdities like blogis, or the mistranslation of Campifons ('Spring of the Field, Field's Spring') for a concept that's recte, aio, Fontium Campus 'Field of Springs, Springfield' (unless it's really Veris Campus). IacobusAmor 04:16, 19 Martii 2010 (UTC)

Well, of course a reference work doesn't necessarily exclude new knowledge—but this one does, by the principle of en:WP:NOR. And even the new arrangements of old facts themselves are discouraged when the sources don't support the arrangement, by en:WP:SYNTH. —Mucius Tever 01:21, 20 Martii 2010 (UTC)

en:WP:NOR (including en:WP:SYNTH) is, according to its own statement, an "English Wikipedia" policy. But many other wikipedias have their own version of that page, and it's probably about time we had one too. Andrew Dalby (disputatio) 09:57, 20 Martii 2010 (UTC)

I don't know where the idea came from that "it's our job to collect knowledge about the world in Latin, not produce it. Sometimes it just happens that there is no knowledge about a thing in Latin." The makers of the 270-odd other Wikipedias are more enlightened than that! Yes, knowledge is largely transmitted though language; fortunately it's become unfashionable in recent years to make a rule against translating it.

Back to the point at issue. We've all said that the edges of this question are hazy. Would it help to sharpen those edges, and produce a consensus, if we inserted the caveat (already used in a different context on this page) names that are commonly translated? This could often be verified in a moment by looking at the interwiki links. If the name is translated into at least some other languages, it would seem pointless to rule out its translation into Latin. Andrew Dalby (disputatio) 13:00, 19 Martii 2010 (UTC)

If I were forced at gunpoint to make a policy, I'd use WP:UE, adding that the term should be accompanied, at its first usage, by a parenthetical remark containing its original language and term in that original language, with transliteration or pronunciation if the original language is not written in a Latin alphabet (a formula would make this easy). ¶ For new terms which are translatable, I'd use circumlocution the first time, and if the circumlocution is too long to repeat thereafter, add a parenthetical remark containing the original language, original language term, transliteration or pronunciation if necessary, and use either the original language term or the transliteration thereafter. The problem that I'd struggle with would be the inflection of that term in cases other than the nominative, as well as the assignation of a gender. ¶ For terms that are not translatable (e.g. last names), use the original term, and parenthetically add the original language, and translation or pronunciation if necessary. Again, I'd struggle with gender and inflection. ¶ I think the parentheticals would serve as an admission that these are translated names, and that they may not carry all the connotations of the original. So that's my stake in the ground :) --Robert.Baruch 21:23, 19 Martii 2010 (UTC)

I think we already try to include the original term in parentheses right after the boldfaced lemma. IacobusAmor 00:06, 20 Martii 2010 (UTC)

I would also add that names that are commonly translated should read names that are commonly translated by users of various language editions of Wikipedia. Unless you can find modern-day native language and origin attestations. When I lived in New York City, I would regularly hear Spanish-language broadcasters refer to "the Verrazano-Narrows Bridge" in an abrupt NYC English accent. However, I would hardly use that as an attestation that "Verrazano-Narrows Bridge" is commonly not translated into Spanish, because that would only be true of the Hispanic population in the New York metro area, but not necessarily of, say, Peruvians in Peru, the Spanish in Spain, or Guatemalans in Guatemala who might want to read a Spanish-language encyclopedic entry or newspaper article in Peru, Spain, or Guatemala, about the bridge. --Robert.Baruch 21:32, 19 Martii 2010 (UTC)

De: "For terms that are not translatable." Some will be imported anyway; and for them, may I propose the Corollarium Amoris: the Latin nominative should phonetically be as close to the original term as possible. (The Finnish & Polynesian systems already follow this plan.) For the English term Chicago, we'd thus have Chicago, -inis, f. (or Sicago, -inis, f.). We know that the stem is Chicagin- (or Sicagin-) because learned attestations give us the accusative Chicaginem and the adjective Chicaginiensis (or Sicaginiensis) here and here. Yet because later & perhaps less informed sources ignored this tradition and attested something else, we've gotten stuck with Sicagum, in which Sicago (or Chicago), most unexpectedly, must be a dative or an ablative, rather than a nominative. IacobusAmor 23:59, 19 Martii 2010 (UTC)

Incidentally, 'Sicagum' actually does follow a rule that the Latin should be as close to the original term as possible. The second declension, after all, is the o-stem declension, indicating a stem 'Sicago-', while 'Chicago, -inis' presupposes that the stem of the word is 'Chicagin-'—a very unusual hypothesis; even ignoring the excrescent 'n', the -inis form is strange, especially considering that -o, -onis words are not at all unusual in Latin—cf. histrio, sabulo, carbo, hundreds of words in -tio, and place names like Olisipo, Pompelo, Hippo.... 'Chicago, -inis', attested or not, doesn't look like a "more informed source" so much as it looks like faux classicization (I'm guessing the example of 'Carthago' is to blame). —Mucius Tever 14:25, 20 Martii 2010 (UTC)

I completely agree, Muci! Chicagum, Tokium, videum, eurum are preferable over Chicaginem, Tokionem, videonem, euronem, since not only the nominative, but also the oblique cases are to be considered. Also Google supports this: Chicagiensis 306 Google Books hits, Chicaginensis 230 hits; Tokiensis 541 hits, Tokionensis 15 hits. But if it is vice versa, the prevailing usage shall always take precedence, regardless any scientific opinions: for example, "milia euronum" is hopelessly more common in plain Google (164 hits) than "milia eurorum" (5 hits). Nevertheless, I hope Iacobus' rule "the Latin nominative should phonetically be as close to the original term as possible" was only meant to apply when we compare some scarcely used Latin alternatives (or when we, for some unknown reason, happen to invent a new Latin name). The thing I enjoy in Latin are its beautiful traditional place names, instead of a mere copy of what local nationalists ruled to be the official sacrosant untouchable name. --Gabriel Svoboda 18:57, 20 Martii 2010 (UTC)

As attested in a link above, the University of Chicago itself, in its commencement ceremonies, used the form Universitas Chicaginiensis, spoken by its own president, William Rainey Harper, who edited books in Latin and Greek. What could be more authoritative than that? What solider an attestation do you want? IacobusAmor 23:34, 20 Martii 2010 (UTC)

I don't think the university owns the name of the city, nor does that constitute even an official use by the university. Pantocrator 23:51, 20 Martii 2010 (UTC)

Evidently, the attested Latin name of the university is not only official, but part of popular lore. From the College Magazine of the University of Chicago in the summer of 2009: "David Strubbe, SB’05, caught a glimpse of his favorite 'in Woodward my first quarter in fall 2001. I never saw it again,' he writes. It read, 'Universitas Chicaginiensis/ Universitatis Chicaginiensis/ Universitati Chicaginiensi/ Universitatem Chicaginiensem/ Universitate Chicaginiensi. I wish I had declined the University of Chicago sooner.'" ¶ Yes, the university doesn't own the name of the city, but it does own its own name: if it prefers to call itself the Universitas Chicaginiensis and the City of Chicago prefers to call itself Sicagum, the world will have to accept the inconsistency. IacobusAmor 00:18, 21 Martii 2010 (UTC)

Good Lord. The motto of the University of Chicago was invented by Professor Paul Shorey, the Latin scholar who edited Horace Odes and Epodes, the school text that for decades served as the standard edition of Horace in many U.S. colleges and universities. If he accepted the name of Universitas Chicaginiensis, who are we to question it? IacobusAmor 00:37, 21 Martii 2010 (UTC)

Well it does seem that that name is in some sense official. I suppose if we make a page on the university, that's what it should be called. Pantocrator 01:02, 21 Martii 2010 (UTC)

The University of Chicago, no matter how experienced Latinists it may have, does not really own the city's Latin name, because the city is mentioned in many more Latin sources than just the universitarian ones. Universitas Chicaginiensis, the specie Oenothera chicaginensis, the restaurant Floreat Taberna Chicaginiensis or the T-shirt Schola Latina Chicaginiensis, Fidelitas do not beat Archidioecesis Chicagiensis or Universitas Loyolae Chicagensis as per Google:

• Chicaginensis 700 hits, Chicaginiensis 93 hits (both adjectives coming from Chicago, -inis)
• Chicagiensis 1570 hits, Chicagensis 568 hits (both adjectives coming from Chicagum)

It can't be any Vicipaedian influence, Google Books show similar results. --Gabriel Svoboda 09:50, 21 Martii 2010 (UTC)

There are other cases where the usual or official Latin name of a university conflicts in spelling with the name that we would normally choose for the city: Edinburgh, I seem to remember, and Nantes as Mucius points out somewhere above. In some cases we might take the university's lead, but in others it's a question of making redirects. Andrew Dalby (disputatio) 09:53, 21 Martii 2010 (UTC)

As for the 'ch' spelling, I don't understand why we can't use 'sh' for that sound, which at least is unambiguous. 'Ch' in Latin is an aspirate or a fricative (depending on your pronunciation), but certainly not a sibilant! (The only classical way is 's', ignoring the difference between the sibilants altogether. I see no reason why the French 'ch' should in general be preferred to the English 'sh'.) Pantocrator 23:51, 20 Martii 2010 (UTC)

We don't use "sh" for that sound, because we don't represent sounds, but letters. Latin is a spelling-based language, since it has got a common orthography, but not at all a common pronunciation. (The i/j/u/v orthographic issues are merely technical details compared to the at least five traditional pronunciations of soft g, for example.) The spelling approach only fails outside the Latin alphabet world, there we might indeed want to find some kind of sound transcription, but America certainly is not the case. I also don't understand why the city is called "Chicago" rather than "Shicago" in English, but we are here to find a Latin name, not to question the vernacular one. The traditional Latin name is Chicagum, which is going to be pronounced sometging like /k(h)ikagum/, and I don't see any problem with it. --Gabriel Svoboda 09:50, 21 Martii 2010 (UTC)

----

Indeed, I think the correct answer as to why we can't use "sh" is because it is not a standard Latin sound. (Some traditional pronunciations have it, though.) And since it is not a standard Latin sound, the least innovative usage would be either to retain the original spelling (since there is no standard spelling to change it to) or to classicize the pronunciation and then spell accordingly (hence Sicagum). (The latter is not always easy—what's the classicizing pronunciation of an alveolar click? Even the affricate 'ch' is controversial in that regard.) Of course, the other reason "sh" is no good for that sound (in general, not just in 'Chicago') is that "sh" is a particularly English spelling. Besides the French "ch", you might just as easily propose "sch" or the Italian "sc(i)" or the Iberian "x"—indeed, "x" is pretty common for "sh" in the early Latin writings on Japan, e.g. Ximum for Shimo, i.e. Kyushu). —Mucius Tever 01:28, 25 Martii 2010 (UTC)

The Lemma

Just a little table to see what other wikis are doing, gathered together into one convenient place. I'll expand this to other "popular" entries and other languages, as well as add subsections for the first mention and subsequent mentions. Hopefully this will let us get a feel of what the Wiki Gestalt does for names. --Robert.Baruch 03:00, 20 Martii 2010 (UTC)

Good idea. I'm adding one or two of the kinds of institutions I was originally thinking about (including one with asteroidal relevance). Please check my work, and add what the Japanese do (I can't read Japanese). At a first glance, what I notice (but no translator, and no reader of large reference works, would be surprised by this) is the stunning inconsistency. Why is "National Gallery" treated so differently from "Natural History Museum"?

[Added:] I can see (again speaking as a translator) a possible reason why Empire State Building is treated differently from Brooklyn Bridge. "Empire State" originates in a nickname, and nicknames are rarely translated. Andrew Dalby (disputatio) 10:03, 20 Martii 2010 (UTC)

Key:

• ⇒ = translated,
• → = partially translated,
• τ = transliterated
• T = partially transliterated, partially translated
• — = not translated or transliterated
• o = name not mentioned after lemma in body of article

This is sometimes the sign of a stub, or sometimes a generic term is used for the name (e.g. "the university").

• x = article does not exist

Sometimes an entry will have two symbols separated by /. This means that the name is translated or transliterated inconsistently throughout the article.

Names originally in English.

Form of the lemma
Article
en:Brooklyn Bridge	⇒	—	⇒	⇒	—	T	⇒	⇒
en:Empire State Building	→	—	—	—	—	τ	τ	T
en:Palomar Observatory	⇒	⇒	⇒	⇒	—	T	⇒	x
en:University of Sussex	⇒	—	⇒	x	⇒	T	x	x
en:Brigham Young University	⇒	—	—	x	⇒	T	T	T
en:Natural History Museum	⇒	—	⇒	⇒	⇒	⇒	⇒	⇒
en:National Gallery (London)	→	—	—	—	—	τ	⇒	⇒
en:New York City	⇒	—	—	—	—	τ	τ	τ
en:Blog	—	—	—	—	—	τ	τ	τ
en:Jimmy Carter	—	—	—	—	—	τ	τ	τ
es:Universidad Autónoma de Barcelona		⇒	x	⇒	x	x	x	x	⇒
es:Universidad Iberoamericana		—	x	x	⇒	x	x	x	—
es:Puente de Vizcaya		—	⇒	⇒	⇒	T	T	T	⇒
es:Mezquita Catedral de Córdoba		—	⇒	⇒	—	τ	τ	x	⇒

Form of the usage in the article
English
en:Brooklyn Bridge	⇒	—	⇒	⇒	—	T	⇒	⇒
en:Empire State Building	→	—	—	—	—	τ	τ	τ
en:Palomar Observatory	⇒	⇒	⇒	o	—	T	⇒	x
en:University of Sussex	⇒	—/⇒	o	x	o	T	x	x
en:Brigham Young University	⇒	—	⇒	x	⇒	T	T	T
en:Natural History Museum	⇒	—/⇒	⇒	o	o	⇒	o	⇒
en:National Gallery (London)	→	—	—	—	—	τ	o	⇒
en:New York City	⇒	—	—	—	—	τ	τ	τ
en:Blog	—/⇒	—	—	—	—	τ	τ	τ
en:Jimmy Carter	—	—	—	—	—	τ	τ	τ
es:Universidad Autónoma de Barcelona		⇒	x	⇒	x	x	x	x	⇒
es:Universidad Iberoamericana		o	x	x	o	x	x	x	⇒
es:Puente de Vizcaya		⇒	o	o	o	T	T	T	o
es:Mezquita Catedral de Córdoba		⇒	⇒	⇒	—	τ	τ	x	⇒

Linguam Russicam addidimus. ¶ Note that Deutsch partly naturalizes the first one, under the name Die Brooklyn Bridge, not The Brooklyn Bridge. Also Das Empire State Building, not The Empire State Building. Also Das Natural History Museum, not The Natural History Museum. So terms that have no gender in English (bridge, building, museum) have gained gender in German. That's a sign of partial translation. IacobusAmor 12:13, 20 Martii 2010 (UTC)

I wouldn't say it's a very good sign of partial translation — just a sign that the language requires a definite article and doesn't have a gender-neutral one (cf. the English École nationale supérieure des Beaux-Arts, which refers to the école). The genders in the German case are those of die Brücke, das Gebäude, and das Museum, and you might just as easily consider the titles to have those nouns "understood", as if it were internally "die Brooklyn-Bridge-Brücke", etc. Note that in the last case, as the German is the same as the English, the "Natural History Museum" article may be considered to have a partially translated title — note that the caption on one of the images does decline 'des Natural History Museums'. As for the missing Japanese entries in that table:

• "パロマー天文台" (Paromā Tenmondai) is Palomar transliterated, observatory translated;
• "サセックス大学" (Susekkusu Daigaku) is Sussex transliterated, university translated;
• ロンドン自然史博物館 (Rondon Shizenshi-Hakubutsukan) is London transliterated, Natural History Museum translated;
• ナショナル・ギャラリー (Nashonaru Gyararī) is National Gallery transliterated.

I've updated the table. —Mucius Tever 14:04, 20 Martii 2010 (UTC)

I've also added some from Hebrew, which is supposed to be the one right next to Russian, but I'm not sure Wiki's SVG->PNG engine is playing nice with right-to-left scripts... --Robert.Baruch 19:39, 20 Martii 2010 (UTC)

It's a pity the languages of the insular Pacific don't have wikis with these terms, as they'd refine this table by showing a different pattern entirely: just about everything would be translated, or at least respelled to suit local phonetics. For example, in Fijian & Samoan (&c.), the name of the city of New York is commonly written Niu Yoka, and in Hawaiian it's Nu Yoka. A placename of more local relevance, Niu Sila (New Zealand) gets 38,000 hits at Google; since the islands are impopulous and have only a tiny number of internet users (compared with the continental world), that should be considered a huge vote in favor of rephoneticization. For The United States of America, the Tongan wiki has Puleʻanga Fakataha ʻo ʻAmelika, the Hawaiian wiki has ʻAmelika Hui Pū ʻia, and the Samoan wiki has Iunaite Setete o Amerika (the most linguistically respectable Samoan-English dictionary gives it as ʻO le Soʻo Faʻatasi o Amerika). We shouldn't be surprised to find that the vast majority of the (several thousand) languages of the world—about seven hundred are spoken in New Guinea—would respell & rephoneticize in this way, not preserve indigenous spellings in the manner seen most egregiously in the Nederlands wiki. IacobusAmor 13:11, 21 Martii 2010 (UTC)

I think that is neither translation nor transliteration, but rather transcription. Yet another dimension. Also, LOL: sco:New York Ceety --Robert.Baruch 15:47, 21 Martii 2010 (UTC)

Call it what you will: it's what happens in writing, and it results in an orthograpy that may be taken as standard for any of the mentioned languages. IacobusAmor 16:55, 21 Martii 2010 (UTC)

Latin in the age of Plautus did this — with Greek υ spelled as u, the aspirates spelled unaspirated, and ζ spelled as ss—but by the classical age whose Latin we tend to prefer the practice had changed; those Romans started transliterating foreign sounds instead: spelling the Greek aspirates with 'h' and borrowing Y and Z wholesale. (And then in later times everything got variable.) Plautus might have written "NVIORC", transcribing by pronunciation instead of using our spelling, but I doubt a classical author would have. Standard Latin was not written phonetically even in native words, as many of the grammarian's peeve lists would indicate. —Mucius Tever 00:42, 25 Martii 2010 (UTC)

Conclusions based on the data

Enumerated above are a number of articles whose lemmas in English consist of a combination of proper nouns (names, localities) and common nouns. Proper nouns can themselves be composed partially or fully of common nouns, for example "New York" and "Empire State". Common nouns can also be neologisms, such as "blog". --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Note that there may be exceptions based on the lemma itself. I don't seek to find each of these exceptions, unless they can be inferred to comprise a class in and of themselves. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Finally, these conclusions are valid only when the source language is English. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Definitions

• Translation: A noun in the source language is converted to the corresponding noun in the target language. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• Transliteration: A noun in the source language is phonetically transcribed into the alphabet of the target language, where the alphabet of the target language is not in the same family as that of the source. This may involve alteration to fit the target language's phonetic patterns. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• Transcription: A special case of transliteration where the target language shares an alphabetical family with the source language. This may involve a change in spelling. and may involve alteration to fit the target language's phonetic patterns. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

The Lemma

• Spanish translates common nouns, but not proper nouns or neologisms. Common nouns in locality names, but not in personal names, are translated. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• German in general translates nothing. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• French is bimodal. Some lemmas are translated according to the Spanish rule, while others according to the German rule. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• Italian and Dutch likewise are bimodal. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• Japanese, Russian, and Hebrew translate common nouns and transliterate proper nouns. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

In all cases, neologisms are either untranslated or transliterated. None are transcribed. (Yes, I know I'm dealing with a sample of one. Let's add a few more.) --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Usage

For the most part, the usage follows the lemma. The exceptions are: --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• German had two articles where the usage was sometimes untranslated, and sometimes translated. This could be ascribed to different editors. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• French had one article where the usage had the common noun translated, but the lemma untranslated. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

• Hebrew had one article where the usage was transliterated, but the lemma only partially transliterated. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Conclusions

For languages which share an alphabetical family, there appear to be two methods, one being the "Spanish" method, and the other being the "German" method. The former translates common nouns, but not proper nouns or neologisms. Common nouns in locality names, but not in personal names, are translated. The latter translates absolutely nothing, preferring instead to use the form of the source language. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

This seems a slight exaggeration of the German position: German does translate "Palomar Observatory", and (as you rightly said earlier) inconsistently translates a couple of other institutions in the text of the relevant articles. But, a very useful summary overall -- thanks! Andrew Dalby (disputatio) 23:48, 25 Martii 2010 (UTC)

For languages which do not share an alphabetical family, the method of choice appears to be to translate common nouns and to transliterate proper nouns. --Robert.Baruch 17:57, 25 Martii 2010 (UTC)

Interesting, thanks for the conclusions. Maybe the results have something to do with the source language being English? German, just like English, is a Germanic language, and a half of the populations of Germany and Austria know English, while Spanish is a Romance language and the knowledge of English is Spanish-speaking countries is lower. In France and Italy the knowledge of English also is lower, but not that low as in Spain or Mexico - hence their bimodality. Only Dutch does not quite fit this theory - a Germanic language with very many speakers knowing also English, yet it is bimodal. --Gabriel Svoboda 18:21, 25 Martii 2010 (UTC)

I haven't added to the table, but I can say that in languages of India, English names of institutions etc. are practically always transliterated, practically never translated. I think this supports Gabriel's interpretation relating to the knowledge of English among speakers of the language concerned. English is very widely known in India; English words like "University" are familiar to a large proportion of speakers and are therefore, I guess, quite acceptable in their (transliterated) English form. Andrew Dalby (disputatio) 23:57, 25 Martii 2010 (UTC)

I think I'd like to put up another table showing non-English sourced names and their en: counterparts. --Robert.Baruch 18:58, 26 Martii 2010 (UTC)

Now your es: examples show that the German language (or the Wikigerman one) really is an odd man out that translates nearly no lemmas, even though only 3.2 % of Germany and 3.3 % of Austria know some Spanish. The other languages usually translate. --Gabriel Svoboda 14:19, 27 Martii 2010 (UTC)