I was reading a thread about Translating Spork and people recommended kulerforko. To my ears and tongue, forkkulero sounds much nicer. It was mentioned that "forkulero is technically either for-kulero (away-spoon) or fork-ul-er-o (a component fragment of a fork person)." This is true, but relies on the presumption that forkulero is the same as or will be mistaken for forkkulero.

This got me thinking about the status of doubled consonants (or so-called gemination) in Esperanto. I am more or less a beginner and would like to know how this is viewed among most Esperantists. I have read that this does not happen within roots, but may often happen in compounds (much like English). However, people seem afraid of doubled consonants as in the above example. To me, knowing that r's are difficult for many people (including myself, when they are smashed close together as in the sequence -krur- that seems to spring up a lot in EO), while gemination is relatively common, I would opt for the latter.

Of course the third (Zamenhofian) option would be to tack an -o on: forkokulero. This seems to me the clearest. Is there any argument (besides length) against everyone using this approach (as against doubled consonants)? Any possible instances of confusion?

On the other hand, I could see people who come from languages such as Arabic or Italian (or even Japanese or Danish) where gemination is highly distinctive (used to distinguish between words), not having the same aversion, so they might be more disposed to naturally construct Esperanto words this way. As such it might be useful to try and be mindful of listening for them. Pali, another very agglutinative language (like Esperanto) that I know a little about makes copious use of gemination: buddha, dhamma, etc. This is usually not an a problem once you get used to it; it is actually one of the less perplexing aspects of its pronunciation.

Is this mostly just a bugaboo for English (and other non-geminate-distinctive language) speakers? Is there an extant forum post or web page about this topic? I haven't been able to find much.

Some authoritative sources (including, as I recall, PAG - I'll need to check) say that, if you find it hard to clearly pronounce a repeated consonant (e.g. ekkrii, allasi, forridi, Finnlando), it is permissible to just produce one longer consonant.

Strictly speaking though, each and every letter should be pronounced distinctly, including doubled consonants and consonant clusters - Rule 9 of the Fundamento explicitly states: Every word is to be read exactly as written, there are no silent letters.

So this seems to be somewhat different from gemination, which deals with length of individual consonants only, if I'm not mistaken? So Esperanto doesn't strictly have gemination - though many speakers will use it rather than reproduce the same consonant twice.

The only common doubled consonant I ever have trouble with is kk; the voiced ones tend to flow together anyway, and some consonants like cx, hx never repeat except in some very infrequent compounds (even then, you can insert a vowel in between; see below).

Of course the third (Zamenhofian) option would be to tack an -o on: forkokulero. This seems to me the clearest. Is there any argument (besides length) against everyone using this approach (as against doubled consonants)? Any possible instances of confusion?

There is nothing wrong with adding the vowel ending for substantive roots - in fact, in some compounds would be exceedingly difficult to pronounce without them, e.g. lingvoscienco. In the written form there is no possibility for confusion because no space separates the parts; in spoken language it is a bit iffier, but normally you would insert a pause if you intended two different words: Sur la tablon, apud la forko, kulero estas metita.

seveer:
Of course the third (Zamenhofian) option would be to tack an -o on: forkokulero. This seems to me the clearest. Is there any argument (besides length) against everyone using this approach (as against doubled consonants)?

That only works if the first component actually takes an -o. That wouldn't work with (for example) the word "superrigardo" (in view on this forum page if you're using Esperanto as the forum language).

Tempodivalse:[...]

Thanks for the reply; I more or less knew about all that. My question was really aimed more at tendencies of word-formation and style, not merely pronunciation. A good example is the apparent flux of spellings for Buddha: Buddo, Budao, Budho. And this is a root, not even a compound. From the little Pali that I know I would tend toward Buddo, and indeed that is the version found in Zamenhof (according to PIV). However it seems that Budho has won out and most of the derivative words come from that. This and examples like the one in the OP intimates to me that there may be some aversion, somewhere, to doubled consonants (in roots at least).

Links to posts/essays/grammars would be useful, but I would also, perhaps moreso, appreciate anecdotal views. Do people avoid creating doubled consonants in word formation? If yes/no what is your native language? Have you noticed this tendency in others and if yes/no, what is their language background?

I would also briefly say that while Zamenhof gives us a pronunciation guide, I don't believe (I could be wrong?) he ever presented it in a scientific phonetic system (like the International Phonetic Alphabet) so to some degree it is open to debate what "pronounced separately" should mean. Nowhere is it commanded that I put a glottal stop between i and o in io to distinguish from ijo, if I can make it clear by including/excluding an emphatic palatal approximant, for instance, that should also be acceptable. I would say that making a consonant "longer" falls into this category. An unvoiced plosive like kk becomes "long" in an entirely different way from ll or nn or rr. These will be more or less difficult depending on your linguistic background.

Bruso:
That only works if the first component actually takes an -o. That wouldn't work with (for example) the word "superrigardo" (in view on this forum page if you're using Esperanto as the forum language).

Sorry, I didn't mean to limit it to -o exclusively (that was just the context). Obviously -a can be used similarly. Would superarigardo be incorrect for some reason? If so, can you cite it?

seveer:

Bruso:
That only works if the first component actually takes an -o. That wouldn't work with (for example) the word "superrigardo" (in view on this forum page if you're using Esperanto as the forum language).

Sorry, I didn't mean to limit it to -o exclusively (that was just the context). Obviously -a can be used similarly. Would superarigardo be incorrect for some reason? If so, can you cite it?

The first component is the word "super". It's not a shortened version of supero or supera.

Would superarigardo be incorrect for some reason?

Yes, because super is a root that, by default, does not have a vowel ending (like most prepositions). I would parse superarigardo as just supera rigardo, which is not the same thing.

Similarly, you cannot insert a vowel after prefixes like ek-, mal- - those have to be attached directly, as becomes apparent from a perusal of the Ekzercaro (and all subsequent precedent).

Do people avoid creating doubled consonants in word formation?

It varies - some people add (where possible) a vowel to compounds more often than others, and sometimes it just sounds ugly without it - e.g. akvofalo.

to some degree it is open to debate what "pronounced separately" should mean.

Fair enough. There has been significant discussion (including in the pronunciation section of PAG, as I recall) on what Rule 9 means, and debate on phonology more generally.

What I meant to say, though, was that the simplest interpretation of Rule 9 is that you repeat the consonant. Given that many people cannot easily distinguish geminated consonants, this has the advantage of being clearer than gemination.

However, this doesn't apply in all cases - I've never heard anyone (well, except myself) say Finnlando with all consonants in the cluster fully enunciated.

This and examples like the one in the OP intimates to me that there may be some aversion, somewhere, to doubled consonants (in roots at least).

There is a vague tendency not to double consonants in new (imported) roots, yes. For example, the old root finn/ has lately come into competition with suom/.

But in general, doubled consonants within a root are rare, because Esperanto imports words from other languages phonetically, not orthographically (so, "parallel" is paralel/ with 2 Ls, not three, because in the source languages you only pronounce two).

Tempodivalse:Yes, because super is a root that, by default, does not have a vowel ending (like most prepositions). I would parse superarigardo as just supera rigardo, which is not the same thing.

Sorry, I'm a bit confused, so bear with me here. At issue here seems to be whether super can be replaced by supera and preserve the meaning, right? You seem to posit that either the prepositional nature or the peculiar lexical difference in this case changes the meaning. If there is something special about the behavior of prepositions as a whole in this regard (compounding), I would appreciate it if you could point me to it. My understanding is that they are morphemes like everything else (in Esperanto) that just happen to have a couple of special super-powers (like cordoning prepositional phrases), but I don't know how this directly affects their use in compounds.

More importantly in this particular case, if I understand correctly, is the peculiar meaning of this particular set of words. Based on my reading of the entires in the PIV there is a contradiction here. Sure, the definitions for supera seem to indicate a distinction, which might be somewhat oversimplified as "literal versus figurative position of superiority" but then when you look at superi (which is fundamenta) you find "Troviĝi super, esti supera." Maybe I'm way off base here, but this indicates to me that if the verb superi means both (under a single entry, therefore synonymously?) to be found super and to be supera, then supera is sufficiently synonymous to be able to use it in this manner in the compound, regardless of the earlier entries that conspicuously exclude this meaning for supera. If this seems incorrect, could you please explain?

I also have to balk a little bit at making a fine distinction between a preposition and the meaning of its adjectival form here, because prepositions are so notoriously vague. Part of the point of prepositions in language seems to be that they are quasi-magical forms that can take on a bunch of idiosyncratic meanings. I would argue that the over in overview (superrigardo) is different from the over in over the rainbow, overwhelmed, worked-over, and overextended. It seems like supera intuitively means something more akin to "superior" but this divergence in meaning is recent even in English and it is wholly correct to say that a ceiling is superior to a floor. Injecting this distinction into superrigardo seems idiomatic. If it is nothing but an idiom, fine, but I get the impression we are arguing around that. I guess I'd like clarification on whether we are talking about a mere convention or some more substantial pattern/rule.

I look forward to your response and would like to reiterate that I am just trying to get a clearer understanding here, not trying to challenge the validity of your statements.

Thanks

seveer:Maybe I'm way off base here, but this indicates to me that if the verb superi means both (under a single entry, therefore synonymously?) to be found super and to be supera, then supera is sufficiently synonymous to be able to use it in this manner in the compound, regardless of the earlier entries that conspicuously exclude this meaning for supera. If this seems incorrect, could you please explain?

There's a passage in PMEG that seems to confirm what Tempodivalse said on this, if it is of any interest:

PMEG:Se la antaŭelemento estas vorteto (tia vorto, kiu ne bezonas finaĵon), oni normale ne uzas ligan finaĵon. Je bezono oni tamen povas uzi ligan E-finaĵon: postsigno → postesigno (faciligas la elparolon), postulo → posteulo (faciligas la komprenon). Tio tamen okazas nur tre malofte.

Aliaj finaĵoj ol E povas aperi post vorteta antaŭelemento nur kiam tio aldonas necesan signifon: unuaeco = “la eco esti unua” (unueco = “la eco esti kiel unu”), antaŭeniri = “iri antaŭen” (antaŭiri = “iri antaŭ io”). La ligon EN (E + N) oni uzas iafoje ankaŭ post radikaj antaŭelementoj: supreniro, ĉieleniro (aŭ ĉieliro), hejmenvojaĝo (aŭ hejmvojaĝo).

Translated: If the first element in a compound is a 'vorteto' (a word which does not require an ending), one does not normally use a linking vowel. If needs be however one can use a linking -e. This is very rare though. Other endings besides -e can appear after a 'vorteto' only when it adds necessary clarity of meaning.

So assuming you accept PMEG's prescription here - which you really should because it's in line with normal language usage - the question would be does the -a in "superarigardo" add necessary meaning. I would say that it does not. You could even say that it adds wrong meaning (as per Tempodivalse's comment), but that's another matter.