You've hit the nail on the head. The biggest hurdle for using Quenya and Sindarin as anything remotely resembling a real language is the severe lack of vocabulary. We have only a few thousands words, while most real-world speakers have vocabularies of 100,000 or more.

Unfortunately, there isn't a clear way to add new vocabulary to the language either. We can invent neologisms using existing roots, but that's only going to get us so far. At some point we are going to need to coin entirely new words off of brand new roots, but I've never heard anyone suggest a good approach for doing that.

+Paul Strack Well, a hundred thousand words may be an exaggeration, many languages can describe the world in 3-5 thousand native lexemes. The problem is not in the limited vocabulary – the real problem is in its lacunas.
BTW, I've found a word for "sugar" in Eldamo. It's a Gnomish glist. I think it may be converted into Quenya like listo or similar, being associated both with lissë and asto.

Oh lovely, sugar = "sweet dust", I'd forgotten all about that. G. glist and glissith both still fit Sindarin phonology, and the root LIS still survives in Tolkien's latter writing.

The second element "dust" = G thith, whose root SISI (actually ÞIÞI) did not survive, but I think the closest latter analog is N/S. list "ash, sand" < LITH. So perhaps primitive lissi-lithsē > Q. lissitse, S glissith "sugar, (lit.) sweet sand" with a bit of ancient haplology to lose the second li.

Also for "kitten" we can probably a diminutive for "cat", perhaps miuincë (pronounced miuyince?).

+Paul Strack (1) I gave the word "sugar" for illustrative purposes, not for necessary discussion of this particular vocabulary entry; (2) In my opinion, listo is a perfect portmanteau of lissë and asto 'dust', the later root √AS is attested in Ety and probably substitutes QL √ÞIÞI; (3) the list is much, much longer…

Sure, Q. listo could be derived from a primitive portmanteau g-lisi-asto > g-listō which would also produce S. glist. I do think LITH is a more likely successor to QL SISI/ÞIÞI than AS(AT), at least for the sense “dust, sand”

I think the example illustrates the problem, though. It’s hard enough to get Elvish students to agree on how to use the vocabulary we already have, much less adding new vocabulary.

the same discussion in FB:
facebook.com - Необхідна перевірка безпеки

+Paul Strack agree on all points.
+Helge Kåre Fauskanger used to say that we need to take twelve or twenty best Quenya scholars, shut them in a closed room and don't let them out till them come to a consensus.
I want this group and that one in FB to become a kind of this room.

I get where you are coming from, but I’ve got 4 or 5 more years of research and data analysis to do before I can make anything close to “definitive” statements (at least definitive for me).

+Paul Strack ok, I see. Your Eldamo database is a great tool and I'll be happy to see it working and expanding.
Nevertheless, I'm going to do some preliminary work for Ilaurëa Coirëa Quenya, meanwhile. It won't hurt anyway.

I very much encourage you to do that work. The assembly of a working Quenya/Sindarin language will be a lengthy, iterative process. Everything I am doing builds on prior work, and I fully hope others will build on what I am doing.

I think when we do this work, extend the vocabulary, we should be conscious that no matter with how much care, love, and respect we approach the creation of neologisms there will always be people for whom it will be a sacrilege.
I also think we should set up a hierarchy of methods of neologisms, and treat it almost like a declaration, a sort of license that is appended to every website that hosts such dictionaries and downloadable wordlists. Something like this:

1. attested words and phrases
2. derivations from attested words using attested pre/suffixes and compounding
3. kennings, phrases
4. derivations from attested roots following attested derivation- and phonetic patterns
5. adoption of early words that fit later phonetic rules
6. adaptation of early words to fit later phonetic rules
7. creating cognates from other Eldarin languages
8. loanwords from IRL languages

Of course the order is debatable, and I may have forgotten something that should be included, but you get the gist.

+Ицхак Пензев as you are active in both social media groups, I would be very very grateful if you could let me/us know if there has been an interesting idea or discussion that is worth sharing here.

I know that the words you listed were just for illustration, but just for fun:

towel: some compound with lanne "cloth"; *parkalanne perhaps, or *málanne (cf. Polish ręcznik from ręka "hand")
toothpaste: we only need to figure out the paste bit, the roots MASAG and KHIM may help
sugar: I like your solution with *listo
pepper: no idea; maybe a loan as it is a migratory word in most IRL languages
rice: idem
cabbage: probably follow the IRL etymology and derive it from "head"
kitten: as Paul says, diminutive of cat
hamster: I like it that in Lithuanian it's called "ground squirrel" or maybe model it on pekkuvo and make it mean "mouth-hide"
turn the lights on: narta i kalma/r, kalta i kalma/r
fry: there is an EQ verb sisi- to look at
come for a visit: in both Latin and in Hungarian it is a frequentative of the verb 'see, behold, inspect'; *kekenna- ?

A parallel idea brainstorming & discussion is going on at https://www.facebook.com/groups/quenyachat/permalink/1556248121111056/ , by the way.

More translation ideas and neoformations, some of which in the very least provide a handy counterbalance for one in deciding how to place his own stones across the river.

+Tamas Ferencz karpakuvo for hamster...karpa meaning the entire mouth?

+Tamas Ferencz I like your ideas, though I would prioritize them differently. An interesting alternative, though, would be to allow some sort of "rating system" for neologisms, perhaps on a site like Parf Edhellen. People could vote for neologisms they like and dislike, which could be converted into a community rating system, and used as a search filter (include only words rated at 60% or better).

Also we have EQ kalar for lettuce, which is from karase which seems to come from an abstract noun from KAS/KAR head.

+James Coish yeah, something like that

+Paul Strack that's a very interesting idea, to have a rating system

+Paul Strack how would you prioritize them? Just out of interest.

+Tamas Ferencz I would prioritize "modernized early forms" over new derivations from late roots. This is for two reasons: (1) I prefer to use forms that Tolkien used at some point over inventing my own and (2) I find that new derivations from late roots often pay "fast and loose" with the meanings of those roots, especially since such primitive formations were among the least stable elements of Tolkien's languages.

The only caveat is that it must be possible to justify the modernized early form within the phonetic and etymological system of Tolkien's later languages, otherwise it can't be used. In other words, when "creating new derivations from attested roots", I think you should first look for early forms that can be adapted into the later system before inventing entirely new ones.

Taken this way, many of these early adaptations would probably also qualify as "derivations from attested roots", except that they are based on words Tolkien actually used at some point rather than invented out of thin air.

This is, however, a personal preference. I know others prefer to avoid early forms if at all possible.

+Paul Strack thank you. I could accept that.

One of the big long-term goals of Eldamo is figuring out how many of these early Qenya and Gnomish forms can safely be adopted or adapted into the later forms of the languages.

+Tamas Ferencz I agree with these...makes sense.

+Paul Strack I agree.

I would love to see how you adopt the Qenya and Gnomish forms...I sometimes have my own ideas, but as you are more skilled, I would take your opinion (and Tamas') over mine.

+Paul Strack I would be open to implement a rating functionality on Parf Edhellen. I imagine it would also be a useful tool for cleaning the dictionary of outdated definitions.

+Leonard W. It wouldn’t need to be complicated. I would suggest a simple “thumbs up” or “thumbs down” vote for “I would use this” or “I wouldn’t use this”. You could then assign individual users a “weight” based on their level of skill or amount of participation.

Then rating = total (up votes * weights) / (all votes * weights).

For words already in the corpus, they could be pre-assigned a certain number of up votes and down votes. Say attested Sindarin words get 100 up votes, attested Noldorin words get 75 up voted and 25 down votes and attested Gnomish words 50 up and 50 down (or something like that).

Anyway, just some thoughts.

Sabbath is over, I am back here. All the ideas you exposed above, are great. I like the concept of hierarchy of methods very much; and indeed, much the same as Paul, I'd give more weight to the early material. It is better to adapt the attested forms, especially taking into attention that we have enough precedents when a root or a word from 1920s pops up in the drafts from 1960s.
I don't know how the rating system can be technically implemented, but the idea is attractive, too.

The list is open for editing here: https://docs.google.com/document/d/1xeENHk1q3tJzUGalq6k-XjjgUwCbIVIhIMtAYZN9ywQ/edit?usp=sharing

For pink, I use carnince "reddish" modelied after luinince "bluish". Malwe as "pale" means yellowish.

Why the adaptation of EQ. haru >> NQ. haro? A final -u is perfectly legitimate in late Quenya (e.g. heru “lord”). I think haru is best left unchanged.

+Paul Strack​ I agree 100% about haru, but please comment it in the document – Lukasz can't see your comment since he is not a member of this group at G+

+James Coish see above

+Ицхак Пензев could you please use Google Spreadsheets instead? It would make it substantially easier to import when we’ve agreed on a vocabulary.

People, I had made a spreadsheet for "My First Dictionary", but then decided it was too early to open it for discussion. I'm going to collect the potential entries first, and then present it to you all.

+Paul Strack I will begin implementation as soon as I have finished the emailing notification service. I will be relocating to Canada on the 3rd of January, so I expect it will take some time before I will have the functionality ready for production. I will let you know.

As always, you can monitor the development on Github.

+Leonard W. Wow, that was quick. I meant it as an offhand suggestion. But if you can get it done soon it could be quite interesting.

Well, the kitten is found! It is titsë (PE21:20,27).