Particles of Uncertainty in Quenya

Particles of Uncertainty in Quenya


A couple of months ago @Paul Strack posted a short but neat analysis of the attested Quenya particles of uncertainty, their possible shades of meaning and use. Since Discord is a platform we cannot link to, I am copying the text here, for further discussion, and to preserve it for posterity:

ma = interrogative/indefinite (probably true)

qui = neutral hypothetical (may or may not be true)

ai/ce = strong hypothetical (probably not true)

au = strongest hypothetical (not true but we are entertaining the idea, past tense only)

la = weak negative (not true, no other connotation)

ū = strong negative (very not true, either impossible, the opposite or with evil sense) 

The challenge here is that when Tolkien introduced ce it was a neutral hypothetical, so a lot of its attested uses align with qui. He only decided it was a strong hypothetical pretty late, where it may or may not have replaced ai 

So we see qe >> ce >> qui for the neutral hypothetical (with a brief flirtation with it(e)/te) and the strong hypothetical ai/au >> (maybe?) 


Submitted by Lokyt Thu, 11/21/2019 - 12:11

As for the discussion on Discord (June the 30th), some of the conclusions there seem not right to me.

At any rate, Q. ai- , qui- and íqua- are indeed "any-"  which is an indispensable quantifier, but one quite different from "some-": in terms of the propositional logic, "any-", "every-" and "-ever" are universal quantifiers (stating that the proposition goes for all entities in consideration), while "some-" is an existential quantifier (the proposition goes for at least one of all the entities).

Now, the mere unspecified mo/quen and ma/qua could potentially work as "someone" and "something": "I need someone" might (with some tolerance) be the same as "I need a person", "something is wrong" might be "a thing is wrong".
But I doubt the same could be said about lúme for "some time", núme for "somewhere" etc.

And you know what? I think the solution might be hidden in námo ("person, someone", but literally something like "is-person, the being one"). This ná- can obviously serve as a prefix stating the very existence of (at least one instance of) what follows – which is exactly what an existential quantifier should do.
So I would tentatively propose forms like *náma "something", *nálume (*nallume? *nálúme? *nalume?) "at some time", *nanome (*nánome? *nánóme?) or *nasse (*násse?) "somewhere"...

Submitted by Lokyt Thu, 11/21/2019 - 20:17

In reply to by Atwe

Yeah, that kind of undermines what I've said, doesn't it :)

Well, in certain contexts, the difference becomes practically nullified ("is someone home?" vs. "is anyone home?"). I believe Tolkien's glosses aim at these situations.
But the distinction is otherwise clear nevertheless, cf. for example "the entwives could be somewhere" vs. "the entwives could be anywhere".

Submitted by Atwe Thu, 11/21/2019 - 14:44

To me the most "vexing" part of all this is that ma is "something" and the interrogating particle at the same time; if it weren't like that, we could have **mana as an adjective "some, some kind of" etc. As in Gnomish.

Submitted by Lokyt Thu, 11/21/2019 - 19:13

In reply to by Atwe

Yeah, it's sort of a pity that Tolkien didn't keep the simple consequent system from the late 1910s (indefinite MA vs. interrogative DA).

BTW, the virtual identity between the interrogative and indefinite morphemes goes all the way back to 1923, when there already was Q. -ma "some" vs. the interrogative particle mai :)
And I definitely cannot exclude the possibility that these m- formants are one and the same in their origin (even in the middle & late periods), despite the development into two different functions.

Submitted by Lokyt Thu, 11/21/2019 - 19:31

In reply to by Atwe

Yep, two IE. roots quite similar in their functions to elvish MIN and ER.

And since you've started with the real-world languages... Slavic has two forms of "some-", one is actually "unknown-" (like "not-known-who", "not-known-where" etc.) and the other is the imperative of "to be" (like "who-be-it", "where-be-it"). The latter reminiscent of Q. námo.
While Latin has even more ways of expressing the existential quantifier, the chief ones being "other" (ali-), "there" (-dam) and negated negative words (nonnullus "not-noone" = "someone"). I admire the last one, as it foollows the propositional logic to a tee :)

Submitted by Atwe Fri, 11/22/2019 - 09:42

In reply to by Lokyt

To which category does the Polish suffix belong? (ktoś, coś, jakiś, któryś etc.)


Hungarian has something similar to nonnullus, there's the colloquial phrase nem semmi "not nothing" which is used as an appreciative interjection: Az nem semmi! "Now that's something!".

Submitted by Paul Strack Fri, 11/22/2019 - 04:50

It occurred to me that if ma can serve as both an interrogative and an indefinite, the same might be true of other question words. So perhaps masse can be both “where” and “somewhere” depending on context.

masse i Elda mare? “where does the Elf live?”

i Elda mare masse “the Elf lives somewhere”

That would leave ai the option of serving as an even more indefinite “any”

One the other hand “every” would be il-

Submitted by Lokyt Fri, 11/22/2019 - 14:05

In reply to by Paul Strack

Some Slavic languages indeed do exactly that - but only in interrogative and multipropositional clauses:
"Was somebody here?" is "was here who?" (while  "who was here?" is "who was here?").
"He has something to eat" (two propositions, having and eating) is "he has what to eat".
But "he ate something" (just one proposinion, eating) requires specifically indefinite pronoun ("something") - the reason being that the unemphatic word order requires the subject to go first and there would be no difference between "who is here?" and "someone is here" if "who" and "someone" alone were not different from each other.

And the last IMHO goes for Q. as well. (I.e. a specific marker of indefiniteness would be needed even though - unless you want to push the subject to an unnatural position only to save yourself an indefinite prefix...)

But to be fair, it's probably true that the distinction would be required only for stuff that can become a syntactic subject. Otherwise, ma- would be enough (expressing interrogativeness when opening the clause × indefiniteness when being positioned anywhere further towards the end).

Submitted by Atwe Mon, 03/30/2020 - 13:34

Another thought that has occurred to me that in some languages (English, Polish that I know of, I am sure others as well) demonstratives — which are inherently definite, aren't they  — can express indefiniteness: I was walking to the pub last night and saw this man on the pavement dressed as a parrot.

Not sure it adds anything meaningful to the discourse, it's just an observation.