So, what can & can’t we put =na on?
What kinds of words and phrases allow the early creolized Chinuk Wawa “Yes/No Question” marker =na?
(Image credit: Scrip)
(Let’s take a second to explain a few things.
About my linguist symbols… “=” signals that na is what’s called a “clitic” form. That’s like a prefix or a suffix, but, instead of attaching onto a single word, we can show that na instead attaches to whole phrases.
Let’s also point out that =na is most commonly found in earlier Chinuk Wawa, in the lower Columbia River region. There are only traces of it in modern Grand Ronde CW, and hardly any sign of it in northern dialects.
Because =na creates a Yes/No question, it’s incompatible with “WH” (or “content”) questions such as ‘who’, ‘what’, ‘how’, ‘when’, and ‘where’. Oddly, though, there are known examples of such theoretically impossible combinations as kata=na for ‘how’ in old written Chinuk Wawa, but not in spontaneous fluent CW. So such chimeras appear to be Whites’ quirky use of the Jargon.
=na can only appear in a main clause, not in a subordinate clause.
To learn about the probable source of =na, see my post “=na ‘Yes/No Question’ from Proto-Salish *nə“.
Also, be advised, today’s post is only preliminary brainstorming about a can of worms…
Okay now, read on!)
I suspect that what’s going on is about (1) focus, (2) things that are able to be predicates, and (3) heaviness.
Using an example from Jack Stillman’s “Chipmunk” story as published in 1936 by Melville Jacobs:
Maika nanich…? ‘Did you see…?’
Because this is a yes/no question, some of the more southerly/earlier dialects would use the yes/no particle =na, added onto the end of the concept being asked about. So in this case, you’d then hear *Maika nanich-na…?* This is really just redundantly clarifying what was already a clear Jargon question; tone of voice alone, not to mention known context and gestures (both of which are important “classic Chinook” features), already make Maika nanich…? clearly a Y/N question about the entire inflected proposition of ‘you seeing (something)’.
But take note, if you were to say instead, *Maika-na nanich…?*, you’d be asking ‘Was it you who saw…?’ That’s a different sort of sentence, because it’s isolating just one component of the proposition, the Subject, to call it into question. By focusing on the Subject with =na, what you wind up asking for is whether there might have been some other Subject who did the seeing.
Taking off from that point:
You may be asking, can’t I isolate the Verb then? We saw that putting =na right after the verb in *Maika nanich-na…?* simply questions the entire phrase. Hmmm!?!?
Yes, there is a way of Yes/No questioning just the verb, and it’s done by putting the verb in the same position where we were able to Y/N query just the Subject — that is, up front!
That placement is what I analyze as Chinuk Wawa’s “Focus” function. It throws the spotlight over to some specific element of a sentence to make sure listeners pay attention to it & give answers as needed. So, you’d then say *Nanich-na maika…*, in the event that you needed to find out ‘Is it seeing that you did… (or was it some other activity)?’
Tiny but cool observation: because there’s just the one Focus position in a CW clauses, you’ll only find one occurrence of =na per clause. I haven’t, for instance, yet found any examples like *mayka mámá=na (or) mayka pápá=na yaka wáwa kákwa?* for ‘Was it your mom or your dad that said that?’
Small observation: I want to suggest that in the real world of human conversations, this focus-querying of a transitive verb would be a comparatively rare type of question. (By ‘real world’, I mean ‘outside of a courtroom or a linguistics classroom’!) 🙂
However, I really do think Y/N-querying of an intransitive or a stative verb (see here about the Fluid-S nature of CW) is common, so in a =na – using dialect, we’d hear a fair amount of questions like *kúri-na Flojo?* ‘Was Flojo running?’, and *chinúk-na mayka?* ‘Are you a Chinook Indian?’
Slightly more important observation: the Focus position isn’t best defined as being ‘first in the sentence’, but more precisely as ‘first in the verbal complex‘. I mean verbal complex as the verb + everything that refines its meaning — from Subjects & Direct Objects, to Negation (which comes before both the Subject & Predicate) & Adverbs (which can occupy various positions); maybe including Indirect Objects too.
Saying it another way: there is a little bit of stuff that has to come earlier in the unit we write as a sentence than the Focused piece, a trivial example being an exclamation ó ‘oh’ or a term of address like nayka shíksh ‘my friend’. You can’t question those little starting tidbits with =na, only the stuff that comes after them in the Focus placement.
But (yes, another but), by mentioning a phrase like nayka shíksh, I have a convenient chance to get into another feature of the behavior of =na. Let’s imagine we’re using this phrase in a different way: not as a term of address, but as a (stative) verbal complex in its own right. Then it would mean ‘It’s my friend’, you know, as a good answer to ‘What’s that way over there that I can’t quite make out?’ My experience of Chinuk Wawa is that that statement can then be questioned with =na. But I’m not entirely sure how! And this is because it contains an attributive part, something that describes the main stative-verb predicate shíksh ‘to be a friend’. So — partly because the use of =na is so rare in spoken CW in my lifetime — I’m not 100% sure about *Mayka shíksh-na?*; I think it’d mean ‘Is it (truly) your friend?’ But I also think it might be able to mean ‘Is it your friend (and not your sister)?’ I have even bigger doubts, but am not sure, about its ability to mean ‘Is it your friend (and not Mary’s)?’ Despite the presence of =na there, I feel forced to add special rising intonation onto the questioned word, to be sure I’m singling it out for Focus and get the answer I want. And when it comes to that, you just don’t need =na there in the first place!
If I give a comparable example, I think I can make more clear what I’m wrestling with there. Another phrase containing an attributive word, this time an Adjective, would be *tk’úp t’síkt’sik* ‘(to be a) white car’. I feel equally comfortable in Y/N-questioning this one, saying *tk’úp t’síkt’sik=na?* and understanding this as ‘Is it (truly) a white car?’ Yet I find it strange to try thinking of the =na as applying only to the last word, interpreting the expression as ‘Is it a white car (and not a white truck)?’ And again, it seems even harder to me to either take *tk’úp t’síkt’sik=na?* as ‘Is it a white car (and not a red one)?’, or, with that same intended meaning, to try =na on the dependent Adjective, *tk’úp=na t’síkt’sik?*. At best, my mind’s ear can hear this string of Jargon words as ‘Is the car white (or not)?’, which by the way has a different intonation pattern. (So it doesn’t need =na anyway!) That is, I could only put =na onto an attributive Adjective if I changed the starting meaning of *tk’úp t’síkt’sik* such that the Adj was a stative verb (‘to be white’), rather than ‘a white…’
George Gibbs 1863:18;
did =na blindly attach to the first word in the verb complex in his time?
Kind of makes you wonder if Chinuk Wawa always had a split between 2 kinds of Y/N question formation: =na for things that can be Focus-able / predicatable material, versus intonation for stuff that’s dependent on that material.
The short summary of what I just said: I’m doubtful that dependent items can be Y/N queried in CW. For my linguist friends: it appears that only those forms that are potential predicates can accept =na.
Here’s another specific thing that can take =na: the negation word wík. Any learner / speaker of modern Grand Ronde CW knows the word wíkna as a tag question meaning “isn’t it?” Surely this usage historically traces back to the general rule I’ve been laying out here. Wík can be a predicate by itself, making an entire single-word statement meaning ‘it’s not’. Therefore wík=na was totally okay back in the day, and was a common enough utterance to provide us with our present word wíkna. Take note though: other than this little word, you’ll hardly find =na at all in the Grand Ronde community now.
Believe it or not, everything that I’m running through today is just a brief outline of the properties of =na, which have not been investigated before by linguists. I fully expect that more will become known about the grammar of this little word.
For instance, as I hinted above, I think that “heaviness” — pretty much amounting to lengthiness — will also prove to be a criterion that prevents a lot of stuff from being Y/N-queried. Here, I’ll find a random long example sentence in the 2012 Grande Ronde dictionary…
…wík mayka tq’ix̣ íkta-shím kʰupa
‘…don’t desire any indiscretion with
other women [who aren’t your own wife].’
That’s a command, and I’ve added a mayka subject into it. (It can now be read either as a statement, or, still as a command.) Now, here’s my attempt at a list of things you can & can’t Y/N-query in it with =na, with Focus-fronting as needed…
- wík=na mayka tq’ix̣ íkta-shím kʰupa x̣lúyma łúchmən?
‘Don’t you want any/some indiscretion with other women?’
- mayka=na wík tq’ix̣ íkta-shím kʰupa x̣lúyma łúchmən?
‘Is it you that doesn’t want any indiscretion with other women?’
- *not OK* íkta=na-shím wík mayka tq’ix̣ kʰupa x̣lúyma łúchmən?
*’Is it (just) some (particular) kind of indiscretion that you don’t want with other women?’*
(Fails because it’s tries to query a dependent adjective and/or a ‘WHat’-type question word.)
- ?maybe? íkta-shím=na wík mayka tq’ix̣ kʰupa x̣lúyma łúchmən?
‘Is it (just) any indiscretion that you don’t want with other women?’
(Marginal, I think, because it tries to query a 2-word complex form and/or a ‘WHat’-type question word.)
- *not OK* kʰupa x̣lúyma=na łúchmən wík mayka tq’ix̣ íkta-shím?
*’Is it with (just) other women that you don’t want any indiscretion?’*
(Fails because it tries to query a dependent adjective.)
- ?not OK? kʰupa x̣lúyma łúchmən=na wík mayka tq’ix̣ íkta-shím?
‘Is it (just) with other women that you don’t want any indiscretion?’
(Fails, I think, because it tries to query an unwieldy 3-word-long element that’s also not clearly part of the verb core.)
Anyway, there’s a good deal more raw research to be done on =na, but I think we’ve been able to see a number of its behavioral traits today that haven’t been pointed out before.