relative roots
Conor McDonough Quinn
quinn at FAS.HARVARD.EDU
Thu Apr 20 14:52:56 UTC 2006
Dia dhaoibh, a chairde!
> Hi Monica - I can't remember if I mentioned this when we talked about
> it before, but I personally like to gloss relative roots with an X.
> So aeN- would be 'in X manner', other relative roots would be 'X
> fast', 'X long', 'X many times', etc. This indicates clearly to me
> that the relative root denotes a variable that needs to be filled in
> by an expression elsewhere in the sentence. Not sure if that works
> for all audiences, but anyway it's another option to consider. I
> think both 'thus' and 'in that way' are very misleading glosses.
I've also been doing this "X"-based approach in my morpheme-by-morpheme
glosses of Penobscot, because it has two advantages:
(a)
"Xmanner" (or "X_manner", etc.) takes up less space than "in...manner".
Intelligible brevity is key for glossings of massively polymorphemeic
words.
(b)
The X is very saliently "unusual": glossings like "thus", if they are
indeed understood by the user, are read as ordinary lexical items, which
relative roots (as far as we see them now) are not. So use of "X", as
off-putting as it can be, signals to the reader "this is more than just
what's familiar" and suggests there's some pattern going on that they
should look up.
This said, I agree with Becky that a majority of laypersons don't find
appeals to the metaphors of algebra (i.e. X as a variable) much more
accessible or less intimidating than "thus", but given the choice, "X" at
least does some things that "thus" doesn't.
Perhaps we could just use brackets, to indicate the fill-in-the-blank
aspect of their use, i.e. replace "Xmanner" with "[]manner". Kind of back
to "in...manner", but still more concise, and, like the "X" approach,
signals gently to the reader that
[]manner
[]point
[]from
[]extent
are part of a unified pattern that they might want to find out about. On
the one hand, David's right: anyone who's unwilling to deal with the
grammar of relative roots in some form or another just isn't going to be
helped much at all by any Algonquian dictionary (that's assuming they are
using it to learn to speak the language like native speakers, which is of
course not the only or necessarily even the majority use of these kinds of
dictionaries).
On the other hand, since we should make dictionaries intelligible to the
more casual reader, I think the "[]' or comparably unobtrusive
consistent-marking-of-relative-root-status-without-using-the-technical-term
approach is pretty reasonable, since people will at least get the idea
that something about manners, points, sources, extents, etc. is involved.
Does this sound like a happy medium?
Till later, keep well.
Sla/n,
bhur gcara
> My first reaction was that using the X would be way too mathmatical-
> looking, but it does get across the fact that something needs to be
> filled. Has anyone considered this or done it? I think it's
> something I would definitely need to get speaker/learner input on
> before doing!
>
> - Monica
>
>
> On Apr 19, 2006, at 6:22 PM, Wayne Leman wrote:
>
> > I agree with Rich on this. And I prefer using an English definite
> > as I gloss relative roots in our Cheyenne dictionary. Early in my
> > work on Cheyenne I used the Bloomfieldian "thus" but after awhile I
> > realized it didn't make much sense to many English speakers, not to
> > mention Cheyenne speakers for whom we want the dictionary to be
> > user-friendly.
> >
> > Hence:
> >
> > Enêhe?eve 'he did it that way'
> > Ehevoo'o 'that's what he said'
> >
> > Wayne Leman
> > Monica,
> > There are a couple of considerations with relative roots.
> > As I have been arguing for about ten years now (and no one seems to
> > notice), relative roots have properties that suggest they are
> > analogous to head marking. What I mean is the following.
> >
> > We translate simple transitive verb forms with something
> > that suggests the object slot which must be filled.
> >
> > waabamaad vta 'see s.o./s.t. (an.)'
> >
> > Since relative roots analogously license clausal complements, we
> > should gloss them analogously with appropriate indefinites:
> >
> > inaabid vai 'look in a certain direction'
> > inaabamaad vta 'see s.o./s.t. (an.) looking like
> > s.t.'
> > apatood vai 'run along a certain route'
> > onjinawaad vta 'kill s.o. for a certain reason'
> >
> > But with 20/20 hindsight, because the head markings all have null
> > definite readings, it would probably be more accurate to gloss them
> > with definites.
> >
> > waabamaad vta 'see him/her/it (an.)'
> >
> > and
> >
> > inaabid vai 'look in that direction'
> > inaabamaad vta 'see him/her/it (an.) looking like
> > that'
> > apatood vai 'run along that route'
> > onjinawaad vta 'kill him/her/it for that reason'
> >
> > I used the indefinite option in my dictionary, but I'm thinking
> > that if I had it to do over again, I'd go with definites.
> >
> > Rich
> >
> >
> >
> >
> >
> > At 6:42 PM -0500 4/18/06, Monica Macaulay wrote:
> >> I got this very helpful message from David Costa and since it just
> >> came to me am taking the liberty of forwarding it to the list. I
> >> think the list is set up so that replies just go to the sender and
> >> not the list, which is silly. I'll check into changing that.
> >>
> >> - Monica
> >>
> >> Begin forwarded message:
> >>
> >>> From: David Costa <pankihtamwa at earthlink.net>
> >>> Date: April 18, 2006 3:06:48 PM CDT
> >>> To: Monica Macaulay <mmacaula at WISC.EDU>
> >>> Cc: Daryl Baldwin <baldwidw at muohio.edu>
> >>> Subject: Re: relative roots
> >>>
> >>> Monica:
> >>>
> >>>> We're currently going through the archaic English words that
> >>>> Bloomfield used
> >>>> in his Menominee lexicon and trying to come up with more
> >>>> colloquial
> >>>> defintions. While thinking about 'thus' and what we could
> >>>> replace it with, I
> >>>> realized that there's an intersecting problem, which is due to
> >>>> the fact that
> >>>> all of the verbs that have 'thus' in their definition - not
> >>>> surprisingly -
> >>>> have the relative root aeN- in them. We were going to change
> >>>> 'thus' to 'in
> >>>> that manner' but it occurs to me that that might be interpreted
> >>>> as a
> >>>> complete definition. So, take the verb that Bloomfield
> >>>> translates as 'it
> >>>> glows thus' - we could change it to 'it glows in that manner'
> >>>> but a
> >>>> dictionary user might not realize that it's a verb that needs a
> >>>> manner adverb
> >>>> - and that using it without one would actually be ungrammatical
> >>>> to a native
> >>>> speaker. Conversely they might not realize how to translate it
> >>>> in a
> >>>> sentence; i.e. if you used this verb with 'brightly' the
> >>>> meaning would be 'it
> >>>> glows brightly' - NOT 'it glows brightly in that manner' or
> >>>> something like
> >>>> that. Have any of you wrestled with this one and come up with
> >>>> a good
> >>>> solution?
> >>>>
> >>> Well, it seems to me that the 'thus'/'in that manner' dilemma and
> >>> the worry
> >>> about people thinking the gloss is a complete definition are
> >>> separate
> >>> issues. In our Miami dictionary, we used 'thus' a lot, but I
> >>> think that was
> >>> just because it's all over the Algonquian literature that way and
> >>> we're so
> >>> used to it. Perhaps in retrospect '(in) that way' or 'so' might
> >>> have been a
> >>> bit more user-friendly since 'thus' is such a marginal word in
> >>> modern spoken
> >>> English.
> >>>
> >>>> A related issue of course is how much info one puts into a
> >>>> dictionary without
> >>>> crossing over the line into being a grammar. I think we
> >>>> probably are all
> >>>> making somewhat different decisions about where to draw that
> >>>> line, and I
> >>>> haven't decided yet where it would be drawn in a case like this.
> >>>>
> >>> And this is the second issue! I think the problem of speakers not
> >>> knowing
> >>> exactly how to use a word grammatically just based on its dictionary
> >>> definition is just unavoidable. In my opinion, at the most one
> >>> could write
> >>> 'relative root' in the gloss along with the form class, then in
> >>> the intro
> >>> refer the user to a grammatical sketch somewhere; or one could
> >>> explain in
> >>> the intro that when a word has that prefix and 'thus' (or 'in
> >>> that manner',
> >>> or whatever) in its gloss, here's what it means, and see the
> >>> grammatical
> >>> sketch. Explaining the details of how to use a relative root ninety
> >>> different times in a dictionary would just drive people crazy,
> >>> and they'd
> >>> just have to refer to the grammar anyway.
> >>>
> >>> I've encountered people (not Miamis!) who want Native American
> >>> languages to
> >>> be spelled just like English, so that they supposedly won't have
> >>> to learn
> >>> any pronunciation rules. When one learns any new language, one
> >>> has to master
> >>> that language's spelling and pronunciation idiosyncracies, and
> >>> one does not
> >>> have the right to expect the rules to be the same as English.
> >>> Grammar is
> >>> the same way -- I've also had people (again, not Miamis) ask
> >>> "can't we learn
> >>> this language without any grammar?" The answer is no, of course --
> >>> Algonquian grammar is SO different from English grammar, anyone
> >>> who wants to
> >>> make meaningful use of an Algonquian dictionary is going to have to
> >>> familiarize themselves with a certain amount of grammar. Using a
> >>> dictionary
> >>> of Spanish or Polish or Swahili would be the same way. And you can't
> >>> make grammar totally transparent in a dictionary.
> >>>
> >>> Anyway, I hope these comments are useful.
> >>>
> >>> Dave
> >>>
> >>>
> >>>
> >>>
> >>>
> >>
> >> Monica Macaulay
> >> Department of Linguistics
> >> University of Wisconsin
> >> 1168 Van Hise Hall; 1220 Linden Drive
> >> Madison, WI 53706
> >> phone (608) 262-2292; fax (608) 265-3193
> >> http://ling.wisc.edu/~macaulay/monica.html
> >
> >
> > --
> > ******************************************************************
> >
> > Richard A. Rhodes
> > Department of Linguistics
> > University of California
> > Berkeley, CA 94720-2650
> > Voice (510) 643-7325
> > FAX (510) 643-5688
> >
> > ******************************************************************
> >
>
> Monica Macaulay
> Department of Linguistics
> University of Wisconsin
> 1168 Van Hise Hall; 1220 Linden Drive
> Madison, WI 53706
> phone (608) 262-2292; fax (608) 265-3193
> http://ling.wisc.edu/~macaulay/monica.html
>
>
>
>
More information about the Algonqdict
mailing list