Fwd: Nahuatl word classes

Galen Brokaw brokaw at buffalo.edu
Thu Jan 3 04:53:40 UTC 2013

It is only misleading if you assume that formal properties are always 
consistent indicators of usage. But that isn't the case. There are 
always irregularities in language usage that are not predictable based 
merely on formal criteria. So you always have to explain the 
peculiarities of usage in all languages. And based on the examples you 
gave, it didn't sound like "kwalli" was used in a way that is exactly 
the same as property words like chichiltic, iztac, etc. Do these other 
property words take the -tin ending in agreement with the nouns that 
they describe? I would also say that in Classical Nahuatl, at least, I'm 
pretty sure that you do actually find "ipan cualli" meaning "in 
goodness." And "cualli" appears as an imbedded noun in verbs such as 
"ninocualtoca" [I consider myself good]. But even if this is not the 
case in a particular dialect of Nahuatl, what you claim are differences 
in the properties of "kwalli" in relation to other nouns is explainable 
as a difference in usage rather than a real difference in properties. It 
seems to me that this is a valid distinction to make. For example, even 
in English, some nouns don't take plural endings. And "kwalli" is a good 
example. We wouldn't say "goodnesses" either, but that doesn't mean that 
"goodness" is not a noun. Based merely on formal rules, you can produce 
"goodnesses," and it is clear what is being done. But it doesn't make 
any sense in our culturally determined conceptual paradigm, so we don't 
produce that form. So, again, the fact that a word doesn't exhibit all 
of the same behaviors as other words of the same class doesn't really 
justify creating a new class for it or putting it in a different class. 
So for that reason, the argument that "kwalli" is a stative rather than 
a noun based on formal properties doesn't make sense to me. Even if you 
give weight to the fact that in Hueyapan Nahuatl, "kwalli" does not 
exhibit all of the properties evident in other nouns, you would still 
have to demonstrate that it exhibits formal properties of some other 
class that other nouns do not.
The classification of "kwalli" as a stative is not based solely on the 
formal argument. It also has to be based on the semantic determination 
that it conveys a stative meaning. And your stative category is based on 
semantics rather than form. "kwalli" is certainly stative insofar as it 
refers to a persistent condition, but all nouns do that. There is 
certainly a qualitative distinction that can be drawn between the 
condition of being "kwalli" and the condition of being "tlakatl," but 
I'm not sure that this is necessarily a universal difference (i.e., that 
speakers of all languages would necessarily conceptualize "kwalli" and 
"tlakatl" as qualitatively different) or that it is all that significant 
for the determination of formal word classes.
It may be true that the "stative" is a useful and necessary word class 
in Nahuatl. I don't know. I'm only commenting on "kwalli," and I'm not 
convinced that "kwalli" justifies such a word class for the reasons that 
I've explained. And what you are saying doesn't respond to those 
reasons. It sounds like you want the syntactical categories to do all 
the work of accounting for usage. But in order to do that, you have to 
abandon formal criteria in favor of semantic criteria. And I don't see 
the advantage or the logic in doing that. To me it would make more sense 
to use "stative" as a sub-class rather than a class of its own. Thus you 
could have stative verbs like "chichiltic" and perhaps stative nouns 
like "kwalli." That way you wouldn't have to abandon formal criteria for 
determining the larger word classes. And the semantics of the word, 
cultural logic, conventions of usage, and so forth would account for the 
variations and anomalies in usage. Yes, that would have to be explained, 
but it has to be explained anyway.

On 1/2/2013 5:41 PM, Magnus Pharao Hansen wrote:
> Many languages use only verbs to describe nouns, and depending on the 
> exact details of analysis and the definition of the word class 
> adjectives they may or may not be said to have adjectives. If for 
> example those verbs that describe properties of nouns behave 
> syntactically different from other verbs, then it can be said that the 
> language has adjectives, but that adjectives are a subclass of nouns. 
> Or it can be said that the language has no adjectives, and that the 
> adjectival function is fulfilled by verbs. This decision depends on 
> theoretical concerns about what how one understand and describes 
> grammar and different linguists do either of the two - some linguists 
> do consider the adjectival function to be universal, others don't.
> The reason I think it doesn't make sense to just say that kwalli is a 
> noun is that it shares virtually none of the properties of other 
> nouns, except the fact that it can make predicates. To say that it is 
> a noun is misleading because in order to explain the words usage we'd 
> have to explain that it is a special noun that is used in a way that 
> is exactly the same as property words like hueyi, iztac, chichiltic, 
> etc, and entirely unlike all other nouns. Therefore I think it makes 
> more sense to say that in Nahuatl there is a class of words that have 
> the primary function of creating stative predicates and that some of 
> those words are deived from nouns and others from verbs.
> My Nahuatl wordclasses are as follows:
> Predicators
> -Verbs
> -Nouns
> -Statives
> Non-Predicators
> -Particles
> On 2 January 2013 13:51, Galen Brokaw <brokaw at buffalo.edu 
> <mailto:brokaw at buffalo.edu>> wrote:
>     Hey Magnus,
>     I have no problems with the morphological criteria for classifying
>     words in Nahuatl. So no problems there. But...
>     I mispoke/miswrote in my example. I didn't mean to say that the
>     language would use verbs exclusively to describe nouns, which I
>     know is what I wrote. I meant that in describing nouns, the
>     language used verbs rather than something else. In other words,
>     the hypothetical language doesn't have adjectives; it uses verbs
>     rather than adjectives to convey the qualities of nouns in
>     addition to the other things that verbs do. Of course, the idea is
>     informed by the way Nahuatl uses verb forms in contexts in which
>     English and Spanish would use adjectives. The point was that in
>     such a case it is misleading to talk about verbs fulfilling an
>     adjectival function as if "adjective" were some kind of universal
>     category. I know you weren't doing this with "kwalli," but it
>     seems to me a similar problem. You seem to want to attribute
>     verb-like qualities to a noun based on what seems to me the
>     pragmatic usage of nouns to convey an idea that could be more
>     fully elaborated into a more "complete" grammatical utterance. I
>     don't see the formal basis for making this argument, and if I
>     understand the issue (which I admit, I may not), then pragmatics
>     provides an explanation for the positive evidence that you give of
>     how "kwalli" actually does behave. The only formal basis for this
>     argument is the negative evidence that "kwalli" doesn't do all the
>     things other nouns do.
>     The fact that a particular word like "kwalli" doesn't exhibit ALL
>     of the same morphological behavior as other nouns doesn't
>     necessarily mean that it doesn't belong to the same category.
>     There are certain expressions in English, and I assume all
>     languages, that while formally grammatical, are not acceptable
>     based on usage because they are semantically strange or illogical
>     given the way the word is culturally defined. So just because you
>     couldn't or wouldn't say it, doesn't mean that it is not formally
>     grammatical. So when you say that you "cannot say 'nokwal,'" I
>     take your word for it. But does this mean that it is merely an
>     expression that is not used for semantic, cultural, and/or logical
>     reasons or does it mean that it violates the in/formal, abstract
>     rules implicit in the speakers' mental grammar. If I understand
>     correctly, all the examples that you give in both posts are
>     negative ones: they are things that "kwalli" doesn't do that other
>     nouns do. But what about what it does? Doesn't the positive
>     evidence that you cite (i.e., what "kwalli" actually does
>     formally) conform to the expectations of nouns? Does it do
>     anything in formal terms that other nouns don't? If "kwalli"
>     behaves formally in ways that other nouns don't, then that kind of
>     evidence might be a more compelling basis for qualifying its
>     classification or for classifying it differently. But if there is
>     no doubt about the fact that "cualli" is a noun historically and
>     it doesn't behave formally in ways that other nouns don't (by
>     which I mean what it actually does rather than what it doesn't
>     do), then I don't see on what basis you would need to call it
>     something other than a noun. So if this is the case, I don't see
>     why the perfect solution for "kwalli" wouldn't be to just call it
>     a noun.
>     Galen
>     On 1/2/2013 12:42 PM, Magnus Pharao Hansen wrote:
>         Hi Galen
>         The form/function issue comes from the fact that Nahuatl is
>         polysynthetic, which means that most of its syntax is carried
>         out by means of morphology, and that syntactic functions are
>         marked through morphology. I.e. when I argue for the use of
>         formal morphological criteria for determining word class
>         membership in nahuatl it is because they are the best
>         diagnostic tools for determining the syntactic functions of
>         words. I.e. a word that takes nominal morphology also tends to
>         have particular syntactic functions.
>         I don't understand your example of a language that uses verbs
>         exclusively to describe nouns - I don't think such a language
>         is possible since verbs are generally understood to be words
>         that have the primary  function of forming predicates. Any
>         wordclass that has the primary function of describing nouns
>         would be a class of adjectives not verbs - unless there is
>         another reason to consider them a subclass of verbs.
>         Regarding kwalli I don't have a perfect solution, it is
>         obviously historically and morphologically a noun - no doubt
>         about that. But you cannot in contemporary Nahuatl say
>         "nokwal" or "kwalko" or "ipan kwalli"  or pluralize it to
>         kwalmeh or kwaltin (except in agreement with the noun it
>         modifies). You also never use kwalli as the argument of a verb
>         "se kwalli" or "in kwalli" (unless followed by a head noun
>         that is being modified by kwalli). In all of these ways it
>         doesn't behave like a noun. Yes it is true that all nouns in
>         Nahuatl can form predicates, like titlacatl, but for most
>         nouns this is not the primary use, for kwalli it is.  I am
>         currently thinking that kwalli, hueyi, chichiltik and other of
>         these words are best characterized as statives, that form
>         intransitive predicates.
>         You can read a more complete exposition of my thoughts on
>         Nahuatl word classes here:
>         http://kuscholarworks.ku.edu/dspace/bitstream/1808/8101/1/KWPL-32-PharaoHansen2.pdf
>         best regards,
>         M
>         On 2 January 2013 11:38, Galen Brokaw <brokaw at buffalo.edu
>         <mailto:brokaw at buffalo.edu> <mailto:brokaw at buffalo.edu
>         <mailto:brokaw at buffalo.edu>>> wrote:
>             Hi Magnus,
>             I have a question about "cualli" too. Like John, I thought
>         your
>             original point was that grammatical categories should be
>         based on
>             form and formal behavior rather than semantic function. And I
>             wholeheartedly agreed with that. I had the same concern,
>         and we
>             have had discussions on the list about this issue before.
>         Now that
>             I've gone back and reread your original response, I see
>         that you
>             do refer to syntactical function, but you then immediately
>             explain, and advocate for, the classification of words in
>         Nahuatl
>             based on morphological form. So I guess I thought you were
>             defining syntactical function in formal terms. So I'm not sure
>             exactly how syntactical function differs from form and
>         whether or
>             not you can divorce it from form or from semantics either, and
>             that seems to cause problems. I agreed with what I
>         understood you
>             to be saying, that grammatical categories should be based
>         on their
>             formal properties and behavior rather than their semantic
>             function. I think this is so in part because "semantic
>         function"
>             always has formal implications that are not necessarily
>         universal.
>             For example, if a particular language uses verbs
>         exclusively to
>             describe nouns, then it is difficult for us to talk about the
>             semantics of this phenomenon without recourse to the
>         grammatical
>             category "adjective." But this does not mean that such
>         verbs have
>             an adjectival function. In the hypothetical language to
>         which I am
>             referring, such expressions are verbal, so they have a verbal
>             function. They only have an adjectival function in relation to
>             languages that have adjectives.
>             This is only indirectly related to my question about
>         "cualli." I
>             think the issue here is a little different. So to get back
>         to the
>             case of "cualli," and at the risk of revealing my profound
>             ignorance, can I ask how the usage of "kwalli" is
>         different from
>             other nouns? Maybe I don't understand your examples, but don't
>             other nouns work the same way? You can say "tehwah
>         titlakati" too,
>             right? But this doesn't mean that "tlakati" works as a
>         verb. If
>             that were the case, then wouldn't all predicate
>         nominatives and
>             direct objects function like verbs when the verb is
>         omitted? It
>             sounds like you are saying something similar to the idea that
>             because a verb is not necessary, therefore nouns work like
>         verbs
>             in Nahuatl. But just because the verb can be omitted
>         doesn't mean
>             that the noun takes on a verbal function. It seems to me
>         that this
>             confuses pragmatics with formal categories and structures of
>             syntax. Maybe I am misunderstanding, but it sounds like
>         you are
>             hesitant to classify it as a regular noun, not because it
>         behaves
>             irregularly from a morphological or formal perspective,
>         but rather
>             merely because it is commonly used without a verb. But it
>         seems to
>             me that such usage has more to do with pragmatic function
>         than it
>             does syntactic function: the fact that Nahuatl doesn't need an
>             actual verb in a complete and acceptable utterance (which
>         could be
>             expressed as a complete grammatical sentence with a verb). Of
>             course, I think this is true of all languages.
>             Galen
>             On 1/1/2013 10:49 PM, Magnus Pharao Hansen wrote:
>                 Dear John and listeros
>                 Thanks for the explanations.
>                 some responses:
>                 2. Ok, so the -x is the remnant of the /yi/ ending
>         (this means
>                 that in La
>                 Huasteca the phonological forms are underlyingly
>         /tokayitl/
>                 and /ma:yitl/).
>                 This would not be recognized by speakers of central
>         dialects.
>                 4. I am not giving an account of how these words are
>         formed,
>                 they are
>                 clearly fromed from verbs and nouns. But they function
>         like
>                 property words
>                 that form stative predicates.
>                 5. kwalli works as a verb in that its primary syntactic
>                 function is to form
>                 predicates "kwalli inon" 'tehwah tikwalli" etc. And it
>         is not
>                 very nouny
>                 ()although obviously it originated as a noun because it
>                 neither accepts
>                 plural or possessive morphology, and hardly ever
>         occurs as the
>                 argument of
>                 a verb as nouns prototypically do.
>                 best,
>                 M
>                 _______________________________________________
>                 Nahuatl mailing list
>         Nahuatl at lists.famsi.org <mailto:Nahuatl at lists.famsi.org>
>         <mailto:Nahuatl at lists.famsi.org <mailto:Nahuatl at lists.famsi.org>>
>         http://www.famsi.org/mailman/listinfo/nahuatl
>         -- 
>         Magnus Pharao Hansen
>         PhD. student
>         Department of Anthropology
>         Brown University
>         128 Hope St.
>         Providence, RI 02906
>         _magnus_pharao_hansen at brown.edu
>         <mailto:magnus_pharao_hansen at brown.edu>
>         <mailto:magnus_pharao_hansen at brown.edu
>         <mailto:magnus_pharao_hansen at brown.edu>>_
>         US: 001 401 651 8413 <tel:401%20651%208413>
> -- 
> Magnus Pharao Hansen
> PhD. student
> Department of Anthropology
> Brown University
> 128 Hope St.
> Providence, RI 02906
> _magnus_pharao_hansen at brown.edu <mailto:magnus_pharao_hansen at brown.edu>_
> US: 001 401 651 8413

Nahuatl mailing list
Nahuatl at lists.famsi.org

More information about the Nahuat-l mailing list