LINGUIST List 12.1127

Tue Apr 24 2001

Review: Rice, Morpheme Order & Semantic Scope

Editor for this issue: Terence Langendoen <terrylinguistlist.org>

Directory

1. Sonya F Bird, Review of Rice, Morpheme order and semantic scope

Rice, Keren (2000) Morpheme order and semantic scope - Word formation in
the Athapaskan verb, Cambridge University Press, 453 pages.

Sonya Bird, University of Arizona


Synopsis

The goal of Keren Rice's book "Morpheme order and semantic scope - Word
formation in the Athapaskan verb" is to argue against the template as a
word formation device in Athapaskan verbs. She proposes instead that word
formation (at least in the case of verbs) is syntactic, and morpheme order
is determined by principles of semantic scope.

	The traditional approach to morpheme order within Athapaskan verbs
is templatic: verbs are associated with a template, and each morpheme
fills a slot of this template. Rice argues that while the templatic
approach can describe many facts of verbal morphology, it does little to
explain these facts. As an alternative, she proposes the Scope Hypothesis,
which not only describes but also explains the ordering of morphemes in
Athapaskan verbs.

	Rice focuses on two characteristics of Athapaskan
languages: "global uniformity" and "local variability". Global Uniformity
refers to features which are invariant across languages. Local Variability
refers to features which differ within and across languages. In reference
to morpheme ordering, Global Uniformity refers to cases in which morpheme
ordering is the same across all Athapaskan languages, and Local
Variability refers to cases in which morpheme ordering is variable, both
within and across languages. Through a detailed discussion of morpheme
ordering in numerous Athapaskan languages, Rice argues that scope
principles explain Global Uniformity, and that Local Variability arises
only when scope principles do not make clear predictions, or when they are
overridden by other principles of the language.

	According to Rice, Global Uniformity - and fixed morpheme order -
is a result of scope principles: the positions in the verb that are the
same across languages are those which are determined by scope
relations. For example, subjects universally have scope over objects, and
therefore consistently appear to their right. (Rice considers verbs as
syntactic constructions which are head last. The wider scope an element
has, the further right it is within the verb; if morpheme A has scope over
morpheme B, then A will appear to the right of B. As for the verb stem,
Rice assumes that it starts at the left of the verb, and moves to the
right at some point in the derivation.)

	Local variability - and free morpheme order - arises in 3
situations: 1) when scope is not involved, 2) when scope allows several
readings, or 3) when scope principles are overridden by other requirements
of the language (phonological, morphological, or syntactic). As an example
of the first situation, the ordering of iteratives and distributives is
variable, both within and across languages. This is explained by the fact
that these two morphemes are part of different systems: distributives are
part of the argument system, and iteratives are part of the event
system. Because they are part of different systems, they do not enter into
a scopal relationship, hence the free ordering.

Layout

The book is split into 6 parts. Part I presents the different approaches
to Athapaskan verbal morphology, and introduces the notions of Global
Uniformity and Local Variability.
	Part II offers a discussion of what Rice terms the lexical
items: preverbs, quantificational elements, incorporates, and other
lexical material. The ordering of these morphemes with respect to each
other is explained based on scope relationships.
	Part III focuses on what Rice terms the functional
items: pronominals, aspectual makers, and qualifiers. Again, the ordering
of these morphemes is explained based on scope relationships. Topics
discussed include an explanation of why 1st and 2nd person subject markers
are not in the same position as 3rd person subject markers, etc.
	Part IV offers of view of the lexicon. In this section, Rice
argues that it is possible to simplify the lexicon by leaving certain
morphemes underspecified in terms of their meaning, and having their
position in the verb (dependent on scope relations) determine their exact
role.
	Part V discusses the implications of the Scope Hypothesis in terms
of linguistic structure, and looks into various related topics: the nature
of the lexical entry, the distinction between inflection and derivation,
the domain of word formation, the role of scope in determining morpheme
order given other restrictions of the language, the consequences for the
Template model, and finally the consequences in terms of historical change
in Athapaskan languages.
	Part VI is an appendix providing additional information on the
Template model, and on morpheme ordering in languages considered
throughout the book. This section also includes a summary of constraints
on verbal morphology, and the differences between languages in terms of
ordering principles, interfaces, differences not related to scope, and
idiosyncrasies.

Critical evaluation

In her book, Rice does an excellent job of making the facts of Athapaskan
verbal morphology seem far more simple and predictable than I ever thought
they could be. The Scope Hypothesis as described and discussed is very
appealing from several points of view. First, it explains the structure of
Athapaskan verbs in a way that is clear and easy to grasp, which is quite
a feet considering its complexity. Second, it makes explicit the role of
each morpheme found in the verb. Third, it constitutes a useful tool for
language learning (at least for those with some linguistics
background). It provides a means of actually understanding morpheme order
based on intuitions about scope relations, rather than simply having to
memorize this order. In summary then, the Scope Hypothesis as presented by
Rice is extremely convincing. This is in part thanks to the way in which
the material is presented, and in part thanks to the thoroughness with
which she covers it.

A. Presentation

Rice's book reads like a mystery novel. New intriguing and seemingly
problematic facts are presented and explained by the Scope Hypothesis one
after another.

	Although the facts are often complex, the their presentation makes
them generally easy to grasp:

1) the facts are presented bit by bit, so that the reader is not
overwhelmed immediately with too much material.

2) the book contains a lot of cross-referencing, which makes it easy to
look back at previous, related, topics of discussion.

3) each section begins with an introduction to what will be discussed, and
ends with a summary of the conclusions drawn. There are also numerous
reviews and recapitulations within the book, which help the reader to keep
track of what has been covered.

	There are times when the facts get a bit confusing for the reader
with little background. For example, the details of the pronominal and
aspectual systems in Part III are sometimes hard to follow. However, this
is due to the complexity of the systems, not to any shortcomings in terms
of presentation.

	The only aspect of the presentation that could perhaps be improved
upon involves the data presented. First, in the morpheme-by-morpheme
translations, the verb stem is generally not translated, but is simply
labeled "verb stem". It would have been useful to have the translation of
the verb stem where possible, no matter how abstract it was. This would
have helped to see what meaning the various prefixes brought. Second, in
the discussion of verbal suffixes, a lot of the examples do not have overt
suffixes in them, which makes it hard to grasp some of the concepts
covered.

B. Content

	The main reason the Scope Hypothesis is so appealing is because
Rice is so thorough in her discussion of the facts:

1) numerous languages are discussed, and the properties of each are
explained.

2) all arguments are laid out clearly and explicitly.

3) examples used illustrate well the points made.

4) categories used for talking about morphemes are always justified, never
simply stipulated.

5) where data is not conclusive, or where questions remain, Rice is clear
about it.

	The nicest feature about Rice's book, in terms of content, is that
in arguing for the Scope Hypothesis, she also argues for - and makes
explicit - the specific role each morpheme plays in the verb. Rice
includes a discussion of which morphemes should be considered lexical
(part of the lexicon) and which are functional. She also renames certain
morphemes such that their name reflects more clearly their role in the
verb. For example, the morphemes traditionally called "conjugation
markers" and "classifiers" are referred to respectively as "situation
aspect markers" and "valence markers". Her terms provides much more
information on the function of these morphemes.

	In a few places, the nature of verbal elements could have been
made a bit more clear. For example, Rice mentions two distinct prefixes
which both have the inceptive meaning: the inceptive subsituation aspect
marker (functional) and a lexical prefix meaning "start point of an
event". There seems to be no reason, outside of position in the verb, that
these morphemes should be considered different in terms of functional
vs. lexical status. It would have been nice to have some external evidence
for this difference.

	The discussion of the valence markers (traditionally called
classifiers) is also somewhat unclear at times. For example, Rice argues
that in some cases, surface forms of verbs exhibit a combination of two
valence markers. However, in all of the examples given, it looks like only
the valence marker with widest scope surfaces. Why Rice argues for valence
markers combining, rather than for elision of the marker with narrowest
scope, is not clear from the examples given.

	Another point that is could have been expanded on a bit more
involves the various hierarchical relations between verbal morphemes. Rice
mentions several hierarchies which determine scope relations, and
consequently morpheme order:

subject > object (subject has scope over object and hence
appears to the right of object)
animate > inanimate
themes > oblique roles
referential pronouns > non-referential ones

Rice does not really tie together the different hierarchies. It would have
been useful to include a discussion of what unifies all of these
hierarchical relationships, i.e. why they all lead to the scope relations
that they lead to.

	While her coverage of the facts and arguments for the Scope
Hypothesis were extremely thorough, some more general issues were not
discussed:

1) the big picture, i.e. how all the elements of the verb come together.

2) related to this, how co-occurrence restrictions between morphemes can
be explained, i.e. whether the morphemes started out next to each other,
and what scope had to do with ordering in such cases.

3) what the Semantic Scope Hypothesis has to say about ordering in
languages in general? How does it translate to other languages with very
different morphological and syntactic properties?

	In general, Rice's book "Morpheme order and semantic scope - word
formation in the Athapaskan verb" presents a very interesting and
appealing perspective on Athapaskan verbal morphology, and is well worth
reading.


Biography

I am currently a 4th year graduate student in Linguistics at the
University of Arizona. I am working on timing issues in the sounds and
rhythmic structure of Lheidli, a dialect of Dakelh (Carrier) - an
Athabaskan language spoken in the northern interior of British Columbia.