LINGUIST List 16.3451

Sat Dec 03 2005

Review: Morphology/Typology: Dressler et al. (2005)

Editor for this issue: Megan Zdrojkowski <>


        1.    Peter Arkadiev, Morphology and its Demarcations

Message 1: Morphology and its Demarcations
Date: 01-Dec-2005
From: Peter Arkadiev <>
Subject: Morphology and its Demarcations

EDITORS: Dressler, Wolfgang U.; Kastovsky, Dieter; Pfeiffer, Oskar
E.; Rainer, Franz
TITLE: Morphology and its Demarcations
SUBTITLE: Selected Papers from the 11th Morphology Meeting,
Vienna, Feb. 2004
SERIES: Current Issues in Linguistic Theory 264
PUBLISHER: John Benjamins
YEAR: 2005
Announced at

Peter M. Arkadiev, Institute of Slavic Studies, Russian Academy of
Sciences, Moscow

This volume is a collection of 18 papers presented at the 11th
Morphology Meeting in Vienna, whose main topic was ''Demarcation
issues in morphology: derivation vs. inflection, compounding vs.
derivation''. Only papers explicitly dealing with the aforementioned
topic were included in the volume. Most articles deal with data from a
particular language, but many are typologically oriented in some
sense or in other. The languages cited in the volume come not only
from Europe, but also from North America, Africa, and Australia.


''Wichita word formation. Syntactic morphology'' by David S. Rood
argues that in Wichita, nearly extinct Caddoan language of central
Oklahoma, there are morphological phenomena which do not fit in the
allegedly exhaustive typology 'inflection vs. derivation vs.
compounding', and which he proposes to call 'syntactic morphology'
due to their direct relevance for the phrasal syntax of the language in
question. In Wichita, a highly polysynthetic language where many
meanings which are in more familiar languages expressed by free
words and phrases are encoded morphologically on the verb, the verb
has special morphemes whose function, according to Rood, is not
related to the verb stem itself, but rather to some constituent external
to the verb. This is illustrated by the distribution of the morpheme re:R-
which appears in constructions meaning 'I don't know ...' and as a
marker similar to definite article (or even, in some contexts, to
anaphoric pronoun). Rood concludes that Wichita, which lacks
configurational phrase structure, uses instead special kind of
morphemes whose main function is 'to hold together the parts of an

''Morphology in the wrong place. A survey of preposed enclitics'' by
Michael Cysouw is a comprehensive survey of a rare phenomenon,
the so called 'ditropic clitics', which are defined by the following
characteristics: (i) their host is not characterizable neither in lexical,
nor in syntactic terms; (ii) the constituent on the other side of the clitic,
on the contrary, is easily definable and is functionally related to the
clitic. In the initial sections of his paper Cysouw discusses various
theoretical and terminological issues bearing on the topic, an
especially notes that many previous studies denied the very existence
of 'ditropic' clitics. Then he turns to the examples of such clitics, which
belong to several types: (i) cross-referencing pronominal clitics (Kugu-
Nganhcara, Djinang, Kherwarian, Udi, Northern Talysh); (ii) clause-
linking clitics (Ingush, Northern Mansi); (iii) noun phrase internal clitics
(Kwakwala, Yagua, Greek). The main conclusion Cysouw draws from
these examples is that 'ditropic' clitics really exist, and, although rare,
must be accounted for by any adequate theory of morphology. Then
he considers several possible ways of explaining this phenomenon,
both synchronic and diachronic, and notes that in many cases such
clitics attach to some pragmatically salient or focused constituent of
the clause.

Jasmina Milicevic in ''Clitic or affixes? On the morphological status of
the future-tense markers in Serbian'' analyses the so called 'analytic'
and 'synthetic' future forms of Serbian from the standpoint of the
Meaning-Text theory (Mel'cuk 1993-2000). The formative cu that
marks future exhibits ambiguous behaviour, but Milicevic quite
convincingly argues that by most relevant criteria (that is,
morphological, lexical and syntactic) it is a clitic, although from the
point of view of morphophonemics it shows some affixal traits.

''The demarcation of morphology and syntax. A diachronic perspective
on particle verbs'' by Corrien Blom discusses the so called 'separable
complex verbs' (SCVs) in Dutch, which consist of a verb and of a
preverbal element corresponding to a postposition. As is well known,
such verbs in Dutch and its close relative German in some contexts
split up, cf. the following examples:

a. dat Jan de boeken opzoekt
that John the books up-searches
'that John looks up the books'
b. Jan zoekt de boeken op.
John searches the books up
'John looks up the books.'

Blom discusses the arguments for treating SCVs as special separable
words and shows that all the properties which purportedly indicate
their wordhood (ability to serve as inputs to compounding and
derivation; valency change induced by the particle; conventionalized
meaning) are actually shared by uncontroversial syntactic phrases.
Blom argues that SCVs are best regarded as a special case
of 'compositional idioms' or 'idiomatically combining expressions' in
terms of Nunberg et al. (1994), that is those phrases whose meaning
is conventionalized but nevertheless compositionally derived from the
meanings of their parts (e.g. 'pull strings'). Such expressions, as
against pure idioms (e.g. 'kick the bucket') allow certain syntactic and
lexical freedom of their parts. Finally, Blom observes that from the
point of view of their diachronic development, Dutch SVCs are in the
middle of the grammaticalization cline from purely syntactic phrases
into morphologized prefixal verbs, while the degree of their
lexicalization may vary from purely compositional and semantically
transparent lexemes to fully idiomaticized verbs.

''When clitics become affixes, where do they come to rest? A case
from Spanish'' by Andrés Enrique-Arias discusses a well-known
problem of whether the order of affixes reflects the original order of
their diachronic sources viz. free forms or clitics. The paper focuses
on the evolution of object markers in Spanish from the 13th century to
present, and is based on a quantified corpus study. In modern
Spanish the position of object markers with respect to the verbal stem
is fixed: they are prefixed to finite verb forms and suffixed to non-finite
and imperative forms. In Old Spanish, however, the position of object
clitics was determined by syntactic factors, mainly by the syntactic
position of the verb, and there was quite a number of cases where
variation in clitic placement was observed. The author then presents a
statistical survey of a selection of Old Spanish texts and shows how
various factors which played a role in clitic placement in Old Spanish
gradually subsided in favour of a more rigid morphological pattern. Of
special interest are such factors as stress placement and 'parallel
processing effect': prefixation of object clitic was particularly favored in
those contexts where its suffixal position would have resulted in a non-
canonical prosodic contour of the word, and there was a strong
tendency to identically align clitics and verbs in sequences of clauses.
The author concludes that object marker placement in Modern
Spanish really reflects the typical order of clitics w.r.t. hosts in Old

''Grammatical hybrids. Between serialization, compounding and
derivation in !Xun (North Khoisan)'' by Bernd Heine and Christa König
is based on the data from !Xun, a Khoisan language spoken in
Southern Angola, northern and northeastern Namibia. This language
has very little affixal morphology, but there is a number of serial verb
constructions where otherwise lexical verbs lose their lexical meaning
and become grammaticalized as markers of various grammatical
meanings. The authors show, however, that despite their
polyfunctionality, these items do not exhibit systematic
morphosyntactic variability, and that whether a given occurrence of a
verb is used as a lexical item or rather as a grammatical marker is
determined mainly by semantic (e.g., the distinction between result
verbs and manner verbs) and pragmatic factors. The authors
conclude that in !Xun there is no clear-cut boundary between lexicon
and grammar, as well as between serialization, compounding and
derivation, since the items in question exhibit certain properties of all
these processes. Such a situation is not unique to the Khoisan
languages, and is attested in the languages of Southeast Asia, e.g.
Chinese and Vietnamese. The authors suggest to model these
phenomena in terms of 'grammaticalization chains', i.e. typologically
attested paths of diachronic development from lexical to grammatical
meanings (see e.g. Bybee et al. 1994, Heine, Kuteva 2002), and
consider the situation in !Xun to be exceptional in the following
respect: usually, semantic and morphological grammaticalization (that
is, cliticization and affixation) go hand in hand, but here the latter
component of the process somehow happened to not operate.

''The borderline between derivation and compounding'' by Laurie
Bauer discusses those quite widespread phenomena (the data comes
mainly from English, Danish, and French) which cannot be
unambiguously classified as either derivation or compounding proper.
First of all, Bauer notes that diachronically derivational morphology
more often then not arises from compounding, and discusses some
borderline cases, which can be characterized by high frequency of
certain elements of compounds and the gradual loss of their semantic
relationship with their lexical sources. Cases of the opposite
development (from derivation to compounding) are, on the contrary,
quite rare and unsystematic. Other problematic cases
include 'synthetic compounds', whose second members (at least in
Danish) do not usually occur as independent words, and in any case
do not allow an uncontroversial analysis, unique morphs (such as
English 'cran-' or 'rasp-' in 'cranberry' and 'raspberry'), 'splinters' such
as 'burger' or '-nomics', neo-classical compounds, and finally
prefixation in French. What characterizes all these rather discrepant
phenomena is their being unstable diachronically, failing to maintain
their status as independent or bound elements.

Geert Booij in ''Compounding and derivation. Evidence for
Construction Morphology'' discusses the data more or less similar to
that of Bauer's article (i.e., such 'borderline cases' as prefixation in
French and Dutch, or semigrammaticalized 'affixoids' like English -
way, -wise or -like), but reaches completely different conclusions.
Booij tries to show that, contrary to Anderson's (1992) claims, internal
morphological structure of both compounds and derived words is
visible to morphological and phonological/prosodic processes, and
thus concludes that there is no such crucial difference between
compounding and derivation that would require handling them with
different formal tools. Then he outlines a theoretical framework which
could capture structural similarities between compounds and affixal
derivatives, viz. the so called 'Construction Morphology', expanding on
the ideas of Construction Grammar, originally proposed as a non-
transformational syntactic framework (Fillmore 1988, Goldberg 1995,
Kay 1997). Construction Morphology regards complex words of all
types as instantiations of morphological patterns of various levels of
generality. The most general patterns are the following, where the
variables x, y stand for phonological strings, and the variables X, Y for
lexical categories (N, V, A):
a. compounding: [[x]X[y]Y]Y
b. suffixation:[[x]X y]Y
c. prefixation: [x [y]Y]Y

These patterns may be associated with certain meanings, and filled by
particular morphological material. Thus, the general compounding
schema is assigned the semantics 'Y with some relation to X', while
the suffixation pattern may be instantiated by, e.g., the following
derivational model:
[[x]Ver]N 'one who Vs'

This last pattern may be further instantiated by a particular complex
word, e.g., 'baker' or 'worker'. This possibility of levels of
representation intermediate between the most abstract patterns and
concrete words is crucial of Construction Morphology, and it can
equally apply to both derivational and compounding schemas. Booij
discusses how various types of affixally derived and compound
formations may be handled by the framework he proposes, dealing
especially with the mechanisms of default inheritance between
different levels of abstraction and of pattern unification, paying
attention also to semantic relations between parts of complex words.
The framework Booij proposes seems to be simple and attractive,
allowing to capture interesting generalizations about both formal and
semantic properties of complex words.

''Selection in compounding and derivation'' by Sergio Scalise,
Antonietta Bisetto and Emiliano Guevara is aimed to show that the
borderline between derivation and compounding lies not only in the
realm of well known formal differences between these two types of
word formation, but also in the (more or less semantically based)
mechanism of selection of the non-head constituent of a complex word
by its head. The authors argue that both in compounding and
derivation there must be a process of 'head-selection', which operates
on the basis of the Lexical-Conceptual Structure (LCS) of both head
and potential non-head of the word. Following Jackendoff (1990) and
Lieber (2003), Scalise et al. consider LCS of a word or a morpheme to
consist of levels: a 'skeleton' containing such grammatically relevant
information as syntactic category and event/argument structure, and
a 'body' of encyclopedic features. The comparison of various types of
derivational and compounding processes reveals the following
systematic differences in head-selection in derivation and
(a) The selection operated by a derivational affix is fixed and constant,
whereas for compounding selection is less strict: what is usually
required is only that the non-head matches at least some information
contained in the head's LCS;
(b) Selection in derivation is less 'syntax-like' than in compounding,
e.g., the non-head in an affixally derived word does not satisfy any of
the head's arguments, while in compounding this is often the case;
(c) Derived words are more predictable than compounds from both
quantitative and semantic point of view -- it is often extra-linguistic
information and pragmatics that play crucial role in the determining of
the possible range of interpretations a compound may have, as well
as its well-formedness, while for the derivational affix the determining
factor is only whether the non-head satisfies the head's rigid
selectional restrictions.

In ''Compounding and affixation. Any difference?'' Pavol Stekauer
claims that from the point of view of his 'Cognitive-Onomasiological
Model' of grammar there is no difference between compound
formation and affixal derivation, since both involve the same
conceptually driven processes of assigning form to conceptual
structures. Also, from his standpoint, there is no difference between
affixal morphemes and lexical units, because both are available to the
process of naming and word-formation. Stekauer also proposes to
redefine the notion 'head of the word', considering it to be semantic
rather than formal. However, this article does not say anything about
how the 'Cognitive-Onomasiological Model' would account for various
purely morphological differences between (at least some types of)
compounds and derivatives.

''On a semantically grounded difference between derivation and
compounding'' by Bernard Fradin argues on the basis of a detailed
analysis of French deverbal agentive derivatives in '-eur' and V-N
compounds that there is an important semantic difference between
affixally derived words and compounds, quite similar to what Scalise et
al. would call difference in selectional properties. While affixal
derivation imposes rigid constraints both on its input and its output,
compounding requires only that the parts of a compound combine in
such a way that best fits some semantic scenario, e.g. a 'causal
structure' (Croft 1991). Fradin also argues that although compounding
involves combining two lexemes, it does not result in a syntactic
structure, since V-N compounds (contrary to the claims found e.g. in
Di Sciullo, Williams 1987) do not form a VP structure. If they did,
compounds such as ''marche-pied'' 'step' (lit. 'walk-foot'), where the N
corresponds to anything but the verb's internal argument, would be

Dany Amiot's article ''Between compounding and derivation. Elements
of word-formation corresponding to prepositions'' is an in-detail study
of French complex words whose initial formative may be used as a
preposition: ''après'' 'after', ''avant'' 'before', ''contre'' 'against', ''sans''
'without' etc. Amiot, using several formal and semantic criteria, shows
that this class of formatives is not homogenous, and that some of its
members (e.g. ''sur-'' and ''contre-'') are real prefixes exhibiting such
properties as endocentricity, ability to combine with words of different
syntactic categories, semantic independence from corresponding
prepositions, while others (such as ''sans'' and ''avant'') show a weaker
degree of grammaticalization.

''Cumulative exponence involving derivation: Some patterns for an
uncommon phenomena'' by Davide Ricca presents some putative
examples of cumulative exponence of derivational categories, and,
more importantly, of derivation and inflection. The former was claimed
(e.g. by Anderson (1992)) to be very rare, and the latter to be non-
existent. All possible examples of these rare phenomena are thus of
great relevance for the problem of 'splitting' the morphology into two
separate modules, viz. derivation and inflection. The data Ricca
discusses is taken mainly from Romance languages, especially Italian,
both literary and dialectal. Instances of cumulation in derivation usually
involve gender or evaluative categories fused with nominalizations of
different kinds, whereas cumulation of derivation with inflection most
notably involves number. Ricca concludes that scarcity of cumulation
in derivation can be related to a weaker paradigmatic structure of
derivational processes where it is hard to establish clear-cut
categories. On the other hand, it is possible to outline some diachronic
sources leading to cumulation of inflection and derivation, viz. (i)
phonological fusion across the derivation-inflection boundary; (ii)
reanalysis of a productive derivational process, already coded
cumulatively, into an inflectional one; (iii) grammaticalization starting
from suppletive lexemes.

Maria-Rosa Lloret in ''Revising the phonological motivation for splitting
morphology'' discusses some peculiar morphophonological facts from
a Cushitic language Wellega Oromo and from Majorcan dialect of
Catalan. In both languages a phonotactically driven process of vowel
epenthesis happens to crucially depend on such aspects of
morphology as nominal vs. verbal domain and inflection vs. derivation.
Lloret argues that previous accounts of these facts in terms of cyclic
phonological rules or underlying allomorphy are inadequate, and
proposes an alternative treatment in terms of a correspondence
surface-oriented Optimality Theory, using output-output
correspondence constraints (see McCarthy 1995, Kager 1999) and
Optimal Paradigms model (McCarthy 2005). This analysis provides
support for the claim that surface paradigmatic relations among
wordforms play an important role on the organization of morphology.

''Derivation versus inflection in three inflecting languages'' by Stela
Manova deals with phenomena on the borderline between inflection
and derivation in three Slavic languages -- Bulgarian, Russian and
Serbo-Croatian. The processes Manova investigates are
diminutivization, 'Movierung' (formation of nouns denoting females
from those denoting males) and imperfectivization of verbs. Manova
discusses various morphological, lexical and semantic properties of
these processes and evaluates them against the common criteria for
distinguishing between inflection and derivation, and concludes that
according to them the phenomena in question do not show great
discrepancy, more or less tending to the derivational pole of the
continuum. However, according to a novel criterion proposed by
Manova, i.e. that of inflection class assignment (in short, 'if a category
can be identified either with a particular inflectional class or with
complementary inflection classes, it represents (non-prototypical)
inflection'), she assigns gender formation and imperfectivization in
these languages to inflection.

Sergey Say in ''Antipassive sja-verbs in Russian. Between inflection
and derivation'' explicitly states that his goal is not to assign the
morphological phenomena he studies to either pole of this dichotomy,
but rather to underpin those properties of various uses of the same
affix which make it so controversial to unequivocally classify them as
either derivational or inflectional. In the first part of the article Say
discusses such uses of Russian sja-verbs as passive, decausative,
reflexive, reciprocal etc. and shows that according to criteria proposed
by Haspelmath (2002: 71) these verbs do not show uniform
behaviour. While purely formal properties of this process seem to
pattern with inflection, its semantic and combinatorial properties are
more like those of derivational affixes, at least in more restricted
reflexive, reciprocal and decausative uses (passive sja-formation are
in most respects like inflection). The second part of the paper is
devoted to the so called 'antipassive' uses of sja-verbs, which fall into
two classes: 'lexical antipassives', which are unproductive and
semantically irregular, and 'grammatical antipassives', which, despite
important similarities with 'lexical' ones, are productive and their
interpretation is context-dependent, not lexically restricted and
idiosyncratic. However, Say shows that even 'grammatical'
antipassives may become lexicalized for some speakers, and thus it is
unreasonable to draw a strict dividing line between them. This paper
convincingly shows that the properties that are thought as defining
inflection and derivation can cut across not only such a polyfunctional
process as Russian sja formation in general, but also a semantically
homogenous phenomenon, like Russian sja antipassivization.

Rok Zaucer in ''Slavic prefixes as state morphemes. From state to
change-of-state and perfectivity'' argues that such properties of
prefixal verbs in Slavic languages as (i) directionality of prefixal motion
verbs and (ii) change-of-state (perfective) meaning of prefixal
derivatives in general can be explained if we assume that prefixes
introduce a stative subevent into the verb's event structure, retaining
the meanings of homophonous prepositions. Moreover, the correlation
between a derivational prefix and verbs inflectional property of being
perfective is only indirect and arises by regular process of event-
composition (cf. similar proposal by Pazel'skaya and Tatevosov 2005).
Also, only those prepositions which can be used in stative copular
constructions denoting location of an object, can have a cognate
perfectivizing prefix, and only those prefixes which have cognate
prepositions are actually perfectivizing. For instance, the delimitative
prefix 'po-' is argued to derive verbs which in many respects pattern
like imperfective rather then perfective ones. Slavic prefixation thus is
argued to be derivational rather then inflectional.

''Delineating the boundary between inflection-class marking and
derivational marking. The case of Sanskrit -aya'' by Gregory T. Stump
discusses the Sanskrit suffix -aya- which is usually regarded as a
causative morpheme. Stump tests this suffix against the following
(i) Distributional parallelism of inflection-class markers: If a mark 'x' of
inflection-class membership appears in particular cells of the paradigm
of a member of inflection class 'A' and some contrasting mark 'y'
appears in the same cells of the paradigm of a member of some
contrasting inflection class 'B', then 'y', like 'x', is a mark of inflection-
class membership (sufficient but not necessary property of inflection-
class markers).
(ii) Semantic contrast between derived stems and their bases: A mark
of derivation signals a particular semantic relation between two
lexemes. A mark of inflection-class membership does not, in itself,
signal a particular semantic relation between two lexemes (weakly
necessary property of derivation class markers).
(iii) Criterion of paradigmatic opposition of inflection-class markings: In
the paradigm of a given lexeme, a mark of inflection-class membership
may be paradigmatically opposed to another mark of inflection-class
membership, but not to a mark of derivation (sufficient but not
necessary property of inflection-class markers).
(iv) Criterion of uniformity of derivational markers: Marks of derivation
are associated with whole lexemes, and therefore occur on all of the
derived lexeme's stems. The appearance of inflection-class markers
may be sensitive to differences among the morphosyntactic property
sets associated with the various cells in a lexeme's paradigm
(necessary but not sufficient property of derivational markers)

When evaluated against these criteria, -aya- turns out to be an
inflection-class marker rather then a true derivational marker, since it
does not appear in all wordforms of Sanskrit causative derivatives.


The papers comprising the volume under review may be classified
according to several criteria. First of all, the papers fall into three main
groups: those which discuss the 'external' demarcation of morphology
from syntax (Rood, Cysouw, Milicevic, Blom, Enrique-Arias, Heine and
König), those which focus on the differences or similarities between
compounding and derivation (Bauer, Booij, Scalise et al., Stekauer,
Fradin, Amiot), and those dealing with the differentiation of inflection
and derivation (Ricca, Lloret, Manova, Say, Zaucer, Stump). On the
other hand, there are papers which argue for demarcation of such
and such components, at least in the languages they discuss (Blom,
Bauer, Scalise et al., Fradin, Amiot, Ricca, Manova), those, which, on
the contrary, claim that such demarcation is unnecessary or
impossible to draw (Heine and König, Booij), while others really focus
not on the question 'does phenomenon A in language C belong to
type C?' but on the very fact that the data in question do not allow
unequivocal characterization in these terms, or on the diachronic
issues (Cysouw, Enrique-Arrias, Say). Another important criterion is
the nature of the argument used in the papers. Some start by showing
that some data do not fall into any of the traditional classes by the
commonly used criteria, but propose a novel criterion which allows
them to assign a non-prototypical phenomenon to a certain class (e.g.
Manova). Other present a more or less detailed analysis of more or
less prototypical cases and argue for a more or less clear-cut
boundary between, for instance, derivation and compounding (e.g.
Scalise et al., Fradin). Still others propose theoretical frameworks
which would capture similarities or differences between certain
phenomena better then the already existing ones (Booij, Stekauer).

The main idea one may draw from the volume as a whole, abstracting
away from the individual papers, is the following one: Demarcation in
morphology is a controversial problem, since along prototypical
instances of inflection, derivation and compounding showing important
differences in morphological, syntactic, and semantic behaviour, there
are quite a lot of borderline cases which do not allow unequivocal
characterization since they share properties of different classes. What
is important, then, and on what linguists should focus their attention is
not the question of theoretical relevance and viability of the very
notions 'inflection', 'derivation' and 'compounding', and the pursuit of
all-or-none classification of the relevant phenomena into these three
types, but the particular criteria and properties which underlie these
classes, and which, as is now well known, do not always cluster in a
straightforward way.

Finally, there are two major critical remarks I think important to make.
The first is that it would have been useful if the authors of the papers
have paid more attention to the work of their fellow-contributors.
There are only a few cross-references in the volume, even when the
authors discuss very similar phenomena, or draw similar arguments
for their conceptions. it is particularly striking in the face of the fact
that the volume is a collection of papers from a conference where all
the authors were present. The second point is that the volume
crucially lacks a large editorial introduction which would not only
outline the structure of the volume and very succinctly summarize the
articles, but would give a broad perspective on the problems
discussed in the volume and of possible approaches to these
problems, as well as some general conclusions which can be drawn
from the discussion of individual cases. This is especially important in
the light of the fact that the articles usually present well articulated and
convincing arguments based on a detailed analysis of empirical data,
arguments which, nevertheless, sometimes lead to diametrically
opposed conclusions.


Anderson, Stephen R. (1992) A-Morphous Morphology. Cambridge:
Cambridge University Press.

Bybee, Joan L., Revere D. Perkins & William Pagliuca (1994) The
Evolution of Grammar: Tense, Aspect and Modality in the Languages
of the World. Chicago IL: Chicago University Press.

Croft, William (1991) Syntactic Categories and Grammatical Relations.
The Cognitive Organization of Information. Chicago, IL: The University
of Chicago Press.

Di Sciullo, Anna Maria & Edwin Williams (1987). On the Definition of
Word. Cambridge, MA: MIT Press.

Fillmore, Charles A. (1988). The mechanisms of Construction
Grammar. In Proceedings of Berkeley Linguistics Society 14, pp. 35-

Goldberg, Adele E. (1995) Constructions. A Construction Grammar
Approach to Argument Structure. Chicago, IL: The University of
Chicago Press.

Haspelmath, Martin (2002). Understanding Morphology. London:

Heine, Bernd & Tanya Kuteva (2002) World Lexicon of
Grammaticalization. Cambridge: Cambridge University Press.

Jackendoff, Ray (1990) Semantic Structures. Cambridge, MA: MIT

Kager, René (1999) Optimality Theory. Cambridge: Cambridge
University Press.

Kay, Paul (1997) Construction Grammar. In Paul Kay (ed.) Words and
the Grammar of Context, pp. 123-132. Stanford, CA: CSLI.

Lieber, Rochelle (2003) Compound interpretation: Lexical semantics,
not syntax. In G. Booij et al. (eds.), Topics in Morphology: Selected
Papers from the 3rd Mediterranean Morphology Meeting, p. 241-253.
IULA, Universitat Pompeu Fabra.

McCarthy, John (1995) Extensions of Faithfulness: Rotuman revisited.
Ms., University of Massachusetts, Amherst
(, ROA # 110)

McCarthy, John (2005) Optimal paradigms. In L. Downing et al. (eds.)
Paradigms in Phonological Theory, pp. 170-210. Oxford: Oxford
University Press (, ROA # 485)

Mel'cuk, Igor A. (1993-2000) Cours de morphologie générale. Vols. 1-
2. Montréal/Paris: Les presses de l'Université Montréal/C.N.R.S.

Nunberg, Geoffrey, Ivan A. Sag & Thomas Wasow (1994) Idioms.
Language, 70, pp. 491-538.

Pazel'skaya, Anna & Sergey Tatevosov (2005). Uninflected VPs,
Deverbal Nouns and Aspectual Architecture of Russian. Paper
presented at FASL 14, Princeton University, May 2005.


Peter M. Arkadiev is a PhD student and junior research fellow at the
Department of Typology and comparative linguistics of the Institute of
Slavic studies of the Russian Academy of Sciences, Moscow. His main
interests are linguistic typology with focus on event and argument
structure and its formal realization, and theoretical approaches to
morphology. He works mainly on Lithuanian, Adyghe and Japanese.