Stump, Gregory T. (2001) Inflectional Morphology: A Theory of
Paradigm Structure (Cambridge Studies in Linguistics 93).
Cambridge University Press, hardback ISBN: 0-19-512600-9, xvi+308pp.
Tania Avgustinova, Saarland University
This book is about Paradigm Function Morphology (PFM) whose
guiding idea is that in the domain of inflectional
morphology, the primary object of analysis (both for the
linguist and the language learner) is the paradigm rather
than merely the word. Stump's objective is to develop this
idea as precisely as possible in order to demonstrate the
particular merits of the theory which it engenders.
Equating the definition of a language's inflectional
morphology with the definition of its paradigm function (a
theoretical construct unique to this theory), PFM presumes
the existence of several different rule types. Chief among
these are paradigm functions, realisation rules, and
morphological metageneralisations. These three basic rule
types are organised hierarchically. A paradigm function's
definition is stated in terms of realisation rules which
give morphological expression to a specified set of
morphosyntactic properties and whose evaluation is in turn
determined by morphological metageneralisations.
Realisation rules are of two types: (i) rules of exponence
directly specify the concrete exponents associated with the
property set being realised; (ii) rules of referral instead
refer the realisation of some property set to some other
realisation rule(s). When two or more realisation rules are
associated with the same morphonological regularity, that
association is expressed by means of a morphological
The book is organised into eight chapters. The main text is
preceded by a list of abbreviations (pp. xiv-xvi) and
followed by notes (pp. 277-191), references (292-300) and
index (pp. 301-308).
Chapter 1 "Inferential-realizational morphology" (pp. 1-30)
offers a brief contrastive introduction to theories of
inflectional morphology, highlighting the evidence which
favours realisational theories over incremental ones,
1. The morphosyntactic properties associated with the
inflected word may exhibit EXTENDED EXPONENCE in that
2. The morphosyntactic properties associated with an
inflected word's individual markings may underdetermine the
properties associated with the word as a whole.
A theory of inflectional morphology that is inferential
rather than lexical and realisational rather than
incremental minimises empirically unmotivated theoretical
distinctions, being compatible with the following
1. There is no theoretically significant difference between
concatenative and nonconccatenative inflection.
2. Exponence ids the only association between inflectional
markings and morphosyntactic properties.
3. An uncompounded word's morphological form is not
distinct from its phonological form.
Chapter 2 "Paradigm functions" (pp. 31-61) presents the
fundamental principles and claims underlying PFM and a
detailed account of the architecture of the theory. The
author examines the nature of morphosyntactic properties
and their relationship to the notions of paradigm and
paradigm function, the nature of realisation rules and
their organisation into blocks, the role of morphonological
rules in the evaluation of realisational rules, and the
role of realisational rules in the definition of a
language's paradigm function.
A detailed exploration of the evidence motivating the
properties of PFM is undertaken in the rest of the book.
Chapter 3 "Rule competition" (pp. 62-95) concerns the
tenability of a principle according to which competition
among realisation rules belonging to the same block is
resolved in favour of the narrowest applicable rule
(Paninian Determinism Hypothesis).
Chapter 4 "Headedness" (pp.96-137) addresses the question
of what determines the incidence of head marking in
inflectional morphology. In PFM the phenomenon of head
marking can be attributed to the Head-Application
Principle, a universal principle for evaluation of paradigm
function in the definition of a headed lexeme's paradigm.
This principles correctly entails the following empirical
1. Paradigm Uniformity Generalisation (PUG): head marking
is an all-or-none phenomenon: if a rood ever exhibits head
marking in its inflectional paradigm, it always does.
2. Coderivative Uniformity Generalisation: Where X and Y
are headed coderivatives (i.e. arise by means of the same
category-preserving rule), either X or Y both exhibit head
marking or neither does.
Chapter 5 "Rule blocks" (pp. 138-168) shows that paradigm
functions make it possible to provide a satisfactory
account of the full range of observable interactions among
realisation-rule blocks in the world's languages. Also, an
independent motivation is presented for postulation of
rules of referral which are essential to the proposed
account of portmanteau, parallel, and reversible position
classes. The view is promoted that inflectional templates
are nothing other than paradigm functions, and thus, all
inflectional paradigms are 'templatic', since they always
involve paradigm functions.
Chapter 6 "Stem alternations" (pp. 169-211) addresses the
question of why one stem is chosen over another in a
particular cell of the inflectional paradigm of a lexeme
that exhibits a variety of distinct stems. Following an
unpublished work of Arnold Zwicky, the author distinguishes
two different sorts of rules serving to determine stem
choice: stem-selection rules (a kind of realisation rule)
and morphological metageneralisations.
Chapter 7 "Syncretism" (pp. 212-241) proposes a theory of
syncretism. The author argues convincingly that syncretism
is not a unitary phenomenon: some types of syncretism are
stipulated, while others are not; of those that are, some
are directional, while others are not.
Chapter 8 "Conclusions, extensions, and alternatives" (pp.
242-276) summarises the evidence motivating the
introduction of paradigm functions into morphological
theory and reviews the principle theoretical claims of PFM.
Furthermore, a paradigm-based conception of inflectional
semantics is presented and the analogy of derivational
'paradigms' to inflectional paradigms is examined. Finally,
the similarities and differences between PFM and Network
Morphology are explored, with particular attention to some
alternative formulations which the latter theory suggests
for the former.
The postulation of paradigm function makes it possible for
PFM to capture several types of generalisation which remain
elusive in other frameworks. The author motivates the
principal characteristics of PFM with empirical evidence
drawn from genetically and typologically diverse languages.
Throughout the work, a conscious attempt is made to be as
precise and explicit as possible in drawing conclusions, as
well as to avoid theoretical preconceptions. A key
assumption is "that the rules and principles determining a
word's inflectional form are properly morphological (being
reducible neither to principles of syntax nor to principles
of phonology), and that the interfaces of these rules and
principles with other grammatical components are in general
Tania Avgustinova received her Ph.D. in Slavic and
Computational Linguistics at Saarland University. In 1998
she was awarded an individual grant from the German science
foundation (DFG) to work on modular language-family
oriented grammar design at the Department of
Computational Linguistics and Phonetics in Saarbruecken.