Blake, Barry J. 2001. Case. Second Edition. Cambridge
University Press, hardback ISBN 0-521-80761-1, xx + 227
pp, Cambridge textbooks in linguistics.
Reviewed by Muriel Norde, Scandinavian department,
University of Amsterdam
CONTENTS
Chapter 1, "Overview" introduces the subject. Case is
defined as "a system of marking dependent nouns for the
type of relationship they bear to their heads.
Traditionally the term refers to inflectional marking,
and, typically, case marks the relationship of a noun to
a verb at the clause level or of a noun to a preposition,
postposition or another noun at the phrase level" (p. 1).
This is clearly a working definition, for as we will see
later on, "case" is also used for a variety of other
systems, both by other scholars and by Blake himself. But
this first definition is clearly exemplified by examples
from Turkish and Latin. Other manifestations than the
marking of dependent nouns are discussed as well. These
are concordial case (adjectives, pronouns and determiners
agreeing in case with their head), a system commonly
found in (older) Indo-european languages; case on non-
nouns (adjectives, pronouns or determiners without a
nominal head), vocatives (as in Latin "domine"
'master!'); ungoverned case (e.g. Latin "me miserum"
(1SG.ACC miserable-ACC) ' unhappy me') and analytic
case markers such as adpositions (whether or not the
latter should be regarded as case markers will be
discussed in the "Evaluation" section at the end of this
review). The final section of this chapter deals with
competing mechanisms for marking head-modifier relations.
These are head-marking (e.g. marking of the verb by
prefixed pronouns; word order (e.g. for distinguishing
subject, direct object and indirect object in English);
adverbs and relator nouns (as in English "on top of");
possessive adjectives (e.g. Old English "mi:n",
originally a genitive form of the pronoun but now
inflected as an adjective for gender, number and case).
Chapter 2, "Problems in describing case systems", is
concerned with distinguishing cases from one another and
describing their meanings and functions, neither of which
is an unproblematic affair. The first problem is that
case paradigms are usually not isomorphic and that not
all distinctions are found in all classes of nouns. In
Indo-european languages for example, there is
considerable syncretism, or neutralization, of case
distinctions (e.g. of the nominative and accusative of
neuter nouns). According to what Blake calls the
traditional method, cases in languages such as Latin
paradigms are aligned according to function (e.g. dative
or ablative) even if not all forms are realized
unambiguously. In spite of its shortcomings (e.g. the
neglect of systematic syncretisms) this method is to be
preferred as the method of description of Indo-european
languages. Blake shows that an alternative method, i.e.
one which only describes case forms, has several
disadvantages. If for instance the latin word "dominae"
'mistress' is characterized as a genitive-dative form,
there would be no way of determining which form of a
third-declension adjective should be used if it modifies
this noun, for third declension adjectives (e.g.
"tristis" 'sad') do distinguish between FEM.SG.GEN and
FEM.SG.DAT. If however one (traditionally) recognizes two
forms, genitive "dominae" (taking the adjective
"tristis") and dative "dominae" (taking the adjective
"tristi"), this difficulty does not arise.
In other languages than Indo-european however, for
instance the Australian Pama-Nyungan languages, the
traditional method is shown to be less useful. In these
languages in which the number of paradigms and functions
is small and there is little or no concord, it is
possible to describe case systems in terms of forms, also
because there is a discrepancy between the case marking
of nouns and the case marking of pronouns. This formal
approach to Pama-Nyuangan has the advantage of explicitly
signaling systematic syncretisms.
A final obstacle in the description of case systems are
cases that are expressed by one case form in some
paradigms but another in other paradigms, such as the
Russian partitive. These are best regarded as "non-
autonomous" cases and treated separately.
The second major problem that Blake discusses in this
chapter is the description of meanings and functions.
Traditionally, the various meanings and functions of a
given case are simply listed, but there have also been
attempts to describe the meanings and functions in a more
systematic way, e.g. by recognizing a general meaning for
each case and / or relating the cases in a system of
oppositions using a minimum of features. This method goes
back to the classical period, but it became prominent in
the works of 13th century grammarians and in two classic
works on case from the 1930s: Hjelmslev 1935 and Jakobson
1936. Blake gives clear summaries of these works, even
though he prefers the traditional analysis because from
the generalized meanings one cannot derive in which
contexts a given case is to be used. Nevertheless, he
argues that it is useful to classify cases in a less
strict way, e.g. by grouping them according to some
general oppositions: nominative versus oblique, core
(nominative, ergative and accusative) versus peripheral
(all other) cases and nonlocal versus local cases. Such
distinctions can be used to capture systematic
similarities between some cases and may help account for
syncretisms.
Chapter 3, "Modern approaches to case" deals with case on
a more abstract level, which has been prominent in
Chomskyan syntactic theories and other frameworks, which
are all more or less influenced by Chomsky. In these
theories the view is taken that case is a universal
entity which exists independently of means of expression
(e.g. inflection). Using data from an Australian language
that he has studied extensively, Kalkatungu, Blake starts
by discussing the concept of grammatical relations and
shows that there is no one-to-one relationship between
case and grammatical relation. He further discusses case
theory in the Chomskyan framework (both in Government &
Binding theory and in the more recent Minimalist
Program), the difference between semantic roles and
grammatical relations, Panini's karaka theory,
Fillmore's influential Case Grammar, Relational Grammar,
John Anderson's Localist Case Grammar and Starosta's
Lexicase. Blake summarizes the main tenets of these
theories with respect to case in a concise yet clear way,
though without positioning himself. In the final section
of this chapter Blake shows how grammatical relations,
cases and semantic relations can be ordered
hierarchically.
In chapter 4, "Distribution of case marking" Blake
describes the distribution of case marking within the
sentence, exemplified with data from a wide variety of
languages. Starting with case marking within the clause,
Blake discusses several means of marking complements and
adjuncts (usually nouns phases or adverbial phrases) to
the predicate (usually a verb). This is the type of case
marking described in the central definition of case given
in the beginning of chapter 1 and includes such basic
examples as the Latin nominative expressing a subject
relation to the verb. Somewhat further from this
definition are case concord (signaling a modifier
relation) or predicates other than verbs. Case marking
within the noun phrase (internal relations) may be used
to denote adnominal relations (e.g. head-genitive).
External relations within the noun phrase concern the
distribution of adpositional case markers and its
relation to basic word order and the distribution of
affixes (word marking versus phrase marking, of which
Blake discusses several types). Other topics discussed in
this richly illustrated chapter are case marking within
the word (prefixes (almost absent) versus suffixes, stem
formatives, compound case marking and multiple case) and
case marking within the subordinate clause (e.g. Greek
and Latin "accusativus cum infinitivo" constructions).
In Chapter 5, "Survey of case marking", Blake provides a
bird's-eye view of case systems and their marking. The
first part of this chapter, "Organization of the core" is
concerned with nuclear relations, such as the nominative
and accusative in accusative languages and the nominative
(sometimes termed absolutive) and ergative in ergative
languages. Blake discusses case marking in active
languages (in which the argument of one-place predicates
is marked sometimes as the agent of a two-place verb,
sometimes as the patient of a two-place verb), mixed
systems (languages with both an ergative and an
accusative case) and direct-inverse systems (in which a
marker on the verb indicates whether the agent is ranked
higher or lower than the patient on the person hierarchy
(1>2>3)). In the final section on core case marking Blake
identifies some criteria to establish the meaning and
function of core cases. The remainder of this chapter is
devoted to so-called peripheral cases: dative, genitive,
partitive and local cases (allative, ablative,
translative etc.). Besides these more or less well-known
peripheral cases there exist a number of other, often
less frequent cases, such as the instrumental or the
comitative (expressing accompaniment). In the final
section of this chapter Blake takes up an interesting
problem, viz. the question of whether inflectional cases
grow and disappear in a certain order, in other words,
whether there exists an inflectional case hierarchy, and
he concludes that there indeed seems to be such a
hierarchy. If a language possesses a certain case, it
usually possesses the higher ranked cases as well,
although not all cases are necessarily realized
inflectionally (which, in my view, reduces the predictive
qualities of such an implicational hierarchy).
Chapter 6, finally, is entitled "Life cycle of case
systems". This chapter first offers a survey of verbal,
nominal and adverbial sources of case markers (both
adpositional and inflectional). These are developments
which are nowadays normally termed processes of
grammaticalization, and it is therefore surprising that
the term grammaticalization is not used in this chapter
(only once Blake speaks of nouns that are "grammaticised").
Also the decay of case systems is touched upon,
illustrated by the loss of inflectional case in Latin
the / Romance languages and in English. Finally, case
inflections may get another grammatical functions, as in
the Australian language Kala Lagau Ya, where case markers
have become time / aspect markers on the verb. In the
finale, Blake concludes that "case makes sense". It is a
practical and economical way to link lexical forms in
grammatical constructions. This is probably why it is
cyclically renewed.
The book concludes with an extremely helpful 12-page
"Guide to terminology", a guide to further reading,
references and extensive indices.
EVALUATION
"Case" is a rich book, written with a contagious
enthusiasm for the phenomenon. For someone whose
favourite subject in school was Latin grammar (a subject
one was supposed to abhor) it is rewarding to read that
"Case has aesthetic properties" (p. xvi). The range of
languages covered is impressive - chapters 4 and 5, in
particular, are veritable treasure troves (One of the few
imperfections being the lack of translations in table 4.3
on Archi case marking and table 4.4 on Yuwaalaray case
marking). Another merit of the book is that it is
theoretically unbiased. Throughout the book the data
themselves form the basis of discussion and analysis, yet
other ("modern") approaches are not ignored. Blake offers
an objective survey of these theories, signals some
potential problems but is not really critical of any of
them. For an introduction such as this book this is the
preferred approach, yet one cannot help being curious
about Blake's own theoretical views. In the following I
will therefore discuss a few of Blake's examples which
reveal his own definition of case and for which he offers
no alternatives. That would probably fall outside the
scope of this textbook, but for the sake of discussion I
will mention a few anyhow.
Blake's central definition will be repeated here: "a
system of marking dependent nouns for the type of
relationship they bear to their heads. Traditionally the
term refers to inflectional marking, and, typically, case
marks the relationship of a noun to a verb at the clause
level or of a noun to a preposition, postposition or
another noun at the phrase level" (p. 1). It soon becomes
evident however that Blake does not generally take the
traditional view that case is realized inflectionally.
Throughout the book, he also considers adpositions as
(analytic) case markers. However, on competing mechanisms
such as head marking and word order Blake remarks
(p.13):): "One could take the view that all these means
of expressing grammatical relations are forms of case
marking" (see also 3.3). This again seems to suggest that
Blake does not regard prepositions as "competing
mechanisms". In my view, it would have been more
clarifying to distinguish between inflectional case
marking and all other means of expressing head-dependent
relations. In his classical article on case in Russian
Jakobson (1971 [1936]:28-29), for example, explicitly
dismisses adpositions as case markers:
Auch das System der präpositionalen Fügungen ist nicht mit
der flektierenden Deklination zu verwechseln, da die
Sprachen, die beide erwänhten Kategorien besitzen, erstens
die syntaktischen Verwendungen eines Kasus mit Präposition
und ohne solche (mittelbare unmittelbare Verbindung)
einander entgegensetzen, und zweitens die Bedeutung der
Kasus und der Präpositionen als zwei besondere
Bedeutungsgattungen deutlich voneinander unterscheiden
[...]
A similar objection might be raised to the figures on p.
15, representing the relation between basic word order
(VSO, SVO, SOV) and case marking. Again, "case marking"
includes adpositions, and to my mind it would have been
more interesting to learn about the relation between
basic word order and inflectional case marking.
Jakobson's observation is also relevant to Blake's
discussion of grammatical and semantic cases (chapter 2,
pp. 31ff.). Here Blake argues that this distinction is
often not clear-cut. In Latin for example, the accusative
(a grammatical case) may also denote the semantic role of
destination and in such constructions it does function as
a direct object. Similarly, Blake argues, the Latin
ablative (a semantic case) may denote the "agent" in
passive constructions, as in "occisus a consule" 'killed
by the consul'. Here the agent is not, of course,
expressed by the ablative, but by a prepositional
construction [a(b) + ablative]. Even if one accepts the
analysis of prepositions as analytic case markers it
would be incorrect to regard "a consule" as a mere
ablative. In order to illustrate the use of semantic
cases to express the agent in passive constructions, it
would have been better to use examples from other
languages. In Old Norse for example, the "bare" dative is
used in similar examples (Nygaard 1905:99), and in
Russian the "bare" instrumental (Kohls et al. 1989:162).
First on p. 73 Blake's analysis is made explicit: "In the
normal practice of modern grammarians of virtually all
schools the choice of case and / or adposition is
disregarded in a situation like this. The preposition
a/ab is not interpreted as having its normal meaning, but
is regarded as a grammatical marker." This is a rather
vague statement, somewhat surprising for a work which is
otherwise quite detailed in its analyses, and by no means
sufficient as a motivation for regarding the adposition
as part of the case. It would have been helpful if the
author had been more explicit about what he regards as "a
situation like this", i.e. under what circumstances the
adposition is considered devoid of any meaning or
function. What is furthermore confusing is that the
author uses the term "a/ab + ablative" besides mere
"abalative" when referring to the agent of Latin
passive constructions (also p. 73 and p. 80).
Similarly, on p.174 Blake claims that in Pennsylvania
German the dative case has become the only way of
expressing possession, so that there is now a genitive-
dative case. However, in the example he mentions: "em
Gaul sei(n) Schwans" (the-DAT horse-DAT tail) 'the
horse's tail', it is again not the bare dative which
expresses possession but a dative plus a resumptive
pronoun. This construction is most probably a
reanalysis of an indirect object construction (as in
German "er hat meinem Vater seinen Hut genommen";
Behaghel 1923:638). In fact the construction is quite
widespread in Germanic and occurs also in languages that
lack morphological case, such as Norwegian, Dutch and
Afrikaans (see Norde 1997:55ff for examples and
discussion of the origin of these constructions). Blake
diminishes the pronoun as a "possessive adjective cross-
referencing the pronoun", but it forms an essential part
of the construction. "True" possessive datives were found
in for instance the older Scandinavian languages (as in
Old Swedish "ör höfthi manni" (out-of head-DAT man-DAT)
'out of a man's head', where it frequently replaces the
genitive in constructions expressing inalienable
possession.
In yet other respects Blake's views are remarkably
traditional, as in the following quote (p. 10): [...] in
English, all prepositions govern the accusative" To me
this is reminiscent of 19th century practice to force
Latin grammar on uninflecting Western European languages.
Similarly, on p. 57 he writes that "In English there is
an inflectional case system, which is confined to
personal pronouns [...] and "who/whom" for some speakers".
Again this is presented as an undisputed fact, but
several authors have been arguing that English has no
case at all (see for instance Hudson 1995). Furthermore,
the author assumes that English has a genitive case for
nouns, a position which no longer seems tenable (see
Norde 2001:247ff. for discussion and references). Whether
or not English has inflectional case may seem like a
detail that falls outside the scope of a general
introduction, but to my mind the question of whether two
forms (as in English I/me) imply two cases is crucial to
the definition of case, and the discussion of such issues
should have been more prominent in this book.
A final objection one might raise against this book is,
trivially, the time of appearance of the second edition.
The most significant changes are additions and revised
interpretations of the data and an update on the section
on abstract case in the Minimalist Program. Additions to
the data are welcome of course, but a revision of the
Chomskyan section seems less relevant, considering that
these section covers some 6 pages only and case has
become less prominent in this framework. What is more,
the section will probably be outdated again in a couple
of years. One theory that did get a considerable boost in
the 1990s is grammaticalization theory, which is relevant
for the section on the life cycle of case systems, yet
this section has hardly been updated at all. One of the
basic textbooks is not mentioned (Hopper & Traugott
1993), nor are such works as Giacalone Ramat & Hopper
1998 or Campbell 2001, or more specialized works such as
DeLancey 1997 on the grammaticalization of adpositions.
Finally, some new "modern" theories have concerned
themselves with case as well, such as Construction
Grammar (Barddal 2001) and Optimality Theory (see for
instance some of the recent additions to the Rutgers
Optimality Archive at http://roa.rutgers.edu/). So, in a
sense, the book appeared "too early", but one could also
say that there is plenty of material for a third edition,
which leaves us something to look forward to!
REFERENCES
Barddal, Johanna. 2001. Case in Icelandic. A synchronic,
diachronic and comparative approach. Lund: Department of
Scandinavian languages/
Behaghel, Otto. 1923. Deutsche Syntax I: Die Wortklassen
und Wortformen. Heidelberg: Winter.
Campbell, Lyle (ed.). 2001. Language Sciences 23: special
issue on grammaticalization.
DeLancey, Scott. 1997. Grammaticalization and the
gradience of categories. Joan Bybee, John Haiman & Sandra
A. Thompson (eds.): Essays on language function and
language type, pp. 51-69. John Benjamins, Amsterdam &
Philadelphia.
Giacalone Ramat, Anna & Paul Hopper. 1998. The limits of
grammaticalization. Amsterdam: Benjamins.
Hjelmslev, Louis. 1935. La catégorie des cas. Étude de
grammaire générale. Aarhus: Universitetsforlaget.
Hopper, Paul & Elizabeth Traugott. 1993.
Grammaticalization. Cambridge University Press.
Hudson, Richard. 1995. 'Does English really have case?'.
Journal of linguistics 31: 375-392.
Jakobson, Roman. 1971 [1036]. 'Beitrag zur allgemeinen
Kasuslehre: Gesamtbedeutungen der russischen Kasus'.
Selected writings II. 23-72. Den Haag: Mouton.
Kohls, Siegfried et al. 1989. Praktische russische
Grammatik. Leipzig: VEB Verlag.
Norde, Muriel. 1997. The history of the genitive in
Swedish. A case study in degrammaticalization. PhD
thesis, University of Amsterdam.
Norde, Muriel. 2001. 'Deflexion as a counterdirectional
factor in grammatical change'. Language Sciences 23:231-
264.
Nygaard, Marius. 1905. Norrøn syntax. Kristiania:
Aschehoug.
About the reviewer.
Muriel Norde is a postdoc researcher at the Department of
Scandinavian Languages at the University of Amsterdam.
She has written a PhD thesis on the history of the
genitive in Swedish, as well as articles on diachronic
morphology, grammaticalization and language contact in
the history of the Scandinavian languages. She is
currently working on a project on the loss of
inflectional case marking in Continental Scandinavian.
|
