Review of Impersonal Constructions
|EDITORS: Andrej Malchukov, Anna Siewierska
TITLE: Impersonal Constructions
SUBTITLE: A Cross-Linguistic Perspective
SERIES TITLE: Studies in Language Companion Series 124
PUBLISHER: John Benjamins
Peter M. Arkadiev, Institute of Slavic Studies, Russian Academy of Sciences, Moscow
Impersonal constructions have only recently become an important topic in
cross-linguistic studies (cf. Lehmann et al. 2000, Creissels 2007, Siewierska
ed. 2008), especially in a European perspective (cf. Lambert 1998, Cabredo
Hofherr 2006). The volume “Impersonal Constructions: A Cross-Linguistic
Perspective” edited by Anna Siewierska (1955-2011) and Andrej Malchukov is the
first large-scale collection of papers on impersonal constructions assuming a
typological perspective and aiming at widest possible coverage of languages of
different genetic and areal provenance and structural type. In addition to such
a broad data-base, the book under review is comprehensive in a no less important
respect, representing a variety of approaches to impersonal constructions (both
functional-typological and generative) and different problems pertaining to this
field, such as the lexical-semantic fields and discourse-pragmatic circumstances
favouring the occurrence of impersonal constructions, the functional-semantic
typology of impersonal constructions, their structural types and relation to
such independently occurring properties of languages as case-marking of
nominals, verbal cross-reference, passive and anticausative, classification of
intransitive verbs into unaccusative and unergative, zero and overt anaphora,
grammatical encoding of animacy etc.
In the Introduction (p. 1-15) Andrej Malchukov and Anna Siewierska give a
succinct and useful general discussion of the conceptual, terminological and
cross-linguistic problems of impersonal constructions and provide a detailed
overview of the structure of the volume and of individual contributions.
The main portion of the book consists of three major part comprising is sum
twenty articles. Part 1 “Typological and theoretical aspects” contains four
papers addressing problems of different degree of generality arising in the
cross-linguistic and theoretical study of impersonal constructions. Andrej
Malchukov and Akio Ogawa in “Towards a typology of impersonal constructions. A
semantic map approach” (pp. 19-56) propose a general functional typology of
impersonal constructions viewing them as encoding certain deviations from the
universal subject prototype (Keenan 1976). Three functional-semantic types of
impersonal constructions are postulated: R-impersonals with non-referential or
indefinite subjects, T-impersonals with non-topical (rhematic) subjects and
A-impersonals with non-agentive (non-volitional, inanimate) subjects. Malchukov
and Ogawa show that different functional types of impersonal constructions show
propensity towards different kinds of morphosyntactic encoding: R-impersonals
often lack a subject argument altogether or use some sort of expletive subject,
or a specialized pronoun with generic reference like German “man”; T-impersonals
(such as presentational constructions) often involve word order inversion and
lack of subject-verb agreement; A-impersonals often display non-canonical
encoding of subject by means of case marking, verbal agreement or both. A
semantic map of “non-prototypical subjecthood” is drawn on p. 42, showing how
different encoding strategies are distributed over inanimate, non-referential,
non-topical etc. subjects. “Transimpersonal” constructions (Malchukov 2008) with
an experiencer object and an indefinite/dummy subject are also discussed at
length, and it is argued that such construction constitute a diachronic link
between R-impersonals and A-impersonals.
In “Overlap and complementarity in reference impersonals. Man-constructions vs.
third person plural-impersonals in the languages of Europe” (pp. 57-89) Anna
Siewierska discusses the areal patterning of two major types of referential
impersonal constructions in the European languages: those involving the
specialized impersonal pronoun like German “man” or French “on”, and those using
a third person plural pronoun or verbal affix. As Siewierska shows,
man-impersonals constitute a trait of the “Standard-Average European” area,
being mostly characteristic of French, German, Dutch, Frisian and Mainland
Scandinavian languages, while third person-plural impersonals mostly occur
outside this area. However, Siewierska shows that “all man-imp languages also
display 3pl-imps” (p. 73), so that in such languages as French and Dutch the two
kinds of impersonal constructions are often in competition. Another typological
issue discussed is the putative correlation of the type of impersonal with the
pro-drop parameter (Holmberg 2005); Siewierska shows that though in Europe the
correlation mostly holds, the theory linking the presence of specific impersonal
pronouns to the requirement of overt subject can account neither for the exact
distribution of man-impersonals nor for the very existence of 3pl-impersonals.
Instead, Siewierska shows that “there is a correlation between pro-drop
properties and the range of uses the different types of non-referential subjects
display” (p. 79): in non-pro-drop languages man-impersonals occur not only in
generic but also in episodic contexts with both non-specific and specific
reference, while in pro-drop languages man-impersonals are usually confined to
generic contexts. 3pl-impersonals show the opposite behaviour.
Werner Abraham in “Verbs of motion. Impersonal passivization between
unaccusativity and unergativity” (p. 91-125) discusses a variety of issues
having to do with the dual classification of verbs of motion in such European
languages as Italian, German and Dutch, where they show properties of both
unaccusative and unergative predicates. Abraham proposes to resolve this paradox
by assuming that verbs of motion, being agentive, are unergative in their
imperfective uses giving rise to impersonal passives, but are unaccusative in
the passive (resultative) participles occurring in perfect constructions.
Another claim by Abraham, now of a more empirical than theoretical nature, is
that impersonal passives occur in languages having several different strategies
for encoding of the passive, e.g. “sein”-passive vs. “bleiben”-passive in German
or synthetic vs. periphrastic passive in Scandinavian languages. However, this
claim is refuted by Lithuanian, which arguably has just one passive strategy and
simultaneously is the language with the widest and most productive use of
impersonal passives (e.g. Timberlake 1982; Wiemer 2006).
Volker Gast and Holger Diessel in “On the distribution of subject properties in
formulaic presentationals of Germanic and Romance. A diachronic-typological
approach” (pp. 127-166) propose a detailed structural typology of presentational
impersonal constructions in Romance and Germanic. The typology is based on three
main parameters: (i) type of existential predicate (one-place vs. copular vs.
transitive); (ii) type of expletive (no expletive vs. weak pronominal expletive
vs. locative expletive) and (iii) whether the language allows verb-initial order
in thetic sentences. Gast and Diessel focus on the synchronic distribution and
diachronic evolution of subject properties (case marking, agreement, raising) in
these constructions, i.e. on the question how these properties are distributed
between the expletive and the referential noun phrase introducing the new
discourse participant. They show that (i) languages where thetic sentences allow
verb-initial order (Romanian, Italian, Portuguese, Catalan) either do not have
expletives in impersonal presentational constructions or such expletives do not
show subject properties, and (ii) that in languages with an obligatory preverbal
constituent (French, English, Norwegian) expletives have acquired subject
properties via the reanalysis of an original transitive predicate denoting
possession as an existential predicate.
Part II “Diachronic studies” contains three articles dealing with the history of
impersonal constructions in Indo-European languages. Michela Cennamo in
“Impersonal constructions and accusative subjects in Late Latin” (pp. 169-188)
shows how the loss of voice contrasts in Latin has led to an emergence of rather
peculiar patterns of case marking with accusative case appearing first on
semantic patients of passives, then on the patientive subjects of unaccusative
verbs, and then on the agentive subjects of unergative and even transitive
verbs. Anna Giacalone Ramat and Andrea Sansò in “From passive to impersonal. A
case study from Italian and its implications” (pp. 189-228) present a detailed
corpus-based study of the development of the reflexive-based si-passive to a
si-impersonal from late Mediaeval to Modern Italian. They pin down the contexts
which could lead to the reanalysis of a passive construction with a generic
agent into an impersonal construction and document the gradual spread of
si-impersonals to different types of predicates and contexts. Giacalone-Ramat
and Sansò also discuss the emergence of the so-called inclusive si-impersonal
construction where the generic/indefinite subject necessarily includes the
speaker. They conclude (p. 226) that “a reanalysis of the passive marker as a
marker of generic human agency is a necessary precondition motivating its
extension to intransitive verbs” and the subsequent emergence of the impersonal
construction. Leonid Kulikov in “Passive to anticausative through
impersonalization. The case of Vedic and Indo-European” (pp. 227-254) discusses
a rare pathway of change leading from passive to anticausative on the basis of
the Vedic Sanskrit data and shows that impersonal passives have served as an
intermediate stage in the reanalysis of the passives of such verbs as “see”,
“hear”, “know”, and “say” into one-place anticausatives.
The third and largest part of the book (“Cross-linguistic variation in
Impersonal constructions: case studies”) comprises thirteen papers devoted to
impersonal constructions in individual languages and language families from all
over the world. Two papers deal specifically with meteorological predicates.
Amina Mettouchi and Mauro Tosco in “Impersonal configurations and theticity: The
case of meteorological predications in Afroasiatic” (pp. 307-322) develop a
typology of meteorological expressions based on the notion of partial vs. total
backgrounding of processes or entities resulting in different morphosyntactic
patterns with the atmospheric event expressed either as a verb or as a noun or
both, and make a strong argument for the recognition of theticity as a primary
factor determining grammatical properties of this type of sentences. Merja Salo
in “Meteorological verbs in Uralic languages - are there any impersonal
structures to be found” (pp. 395-438) shows on the basis of the data from 14
languages that though many meteorological expressions in Uralic contain some
sort of (not necessarily agentive) subject argument and thus are not
structurally impersonal, they are close to impersonal constructions
functionally. Unfortunately, these two papers do not refer to each other.
Two more papers dealing not with a single language but with a group of related
languages are devoted to the languages of Oceania. Claire Moise-Faurie in
“Impersonal constructions in some Oceanic languages” (pp. 581-606) discusses the
data of Polynesian and Kanak (New Caledonia) languages, where the following
strategies of encoding impersonal constructions occur: subjectless clauses with
avalent predicates (mostly for natural events), dummy pronouns (for some modal
predicates), agentless use of otherwise transitive verbs, encoding of the
notional agent as a possessor or an obliquely marked adjunct, zero anaphora,
and, rarely, specialized impersonal pronouns. An important distinction is drawn
between impersonal verbs which are not able to occur in syntactic structures
implying a referential subject or agent, and impersonal uses of otherwise
regular transitive and intransitive verbs. Jean-Christophe Verstraete in
“Impersonal constructions in Umpithamu and Lamalamic languages” (pp. 607-625)
describes a non-trivial kind of subjectless construction attested in several
Pama-Nyungan languages of northeastern Australia. In these languages,
experiencer predicates denoting “sudden, involuntary physical processes” (p.
613) are encoded as transitive verbs with an absolutive noun phrase expressing
the body-part and an accusative or oblique personal enclitic referring to the
experiencer. These structures are shown to be impersonal even in the presence of
an inanimate ergative noun phrase, since they do not admit a nominative personal
enclitic, which is otherwise able to cross-reference an ergative subject.
Verstraete draws functional parallels between such impersonal constructions and
voice, showing that they mostly occur when “an inanimate agent-like element
affects a human undergoer-like element” (p. 622).
Several papers discuss in detail just one particular construction. Doris L.
Payne in “The Maa (Eastern Nilotic) Impersonal constructions” adduces
morphosyntactic evidence showing that the “passive” construction in this
language does not contain a subject argument. Payne’s argumentation largely
repeats that of Mel’čuk’s (1997) discussion of the same construction. A large
part of this paper is devoted to a detailed corpus-based account of the
semantic-pragmatic functions of the Maa impersonal. Anna Kibort in “The elephant
in the room: The impersonal -ne/-te construction in Polish” (pp. 357-393)
provides a detailed description of a little studied construction, whose main
peculiarity is its underspecification with respect to the exact syntactic
structure it instantiates (verbal passive or non-passive adverbial). Anna
Bugaeva in “A diachronic study of the impersonal passive in Ainu” (pp. 517-546)
presents an account of morphological and syntactic properties of a peculiar
impersonal passive construction showing non-trivial dialectal variation, and
proposes a grammaticalization scenario for its development. Synchronically, the
impersonal in the Ishikari dialect is interesting in that it employs two
different impersonal markers for transitive verbs with 1st person vs. non-1st
person objects. From a historical perspective, Bugaeva shows that the most
productive impersonal marker in Ainu goes back to an existential verb attaching
to a nominalized predicate.
The remaining papers describe and discuss whole classes of impersonal
constructions in particular languages. Alain-Christian Bassène and Denis
Creissels in “Impersonal constructions in Jóola-Banjal” (pp. 285-306) provide an
overview of the morphosyntactic and semantic properties of different kinds of
impersonal constructions in an Atlantic language where such constructions can be
easily defined in structural terms as those lacking a subject cross-referencing
prefix on the verb. Ruth A. Berman in “Revisiting impersonal constructions in
Modern Hebrew” (pp. 323-355) addresses several kinds of impersonal constructions
from a discourse-pragmatic and developmental perspective, discussing such issues
as the correlation between grammatical impersonality and expository genres of
discourse, speaker’s age and education level, etc.
Three papers discuss impersonal constructions in polysynthetic languages with
highly complex verbal morphology. Edward Vajda, Andrey Nefedov and Andrej
Malchukov in “Impersonal constructions in Ket” (pp. 439-458) show how several of
the numerous intransitive conjugations in this endangered Yenisseyic language
well-known for its idiosyncratic verbal morphology can be accounted for as
involving impersonal (more precisely, transimpersonal) patterns using a frozen
3pl inanimate agent marker da-. “Impersonal verbs in Central Alaskan Yupik
(Eskimoan)” by Osahito Miyaoka (pp. 459-488) gives a comprehensive account of
how two kinds of impersonal verbs (simplex and derived) interact with a complex
system of valency changing derivations in Yupik. Simplex (“primary”) impersonals
in Yupik include meteorological verbs as well as predicates denoting colors and
qualities. The derived impersonal structure is created by the productive
necessitative suffix; interestingly, such complex predicates are undergoing
reanalysis and losing their impersonal morphosyntax. Lynn Drapeau in
“Impersonals in Innu” also draws a distinction between lexical and derived
impersonals in this Algonquian language with direct-inverse marking and an
important role of animacy distinctions in morphosyntax and verbal morphology.
The class of lexical impersonals in Innu, in addition to cross-linguistically
recurrent predicates denoting natural events, contains many lexemes which
correspond to nouns in the languages of Europe, e.g. expressing topographic
notions such as ‘bay’. A clear link between morphosyntactic impersonality and
pragmatic theticity as observed in Innu as well: “event-centered” statements not
forming a topic-comment structure are often expressed via impersonal structures,
and this is the main discourse function of the derived impersonals formed by
affixes, different for transitive and intransitive stems, suppressing the subject.
In contrast to many languages where impersonals are marked by special verbal
morphology, in Mandarin, as show Yi Yan and Anna Siewierska in “Referential
impersonal constructions in Mandarin” (pp. 547-580), subject/agent backgrounding
is manifested by use of various nominal strategies, such as special generic
nouns and pronouns, including zero pronouns, or a cross-linguistically
non-trivial construction involving an existential predicate and a generic noun.
Interestingly, the typologically common strategy of forming referential
impersonals with the help of 3Pl pronouns is uncommon in Mandarin. The paper
contains a useful table (p. 578) comparing Mandarin and English impersonal
The book is undoubtedly a welcome and useful contribution to language typology.
Impersonal constructions have not been subject to a detailed and comprehensive
cross-linguistic analysis before, and this volume successfully fills this gap.
The editors can be praised for having been able to establish a good balance
between descriptive and theoretical studies, as well as between synchronic and
diachronic perspectives. The cross-linguistic coverage of the volume is almost
comprehensive, with a slight bias towards Africa and Eurasia and a regrettable
lack of Papuan and South and Meso American languages. Inclusion of several
papers dealing with whole language families or areas instead of individual
languages, thus addressing issues of intra-genetic and areal typology, is also
an example to be followed.
Of the problems addressed in the volume that concerning the very definition of
“impersonal construction” is among the hardest to solve. From a eurocentric
perspective, usually reproached in contemporary typological studies, an
impersonal construction is one that lacks a “canonical” referential subject.
However, since the notion of subject itself has been rejected as not applicable
to all languages and not allowing for cross-linguistic identification (cf. e.g.
Dryer 1997, Haspelmath 2010), the typological validity of the notion “impersonal
construction” is subject to doubt, too. However, the editors and authors of the
volume have decided to assume this formal-syntactic definition of impersonal
construction and to test its applicability to different languages. The results
are very instructive, showing (i) that it is possible to identify constructions
with “defective” subjects or no subjects at all in typologically very different
languages and do so on a non-arbitrary basis, and (ii) that constructions so
identified show a striking degree of similarity across languages with respect to
the lexical-semantic types of predicates usually occurring in such constructions
(e.g. meteorological verbs or experiencer predicates) and discourse-pragmatic
functions associated with them (e.g. agent-defocusing and theticity). This
suggests that appropriately defined syntactic structures and morphosyntactic
patterns may indeed serve as basis for cross-linguistic comparison.
What the editors, however, to my regret have not done is to make the volume more
coherent. There are almost no mutual cross-references among the individual
contributions, even when the authors address similar issues. It would have been
also very useful if the volume contained an afterword summarizing the findings
of all the contributions and making typological generalizations about various
features of impersonal constructions.
The other serious drawback of this otherwise excellent book is a huge number of
typos and typesetting lapses, both in the text and in the examples. To mention
just one, ex. (5) on p. 58 (the only example from Lithuanian found in the whole
book) contains a typo (“mayt” instead of “matyt”) and three (sic!) wrong
glosses: “her” instead of “he:instrumental”, “was” instead if “will be” and
“complaining” instead of “complained”. Sometimes it looks as if the articles
have not been proof-read by the authors or editors.
Cabredo Hofherr, Patricia. 2006. ‘Arbitrary’ pro and the theory of pro-drop. In
Arguments and Agreement, P. Ackema, P. Brandt, M. Schoorlemmer, and F. Weerman
(eds.), 230-261. Oxford: Oxford University Press.
Creissels, Denis. 2007. Impersonal and anti-impersonal constructions: A
typological approach. Ms., Université Lyon 2.
Dryer, Matthew. 1997. Are grammatical relations universal? In Essays on Language
Function and Language Type: Dedicated to Talmy Givón, J.L. Bybee, J. Haiman,
S.A. Thompson (eds.), 115-143. Amsterdam, Philadelphia: John Benjamins.
Haspelmath, Martin. 2010. Comparative concepts and descriptive categories in
cross-linguistic studies. Language 86 (3): 663-687.
Holmberg, Anders. 2005. Is there a little pro? Evidence from Finnish. Linguistic
Inquiry 36 (4): 533-564.
Keenan, Edward L. 1975. Towards a universal definition of subject. In Subject
and Topic, Ch. Li (ed.), 303-333. New York etc.: Academic Press.
Lambert, P.-Y. 1998. L’impersonnel. In Actance at valence dans les langues de
l’Europe, J. Feuillet (éd.), 295-347. Berlin: Mouton de Gruyter.
Lehmann, Christian, Shin, Yong-Min, Verhoeven, Elisabeth. 2000. Person
Prominence and Relation Prominence. On the Typology of Syntactic Relations with
Particular Reference to Yucatec Maya. München, Newcastle: LINCOM Europa.
Malchukov, Andrej. 2008. Split intransitives, experiencer objects and
‘transimpersonal’ constructions: (re‑)establishing the connection. In Typology
of Semantic Alignment, M. Donohue and S. Wichman (eds.), 76-101. Oxford: Oxford
Mel’čuk, Igor. 1997. Grammatical cases, basic verbal construction, and voice in
Maasai: Towards a better analysis of the concepts. In Advances in Morphology, W.
Dressler, M. Prinzhorn & J. Rennison (eds.), 131-170. Berlin, New York: Mouton
Siewierska, Anna (ed.). 2008. Impersonal Constructions in Grammatical Theory.
Special Issue of Transactions of the Philological Society, 106 (2).
Timberlake, Alan. 1982. The impersonal passive in Lithuanian. In Proceedings of
the 8th Annual Meeting of the Berkeley Linguistics Society: 508--524.
Wiemer, Björn. 2006. Relations between Actor-demoting devices in Lithuanian. In
Passivization and Typology. Form and Function, W. Abraham, L. Leisiö (eds.),
274-309. Amsterdam, Philadelphia: John Benjamins.
| ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Peter M. Arkadiev, PhD in linguistics (2006), is a senior research fellow
in 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 a focus on case marking and argument
structure and its formal realization, and tense-aspect-modality. He works
mainly on Lithuanian and Adyghe.