Review of The Expression of Information Structure
EDITORS: Fiedler, Ines; Schwarz, Anne
TITLE: The Expression of Information Structure
SUBTITLE: A documentation of its diversity across Africa
SERIES: Typological Studies in Language 91
PUBLISHER: John Benjamins
Mahamane L. Abdoulaye, Abdou Moumouni University, Niamey, Niger
This book presents a selection of papers given at the 2005 Focus on African
Languages conference organized by the Collaborative Research Center for
Information Structure (University of Potsdam and Humboldt University, Berlin).
The book is the third on focus in African languages edited by the researchers of
this institution, which is at the forefront of focus studies in the world's
languages. The editors of the volume emphasize in their introduction that the
book explores a large spectrum of information structure notions and their
expression in a wide variety of African languages, some of which are
little-studied. Besides the introduction, there are 13 contributions discussing
about 37 languages from all four Greenbergian African phyla. The book has a
table of contents and language and subject indices.
In the first contribution, ''Information structure marking in Sandawe texts'' (pp.
1-34), Helen Eaton deals with the information structure notions of focus, topic,
contrast, and thematic prominence in Sandawe (presumed Khoisan, Tanzania, 40,000
speakers). The author refers to the definition of focus from Lambrecht (1994).
The data is drawn from texts of various genres written by literate native
speakers. Eaton shows that in Sandawe, information structure notions are
expressed mostly through morphemes and less through word order shift. There is
roughly a 3-way split in the marking of information structure depending on
whether the sentence is realis, imperative/subjunctive, or irrealis. The
clearest case of information structure marking happens in realis clauses where a
pronominal clitic (the nature and origin of which is not discussed) marks
lexical items (NPs, PPs, temporal adverbs) as focused (or rather, as contained
within the focus domain of the sentence (p. 10). However, with function words
(such as discourse adverbs or conjunctions), the pronominal clitic marks the
following information as thematically prominent. Finally, the pronominal clitic
also appears on conjunctions introducing narrative events, subjunctive verbs,
and repetitive actions. Despite this polyfunctionality, the pronominal marker is
about the only explicit focus marker in the language. For the subject noun of a
realis sentence, the author discusses a particle called subject focus, but the
role of this particle is debatable since the only relevant example given
concerns an independent pronoun, which can be interpreted as being emphasized
(in an ''as for him'' construction; cf. example 26, p. 18). Similarly debatable
(as the author herself recognizes, cf. p. 24) is the focus-marking role of word
order shift in imperative and subjunctive sentences, or the focus-marking role
of tone change in irrealis sentences. Overall, it is not clear how subject focus
is marked in this language.
In the second contribution, ''Topic and focus fields in Naki'' (pp. 35-67), Jeff
Good, based on question and answer elicitation data, claims that Naki (Bantoid,
Cameroon, about 4,000 speakers), an SVO language, has postverbal focalization.
The nature of information structure notions is not an issue in this contribution
and the author simply assumes the topic and focus definitions given in Lambrecht
(1994). Instead, the author deals with the proper structural analysis of the
postverbal position, which in Naki is targeted by focused elements (including
the subject, although this argument has an alternative cleft-like construction,
which was not discussed in the paper). The shift into postverbal position is,
for some TAMs, accompanied by a tonal change on the verb, and this tonal change
is sometimes the only formal difference between a basic SVO sentence (''the lion
killed the hunter'') and a subject focus OVS sentence (''the HUNTER killed the
lion''; cf. p. 47). That is, the postverbal shift of some constituent may induce
the direct object to move to preverbal position. Jeff Good proposes that Naki
information structure is articulated into two fields around the verb: A
preverbal topic field, with a freer word order, and a postverbal field, with a
much stricter word order (the focused item must be adjacent to the verb). The
author claims that the field approach is better than the cartographic approach
of the generative type that assumes that focal elements get moved along the tree
branches into a specific position (such as the Focus Phrase). In fact, the field
approach allows the author to seriously question the usefulness of the
grammatical notions of subject and direct object.
In the third contribution, ''The relation between focus and theticity in the Tuu
family'' (pp. 69-93), Tom Güldemann studies a cleft-like construction in four Tuu
(Southern Khoisan) languages: N|uu, |Xam (extinct), Western |Xoon, and Eastern
|Xoon. The information structure notion dealt with is simply ''focus'' as defined
by Dik (1997), whose overall focus typology the author adopts (in particular the
distinction between assertive focus and contrastive focus). A second major
theoretical reference for the author is Sasse's (1987) distinction between
categorical and thetic statements. Important for the paper also is the
distinction between entity-central and event-central thetic statements. In the
first type, entities are predicated to exist (''my sister died'') while in the
second type an event is presented (''[it is that] Mum is hitting me''). The author
worked with texts, naturally produced discourse, and also question and answer
elicitation. The four languages are basic SVO and the cleft construction takes
the focalized constituent to the beginning of the clause and marks it with a
focus particle that is related to an identification copula. What is remarkable
is that in all four languages, the cleft construction has a second distinct
function: It is also used to mark the subject of thetic sentences, which
normally are thought to express sentence focus (cf. Lambrecht 1987, 1994).
Therefore, the author rejects the notion of sentence focus and instead claims
that entity-central thetic statements are structured like constituent focus. In
one case, the default information structure (the categorical statement) is
disrupted to mark the prominence of a constituent (the focused term). In the
other case, the default information structure is also disrupted whereby the
thetic predicate is downgraded and the thetic subject foregrounded. So, given a
cleft sentence in the Tuu languages, only the semantic role of the focused
argument and discourse context can tell whether it is an ordinary focused
constituent or a thetic subject.
In the fourth contribution, ''Focus marking in Aghem: Syntax or semantics?'' (pp.
95-116), Larry Hyman revisits the well-known case of Aghem (Western Grassfields,
Cameroon). The information structure notions referred to include contrastive
focus, auxiliary focus, and inherent focus. The data, collected by elicitation,
is drawn mostly from previous publications on the subject by the author and his
associates. At first sight, Aghem seems well endowed when it comes to strategies
of marking focus. Aghem has basic SVO word order which can be interpreted as
expressing topic/comment articulation, sentence focus, verb focus, or direct
object focus. In fact, the position immediately after the verb is the general
focal position so that any constituent can be moved there to receive focus. In
this case, the direct object, if there is one, is moved away into the
defocalizing preverbal position. Wh-words, too, obligatorily go to the position
immediately after the verb. Aghem also has a distinct focus type, auxiliary
focus, which can be extrinsic or intrinsic. Extrinsic auxiliary focus stresses
the truth value of the proposition (cf. ''he DID eat fufu'') and is marked by
alternate forms of the present tense, today's past and the general past.
Intrinsic auxiliary focus concerns the negative and the imperative, which,
although they do not mark any particular item as focused, seem to forbid the
focus of any element in the position immediately after the verb. The author
proposes that negation and the imperative are inherently focused. Aghem also
seems to mark contrastive focus through changes in the internal syntax of NPs:
Focused NPs appear with a zero (non-overt) determiner, while non-focused NPs
appear with an overt determiner. In the literature on Aghem and related
languages, focused and non-focused NP constructions are referred to as the
A-form and the B-form, respectively. It is exactly this A-form/B-form
alternation that Hyman revisits in the paper to offer a new analysis. The author
proposes that the alternation cannot be accounted for in semantic/ functional
terms and that a purely formal account is necessary. According to Hyman, the
B-form (with overt determiner) appears in contexts where a covert (zero)
determiner would not be properly governed. The author goes further to suggest
that in every language he has looked at, the mismatch between focus semantics
and its expression (i.e., focus semantics without marker or focus marker without
focus semantics) is such that grammar mediation must be postulated (p. 110).
In the fifth contribution, ''On the obligatoriness of focus marking: Evidence
from Tar B'arma'' (pp. 117-144), Peggy Jacob, based on original data, presents
the first description of the information structure of Tar B'arma (Central
Sudanic, Southern Chad, 45,000 speakers). The information structure notions
mentioned are: Focus, topic, given, and the types of focus found in the typology
proposed by Dik (1997). She cites at least three definitions of the term
''focus''. Jacob is also the only author to have tried to define what information
structure is, which she says ''reflects the organisation of an utterance
according to the temporary state of knowledge of the interlocutors'' (p. 120).
The data is based on elicitation using a question and answer frame. The language
has a basic SVO word order. In this language, the subject stands out against all
other constituents in that when it is focused, it has an obligatory marking that
involves a (vacuous) fronting and a focus particle (not related to a copula).
All other constituents can be focused with or without marking depending on the
preceding question: If the wh-question has an in-situ wh-word, the answer, too,
will have an unmarked in-situ focused constituent. If the wh-question has a
fronted wh-word, then the answer will have a fronted focused constituent. This
leads Jacob to claim that focus marking in the language is controlled by
grammatical constraints (subject category and structure of preceding
wh-question). However, for this reviewer, this claim would need to be tested
with more naturalistic data where not all focused sentences are preceded by a
In the sixth contribution, ''Focalisation and defocalisation in Isu'' (pp.
145-163), Roland Kießling takes a new perspective in the analysis of the A-form
and B-form of Aghem by studying the corresponding forms in the related Isu
language (Western Ring, Cameroon). The information structure notions mentioned
in the paper are: Topic, focus, background, auxiliary focus, and defocalization.
The data are based on narrative texts complemented by elicited material. Like
Aghem, Isu has an S-Aux-V-O-X basic word order. The position immediately after
the verb takes focalized constituents, while the position immediately before the
verb takes defocalized constituents. For example, wh-words in questions and the
constituent answering them in the corresponding answers must appear in the
position immediately after the verb. However, Isu also has other focusing
strategies for auxiliary focus, predicate (comment) focus, sentence focus and
verb focus. These focus types may leave the direct object in the position
immediately after the verb (IAV), which then must be defocused. According to the
author, this defocalization process is achieved through the A-form/ B-form
alternation of the NPs. The A-form appears in a focused IAV position where the
noun has a class prefix. The B-form appears in non-focused positions (including
the IAV position when focus is shifted elsewhere in the clause). In the B-form,
the noun is followed by an enclitic made up of the class marker and a
defocalizing morpheme. Taking the A-form to be basic, the author claims that Isu
in fact ends up with a mismatch of markedness relations: The pragmatically
neutral B-form is marked (as defocalized), while the pragmatically charged
A-form is left unmarked (at least in a diachronic perspective).
In the seventh contribution, ''Discourse function of inverted passives in
Makua-Marevone narratives'' (pp. 165-192), Oliver Kröger studies the information
structure in three narrative texts in Makua-Marevone (Bantou, Mozambique). The
notions mentioned by the author, drawn from information structure studies and
text analysis, are: Topic, focus, assertion, theticity, presupposition,
prominence, identifiability, activation, etc. However, the main point of the
paper is the discourse function of the inverted passive construction. The
theoretical frameworks are Lambrecht's (1994) model of information structure
(with its two dimensions of relations (topic/ focus) and references (activation
statuses), Givón's (1984) model of prominence scale, and Dooley and Levinsohn's
(2000) model of participant and prop roles in narratives. Makua-Marevone has
basic SVO word order. In this configuration, a narrative subject, usually an
animate referent, has high prominence (it is frequent and active throughout the
narrative), a direct object has high prominence if animate but low prominence if
inanimate, while adjuncts are low prominence no matter their animacy.
Makua-Marevone also has a postposed subject in a VS construction that expresses
thetic statements where the subject still has a high prominence. Makua-Marevone
further has a passive where the patient becomes subject/ topic with a
semi-active status. Finally, Makua-Marevone has an inverted passive
construction, i.e., the combination of passive and subject postposing, which is
used to promote inanimate referents (normally low prominence) to postposed
subject position, where they acquire high prominence. The author concludes that
the inverted passive, by violating one of the principles of the narrative script
(''inanimate referents are props''), alerts the listener to the unexpected role of
In the eighth contribution, ''Topic-focus articulation in Taqbaylit and Tashelhit
Berber'' (pp. 193-232), Amina Mettouchi and Axel Fleisch compare emergent
discourse-configurationality in Taqbaylit (Berber, Algeria, 5 million speakers)
and Tashelhit (Berber, Morocco). The information structure notions mentioned are
emphasis and contrast and the usual concepts of topic, focus, thetic,
categorical, argument focus, etc. Although the authors referred to Lambrecht and
Sasse, they did give their own definition of argument focus, which is ''a type of
emphasis that singles out one particular constituent and contrasts it with
conceivable alternatives. The corresponding construction assigns a new
information status to the focused constituent, combined with a notion of
counterexpectation''. As we will see in the evaluation section, as convoluted as
this definition may appear, it represents a welcome departure from the standard
definitions of focus. Finally, in their discussion of clause structure, the
authors borrow terms from Role and Reference Grammar (cf. Foley and Van Valin
1984), though the framework is not explicitly cited. The data for Taqbaylit was
drawn from spontaneous speech in various genres while a narrative text was used
for Tashelhit, complemented with elicitation. Taqbaylit and Tashelhit are basic
VSO languages that admit word order variations to signal information structure
(though variation is more restricted in Tashelhit). Taqbaylit in fact has a
pronominal argument next to the verb fulfilling the syntactic role, while
eventual co-referring lexical NPs fulfill the reference role and are positioned
depending on information structure requirements. The (arguably) basic VS/VSO
order usually conveys thetic statements (subject is not topic). Predicate focus
(= topic/ comment, = categorical statement) is expressed through the SV/SVO word
order with emphasis or contrast on the topical subject. Argument focus,
including subject focus, is expressed through a cleft construction with emphasis
or contrast on the clefted NP. The SVO order, which involves topicalization, is
well distinct from the cleft structure, which uses a marker next to the fronted
NPs and a relative clause for the rest of the proposition. Tashelhit, too,
expresses thetic statements with VS/VSO order and allows contrastive
topicalization on NPs in SV/SVO order, though in a more restricted way. For
argument focus, it uses a distinct cleft construction. Despite the general
cross-linguistic tendencies in this regard, the authors claim the processes
fronting lexical NPs did not lead to a VSO to SVO word order shift and that in
the two languages word order variations only code information structure.
In the ninth contribution, ''Focus in Atlantic languages'' (pp. 233-260), Stéphane
Robert, based on the study of 17 Atlantic languages (West Africa), shows that
focus expression can mesh with almost every aspect of verb morphology. The
information structure notions cited are: Rheme, focus, verb focus, argument
focus, etc. Robert puts forth her own theory about the nature of focus and
defines the rheme as the ''informative part of an utterance'' (p. 239). The rheme
can be a ''focus'' only if it ''corresponds to a syntactic constituent of the
sentence and is morphologically marked'' (p. 234, 248). Working with 17
languages, the author relied on data from published sources, including her own
previous works on some of the languages. Atlantic languages vary in the extent
to which they bind focus with verb morphology. On the one hand, the Mey language
for example has four conjugations for focusing verbs, subjects, objects, and
circumstantial phrases. Fula binds focus marking with TAMs and diathesis, so
that for example the markers coding perfective and argument focus are: -i for
active voice, -ii for middle voice, and -aa for passive voice (i.e., -i would be
glossed 'perfective, argument focus, active voice'). In Wolof, on the other
hand, an agreement pronoun codes information structure and TAM categories, so
that the particle ''la'' is glossed as 'perfective, non-subject focus, subject
agreement pronoun'. The focused NP is also fronted. Joola and Seereer defocus
the verb to indicate subject focus. Finally, some languages do not use verbal
morphology at all and resort to particles and pronouns to express focus.
Globally, there is a continuum between strongly morphological systems and
analytical systems. The situation in Atlantic languages leads the author to
reconsider the structure and function of the split focused proposition (such as
''it was John that we saw yesterday''). She claims that the assertive part of the
sentence codes identification and qualitative designation, while the presupposed
part codes a temporal relation and the existence of the subject (or the focused
element in general). This for her explains why in Atlantic languages (which have
no cleft construction), focus expression mixes with verbal morphology and why
focalization takes supplemental values such as explanation (of states of
affairs) or the intensification of the verb action.
In the tenth contribution, ''Topic and focus construction asymmetry'' (pp.
261-286), Ronald Schaefer and Francis Egbokhare contrast the grammatical
properties of topic and focus constructions in Emai (Benue-Congo, Nigeria). The
information structure notions cited are topic, focus, and shared information.
The authors adopt the notion of cognitive files for events and participants
developed by Givón (1983) and Du Bois (1987). In the context of a topic
construction, the participant file is shared between speaker and hearer but not
the main clause event file, which only the speaker holds. In the context of a
focus construction, the event file is shared but not the participant file (p.
264). The authors set out to show that these features account for most of the
differences observed between topic and focus constructions. They use data from a
texts collection and an extended elicitation carried out while writing a
dictionary and a grammar. The authors show that the main clause of topic
constructions, by virtue of being speaker-only knowledge, allows the imperative,
the hortative, and various particles (such as 'after all, of course, mistakenly,
a lot, a bit, etc.') that manipulate the truth value or the intensity of the
event. The main clause of focus structures, which conveys shared information,
resists such manipulations. However, differences (or similarities) in the
resumptive strategies for topic and focus NPs in the sentence could not entirely
be linked to the information structure notion of shared/ non-shared information
(the focus construction, as one may expect, has a preference for zero pronoun in
the main clause). On the side of the NPs, the authors show that the topic
position allows only definite nouns, allows partitive ''some'' and alternative
''another'', but rejects emphatic reflexives, restrictive ''alone'', or the ''of that
kind'' and ''of different kind'' modifiers. The focus position behaves
contrastively with all these items. The authors conclude that information
structure indeed influences the grammatical form of NPs and clauses. This
overall conclusion is reasonable with regard to the Emai data but, as one may
expect, the correlations described between topic and focus statuses and
grammatical form may or may not carry over into other languages. For example,
the English translations of many of the starred Emai sentences are fine. Also,
the authors' characterization of topic and focus applies only to the core cases
(for example, in ''do you like beans?'' / ''BEANS I like'', the NP and the main
clause in the reply are all shared information).
In the eleventh contribution, ''Verb-and-predication focus markers in Gur'' (pp.
287-314), Anne Schwarz tracks the information structure usage of cognate ‘mE’
particles in four Gur languages (West Africa): Konni, Buli, Dagbari, and Gurene.
The information structure notions cited are: Emphasis, contrast, truth value
focus, operator focus, etc. The author cites Dik's (1997) definition of focus
and Hyman and Watters' (1984) distinction between information focus and
contrastive focus. The data was gathered through the administration of a
questionnaire designed to test the conditions of the appearance of the particle
‘mE’ in the languages. All of the languages have basic SVO word order and word
order change is not used for information structure marking. Instead, the
languages rely on particles for the various focus types, including verb and verb
operator focus. Despite some variability, the author shows that globally the
cognate particles are exclusively used for focusing the lexical semantics of the
verb ('I CLEANED the oranges'), the truth value of the verb action or the TAM
operators ('I DID clean the oranges'). More crucially, in the opinion of this
reviewer, the particles seem especially favored for marking emphasis and
contrast, i.e., in the four languages, the use of the particles is increased in
contexts of high emphasis and contrast (for example, there is less restriction
on the TAMs that admit the particle). In a context of weaker emphasis and
contrast, some of the languages actually resort to other means of marking focus.
In the twelfth contribution, ''Why contrast matters: Information structure in
Gawwada (East Cushitic)'' (pp. 315-348), Mauro Tosco proposes a new analysis of
an information structure particle that can apply both to topic and focus NPs.
The information structure concepts dealt with are: Contrast, topic, and focus.
Overall, the system and definitions found in Lambrecht (1994) are adopted. The
author says that he purposely used narrative texts as data source to break from
the tradition of studying focus in the confines of the wh-question and answer
context. Gawwada has a basic SOV word order where the verb and the pronominal
arguments constitute the verbal group (which also has the SOV configuration).
Lexical NPs, if specified, are in the periphery of the clause. The basic SOV
word order expresses the topic/ comment articulation, while an alternate OSV
order expresses thetic statements. However, the most significant aspect of
information structure marking in Gawwada is the fact that ''focus'' as such seems
not to be marked at all. Instead, a particle -kka/k, which would be closest to a
focus marker, applies both to topic and focus NPs and can be analyzed simply as
a contrast marker. Given this situation, Tosco, so to speak, rescues the notion
of contrast as a linguistic category, contra, for example, Lambrecht (1994), who
dismisses it as being an effect induced by conversational implicatures.
In the thirteenth and last contribution, ''Focus and the Ejagham verb system''
(pp. 349-375), John Watters starts by proposing a complete typology of focus
structures that combines two dimensions: (i) The scope of focus, i.e., the
element under focus (a term, the verb, or a sentence operator such as truth
value and TAMs); and (ii) the communicative point of focus, whereby assertive
focus (a simple declarative statement) is distinguished from contrastive focus
(a complex form of information). This gives a set of six types of focus, a
subset of which the author claims would be instantiated in any one language. The
contribution is based on elicited material usually in the context of
wh-questions and their answers. According to Watters, Ejagham (Ekoid Bantu,
Cameroon and Nigeria, about 170,000 speakers), in particular its western
dialect, has two forms of the perfective and imperfective TAMs. One form, called
Operator Focus (=auxiliary focus) Form, appears in neutral assertive (topic/
comment) structure, in narrative event lines, and when the truth value or the
perfective/ imperfective semantics of the clause is contrasted (cf. 'he DID eat
the yams', 'he IS [NOW] eating the yams'). The other form, called the
Constituent Focus Form, appears when the lexical semantics of the verb or one of
its terms are focused. However, the Constituent Focus Form also appears in
relative clauses, cleft structures, wh-questions, and in answers following ''what
happened''-questions (which in many languages have a clefted form, cf. French
''que se passe-t-il?''; ''C’est Ali qui frappe Salif'' 'it is Ali [who is] hitting
Salif'; cf. Güldemann p. 88). Finally, the author shows that TAMs other than
perfective and imperfective have only one form each and that they do not mark
focus. Given such a situation, this reviewer thinks that Ejagham essentially
does not directly mark the focused constituent itself but rather marks the
presupposed (out-of-focus) part of the sentence. Indeed, many languages mark
both the focused material (for example a clefted NP) and the presupposed
material (for example the relative clause in English cleft sentences). Ejagham
seems to have lost the focus markers and retained only the presupposition
markers, and this, too, is restricted to the perfective and imperfective. Hausa
has a comparable situation where, although focus material is overtly marked with
fronting and a copula, special perfective and imperfective forms mark
presupposition in nearly the same contexts as in Ejagham (cf. Abdoulaye 2007).
The papers are written by specialists on the respective languages, mostly using
data they themselves gathered. They handle a variety of phenomena using diverse
approaches. This surely has the advantage of bringing a wide coverage, but it
also has drawbacks. For example, in the entire book there is only one place
where the notion of information structure is defined (Jacob, p. 120). Without an
agreed-upon definition, however tentative, of the functional domain, it will be
difficult to seek the relevant linguistic expressions. This is a key procedure,
for example, in typological studies (see Stassen 1997). One issue with focus
studies, which the book did not overcome, is the multiple terminologies
sometimes used to refer to the same phenomena (witness for example Güldemann's
attempt to reconcile Lambrecht's and Sasse's typologies, p. 86, where he agrees
with both on certain points and disagrees with both on other points). There is
even a point where standard terminology can get into the way of the analysis.
For example, the notion of focus seems to be intractable. Jacob, besides her
interesting definition of information structure, tries to fit together at least
three different definitions of ''focus'': (i) Information in a sentence speaker
assumes listener does not share (i.e., opposed to presupposition, Jackendoff
1972); (ii) most significant or salient information in the clause (Dik 1997);
and (iii) a category involving ''the presence of alternatives that are relevant
for the interpretation of the linguistic expressions'' (Krifka 2007), a
definition that equates focus with contrast. And there are many other
definitions of focus in the book and elsewhere. Clearly one must now throw away
the notion and the term “focus” and concentrate on simpler, more definable
notions, the expressions of which can then be sought in the languages. There are
three promising attempts in the book in this regard and all of them bring forth
the same concepts: Emphasis and contrast. Indeed, in one way or another, the
contributions by Mettouchi and Fleisch, Schwarz, and Tosco underline the
importance of emphasis and contrast (elsewhere, see Caron 2000:34, Abdoulaye
2006:1163, 2007). The three contributions show that these are the categories
that regularly get expressed in remarkable ways in the languages studied.
Abdoulaye, Mahamane L. (2006) Existential and possessive predications in Hausa.
Linguistics 44: 1121-1164.
Abdoulaye, Mahamane L. (2007) Profiling and identification in Hausa. Journal of
Pragmatics 39: 232-269.
Caron, Bernard (2000) Assertion et preconstruit: topicalisation et focalisation
dans les langues africaines. In: B. Caron, Ed., Topicalisation et focalisation
dans les langues africaines, pp. 7-42. Louvain/ Paris: Peters.
Dik, Simon C. (1997) The theory of functional grammar, part 1: The structure of
the clause. Berlin/ New York: Mouton de Gruyter.
Dooley, Robert and Stephen H. Levinsohn (2000) Analyzing discourse: A manual of
basic concepts. Dallas: SIL International.
Du Bois, John W. (1987) The discourse basis of ergativity. Language 63: 805-855.
Foley, William A. and Robert D. Van Valin, Jr. (1984) Functional syntax and
universal grammar. Cambridge: Cambridge University Press.
Givón, Talmy (1983) Topic continuity in discourse: An introduction. In: T.
Givón, Ed., Topic continuity in discourse: A quantitative cross-linguistic
study, pp. 1-41. Amsterdam/ Philadelphia: John Benjamins.
Givón, Talmy (1984) Syntax: A functional-typological introduction, vol. 1.
Amsterdam/ Philadelphia: John Benjamins.
Hyman, L.M. and John R. Watters (1984) Auxiliary focus. Studies in African
Linguistics Supplement 15: 233-273.
Jackendoff, Ray (1972) Semantic interpretation in Generative Grammar. Cambridge,
MA: MIT Press.
Krifka, Manfred (2007) Basic notions of information structure. In: C. Fery et
al., Eds., Interdisciplinary Studies on Information Structure (ISIS) 6: 13-56.
Lambrecht, Knut (1987) Sentence focus, information structure, and the
thetic-categorical distinction. Berkeley Linguistics Society 13: 366-382.
Lambrecht, Knut (1994) Information structure and sentence form: Topic, focus and
the representation of discourse referents. Cambridge: Cambridge University Press.
Sasse, Hans-Jürgen (1987) The thetic/ categorical distinction revisited.
Linguistics 25: 511-580.
Stassen, Leon (1997) Intransitive Predication. Oxford: Clarendon Press.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Mahamane L. Abdoulaye teaches linguistics at the Abdou Moumouni University,
Niamey. His main research focuses on Hausa and Zarma Chiine morphology,
syntax, and semantics.