Review of Latin Embedded Clauses
AUTHOR: Lieven Danckaert
TITLE: Latin Embedded Clauses
SUBTITLE: The Left Periphery
SERIES TITLE: Linguistik Aktuell/Linguistics Today 184
PUBLISHER: John Benjamins
Andreas Trotzke, Department of Linguistics, University of Konstanz
“Latin Embedded Clauses: The Left Periphery” by Lieven Danckaert (henceforth,
LD) deals with word order phenomena in Latin embedded clauses. The book focuses
on a subtype of embedded clauses, adverbial clauses (ACs), and investigates the
pattern in which one or more constituents occur to the left of a subordinating
conjunction. This monograph grew out of LD’s 2011 dissertation from Ghent
University. Since this study was supervised by both a Latin philologist and a
formal linguist, it aims at being accessible to readers not trained in formal
linguistics, as both the author and the publisher point out.
Chapter 1 (“Introduction”) specifies the topic and the methodology of the study.
LD first sketches earlier accounts of fronting elements to the left of a
subordinating conjunction, which he calls ‘Left Edge Fronting’ (LEF). He then
foreshadows his claim that there are two subtypes of this operation (LEF1 and
LEF2) with different syntactic and interpretive properties. After introducing
LEF as the main topic of the book, LD describes the corpus he studied, in which
texts from 180 BC to 120 AD were taken into account. He then proceeds by
pointing out that Latin is a discourse-configurational language, that is, he
sketches the claim that different Latin ordering patterns correspond to
different pragmatic interpretations. He then highlights the suitability of a
hierarchical approach to syntax for investigating these patterns and provides,
as an addendum, a ‘crash course’ in generative syntax, sketching basic concepts
like movement and locality.
Chapter 2 (“The internal syntax of Adverbial Clauses (ACs)”) provides an
introduction to the syntax of ACs and to Latin ACs in particular. LD presents
arguments in favor of an operator movement derivation of ACs. He then discusses
a class of Latin ACs which do allow for ‘Main Clause Phenomena’ (MCP), that is,
for operations that are available in root clauses but unacceptable or degraded
in embedded clauses. The contrast between ACs that allow for MCP (i.e.
peripheral ACs) and that do not allow for MCP (i.e. central ACs) is then
exemplified by a case study on the Latin particle *quidem*. LD provides an
analysis that correctly predicts that *quidem* cannot occur in the left
periphery of central ACs, while it can occur in the left periphery of peripheral
Chapter 3 (“The left periphery of embedded clauses”) presents a more detailed
picture of the decomposition of functional projections in the left periphery of
embedded clauses and provides an overview of the results of LD’s corpus study.
After sketching the approach of disjoining the clause-typing operator (placed in
the projection ForceP(hrase)) and the lexical element traditionally called
‘subordinating conjunction’ (placed in Fin(iteness)P), LD presents the results
of his corpus study on word order in Latin ACs. These results reveal that LEF
occurs most frequently in clause-initial ACs. Moreover, relative and
demonstrative pronouns are exclusively found in a LEF position in clause-initial
ACs. Based on this left-right asymmetry, LD formulates the distinction between
two types of LEF: pronoun fronting in initial ACs (LEF1) and XP-fronting in both
initial and final ACs (LEF2). This distinction sets the agenda for the rest of
the book; the syntax of LEF1 is analyzed in Chapters 4 (on relative pronouns)
and 5 (on demonstratives), and LEF2 is investigated in Chapters 6 and 7.
Chapter 4 (“The syntax of island pied-piping: Evidence from Latin relative
clauses”) focuses on the Latin phenomenon referred to as ‘Relative
Verschränkung.’ That is, LD examines a structure that contains a relative clause
S(entence)2, where the antecedent of the relative pronoun is located in the
superordinated clause S1, the extraction site of the relative pronoun is
embedded in S2 within another clause S3, and S3 occurs in a leftward position in
S2. As LD also briefly mentions in Chapter 3, the main property of this
structure is a left-right asymmetry. He argues that this asymmetry can be
captured if one assumes a type of topicalization involving clausal pied-piping.
Accordingly, the structure is derived in three steps: ‘internal (i.e.
non-terminal) movement’ of the pronoun to the edge of the embedded clause;
feature percolation; and clausal pied-piping, targeting the left periphery of
the superordinated clause. LD then extends this analysis to biclausal patterns
with a ‘relatif de liaison’ (i.e. with a wh-word introducing a non-integrated
non-restrictive relative clause) in the leftmost position.
Chapter 5 (“Clausal pied-piping by topics”) addresses the question if non-wh
topics can act as clausal pied-pipers as well. LD argues that the two patterns
of LEF (the pied-piper being a wh-element introducing either a restrictive or a
non-integrated non-restrictive relative clause) form a coherent class with
structures where demonstratives undergo LEF. In particular, he shows that
(independent or attributive) forms of the pronouns *is* and *hic* (which he
calls ‘IS-type pronouns’) can act, just like relative pronouns, as clausal
pied-pipers. He shows that these pronominal expressions receive a topical
interpretation and that the pattern they involve can best be analyzed by the
topicalization strategy that involves clausal pied-piping that he sketched in
Chapter 6 (“LEF2: Presentational foci in CP”) turns to cases where the LEF
constituent is neither a wh-expression (as discussed in Chapter 4) nor a
demonstrative pronoun (as discussed in Chapter 5). The crucial feature of this
LEF subtype (LEF2) is that it is also available in clause-final embedded
clauses. Based on the observation that non-(or less) referential elements can
undergo LEF2, LD argues that LEF2 should not be characterized as Germanic-type
scrambling. Based on the observation that non-specific elements (e.g.,
*aliquis*) or bare quantifiers (e.g., *omnia*) can undergo this movement, he
argues that LEF2 should not be characterized as the Romance-type topicalization
strategy of ‘Clitic Left Dislocation.’ After ruling out these options, LD claims
that LEF2 is an optional focalization strategy that alternates with marking the
focus in situ by prosodic stress.
Chapter 7 (“The syntax of LEF2: A synchronic and diachronic perspective”)
addresses the diachronic evolution of LEF2. LD points out the observed decline
of this phenomenon and argues that this is related to a change that took place
in the same period. More specifically, he shows that there is a correlation
between the loss of LEF2 and the increased frequency of VO word ordering
observed in the history of Classical Latin.
Given that there are very few studies on Latin word order that make use of
descriptive means provided by formal linguistics (cf., e.g., Devine & Stephens
2006), this book fills a gap in syntactic research and is a genuine contribution
to the field. As pointed out at the outset of this review, LD makes clear that
“the material in this book is meant to be relevant for classical philologists as
well as for formal syntacticians” (xv). As for classical philologists, the book
is written in a very clear and reader-friendly fashion, and both the general
introduction to formal syntax (28-51) and the introductory parts of each chapter
expand theoretical background in a subtle, non-overwhelming form, by
accompanying the presentation of theoretical claims with a healthy dose of
relevant data illustrating these claims. As for formal syntacticians, although
one need not agree with every detail of the proposed analyses and conclusions,
LD always discusses potential counterarguments and highlights what has to be
explored in future research to test his claims (e.g., Section 6 of Chapter 5 on
analyzing the left-right asymmetry in terms of Romance ‘Clitic Left Dislocation’
and the discussion of LEF as postsyntactic P(honetic) F(orm)-movement at the end
of Chapter 6). Moreover, the book is also suitable for scholars who draw their
linguistic expertise more heavily from Germanic languages. Instead of simply
cataloguing Latin examples, LD takes the time to discuss the results of his
corpus study in light of parallels to related phenomena in languages like Dutch,
English, and German, and this will no doubt be illuminating to many readers.
The overall organization of the book is stringent. As already summarized above,
after providing the necessary theoretical background and the distinction between
LEF1 and LEF2 in Chapters 1-3, Chapters 4-5 deal with LEF1 and Chapters 6-7
investigate LEF2. However, it would have been desirable (for the present
reviewer, at least) to have a general conclusion at the end of the book that
ties together all insights of the study. Since parts of the book are quite
inventorying (but show LD’s remarkable erudition), and since the intermediate
conclusions/summaries at the end of each chapter are sometimes rather short, the
direct impact that this work will make on the field might be considerably
strengthened by a future paper in a relevant journal that presents the results
in a more concise way.
These shortcomings aside, there is much to recommend about the book. The
empirical domain of ACs is very well chosen, allowing the reader to gain
fundamental insights on the overall structure of the Latin clause, since ACs
have the property of islandhood, show an equal left-right distribution, and, as
LD points out, are very frequently attested, and thus easy to retrieve in
corpora. Given this suitable empirical domain, LD adopts the framework of
cartography to analyze structures. Here lies one of the main strengths of this
book. The cartographic approach decomposes functional projections into simple
structural units. Thus, in the eyes of linguists working within this framework,
“[l]ocal simplicity is preserved by natural languages at the price of accepting
a higher global complexity, through the proliferation of structural units”
(Rizzi 2004: 8). As for LD’s focus on the clausal left periphery, it has been
pointed out in the literature that there is a fundamental problem of
presupposing discourse-related features (such as [foc(us)]) in the syntactic
representation when one is committed to the perspective of the Minimalist
Program. According to Chomsky (1995: 228), “any structure formed by the
computation [...] is constituted of elements already present in the lexical
items selected [...]; no new objects are added in the course of computation
apart from rearrangements of lexical properties.” In other words, this
‘Inclusiveness Condition’ implies that syntactic operations can only refer to
lexical features. Of course, lexical items cannot be viewed as inherently
focused etc. Consequently, such features, as Neeleman & Szendrői (2004: 155)
note, “must be inserted after an element has been taken from the lexicon,” and
thus the postulation of such features violates the minimalist ‘Inclusiveness
Condition.’ LD shows, however, that approaching syntactic structures (and
especially the clausal left periphery) from a cartographic perspective can be
incredibly fruitful. Since proponents of this approach are committed, by and
large, to a rigorous methodology of description, LD can rely on a large amount
of previous work and thereby refine our picture of the overall structure of the
Latin clause. Hence, a lesson that might be learned from LD’s work is that these
different strands of generative linguistics, focusing on different levels of
adequacy, should complement each other rather than being seen as incompatible
approaches. This is not a trivial issue, since many minimalist scholars, as the
reviewer has noted at various conferences, tend to prejudge an analysis when
recognizing that it is ‘hopelessly cartographic’ (making the term ‘cartographic’
a general verdict like – to exaggerate a little – the term ‘metaphysics’ was
used as an invective by the Vienna Circle in the early 20th century).
One aspect where cartographic research may help in gaining important insights is
the exact (interpretive) nature of the topicalization operation described by LD
in the context of LEF1. Clausal pied-piping has been argued to exist in a number
of languages, for instance, in Bavarian (cf. Bayer 2001). Since the main
advantage of this analysis is that it readily explains the left-right asymmetry,
LD adopts “a Bayer-style analysis to Latin topicalization examples” (238).
According to Bayer (2001), the movement type ‘Emphatic Topicalization’ (ET)
targets the specifier of a complementizer and prevents the embedded clause from
remaining in its base position. Accordingly, the entire embedded clause must
move to the left periphery of the clause that immediately dominates it. This
clausal pied-piping is necessary because, according to Bayer (2001), ET induces
the feature [etop] that can only be interpreted in root or root-like clauses.
Recently, it has also been demonstrated that ET exists in Bangla (cf. Bayer &
Dasgupta subm.). It is not yet clear if the interpretive properties of ET can be
captured in terms of information structure or if other notions related to the
‘expressive’ dimension of meaning (cf. Potts 2007) might be more appropriate in
this regard (like ‘emphasis,’ cf. Frey 2010, or ‘mirativity,’ cf. DeLancey 1997
and Cruschina 2011); cartographic work like LD’s helps make progress on these
issues and sheds further light on the interpretive properties of the clausal
Whether or not this book on the syntax of Latin is intelligible to readers
outside of formal linguistics remains to be seen. While it may perhaps be more
appropriate to be a little ‘theory savvy’ when approaching this book as a
reader, the author does a good job of clarifying formal ideas. In sum, the book
is very instructive in showing what modern syntactic research (within the
cartographic framework) can teach us about the structure of a linguistically
under-researched language like Latin.
Bayer, Josef. 2001. Asymmetry in emphatic topicalization. In Caroline Féry &
Wolfgang Sternefeld (eds.), Audiatur Vox Sapientiae: A Festschrift for Arnim von
Stechow, 15-47. Berlin: Akademie-Verlag.
Bayer, Josef & Probal Dasgupta. subm. Emphatic Topicalization and the structure
of the left periphery: Evidence from German and Bangla. Submitted to Syntax.
Chomsky, Noam. 1995. The Minimalist Program. Cambridge, Mass.: MIT Press.
Cruschina, Silvio. 2011. Discourse-Related Features and Functional Projections.
Oxford: Oxford University Press.
DeLancey, Scott. 1997. Mirativity: The grammatical marking of unexpected
information. Linguistic Typology 1, 33-52.
Devine, Andrew & Laurence Stephens. 2006. Latin Word Order: Structured Meaning
and Information. Oxford: Oxford University Press.
Frey, Werner. 2010. Ā-Movement and conventional implicatures: About the
grammatical encoding of emphasis in German. Lingua 120, 1416-1435.
Neeleman, Ad & Kriszta Szendrői. 2004. Superman sentences. Linguistic Inquiry
Potts, Christopher. 2007. Conventional implicatures: A distinguished class of
meanings. In Gillian Ramchand & Charles Reiss (eds.), The Oxford Handbook of
Linguistic Interfaces, 475-501. Oxford: Oxford University Press.
Rizzi, Luigi. 2004. On the cartography of syntactic structures. In Luigi Rizzi
(ed.), The Structure of CP and IP, 3-15. Oxford: Oxford University Press.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Andreas Trotzke is a postdoctoral scholar in the Department of Linguistics,
University of Konstanz. His research interests include left peripheral
syntax, the syntax-pragmatics interface, and the recursivity of syntactic
embedding. He is also interested in recent developments in biolinguistics
and their connections to performance-oriented linguistics.