Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  The Development of Latin Clause Structure

Reviewer: Bruno O. Maroneze
Book Title: The Development of Latin Clause Structure
Book Author: Lieven Danckaert
Publisher: Oxford University Press
Linguistic Field(s): Historical Linguistics
Subject Language(s): Latin
Issue Number: 30.2818

Discuss this Review
Help on Posting

“The development of Latin clause structure”, by Lieven Danckaert, is a study on Classical Latin word order and other related phenomena. It is the 24th volume of the “Oxford Studies in Diachronic and Historical Linguistics” series and, as described in the “Acknowledgements”, synthesizes five-years of research the author carried out between 2011 and 2016. Its 356 pages are divided in six chapters enriched with a glossary of important terms and three indexes (index locorum – where to find Latin quotations, author index and subject index).

The first (and longest) chapter (entitled “What is at stake”) presents the phenomena that will be analyzed as well as most theoretical concepts necessary to understand the analyses. As the author states, the book has two main goals: the first one is to describe and analyze two “word order alterations in the Latin clause, namely the variable distribution of the orders ‘object-verb’ (OV) and ‘verb-object’ (VO) […] and the alternation between the orders ‘non-finite verb-auxiliary’ (VAux) and ‘auxiliary-non-finite verb’ (AuxV)” (page 1). The second goal addresses “the question of which type of theoretical approach is best suited to describing and analysing facts of Latin word order” (page 3). The author then demonstrates the highly flexible Latin word order with many examples extracted from a corpus (pages 3-9) and presents an overview of the theoretical approaches that try to explain this flexibility (pages 9-22). Four groups of approaches are described: non-configurational, semi-configurational, hybrid and fully configurational. The author’s position is that the fully configurational approach is superior to the others, a statement that will be argued for in the remainder part of this chapter.

As part of this argument, the author intends to answer the question “does Latin have a VP constituent?” (page 30). Some pieces of evidence are brought up: the first one is the “Negation-Verb Ordering Restriction” (NegVOR for short), a generalization that Danckaert is probably the first to describe. This generalization goes as follows:

“In Latin, the marker of sentential negation ‘non’ always linearly precedes the hierarchically highest clause-mate verb, but it can either precede or follow all other verbs in the same clause.” (page 41).

Danckaert also proposes an explanation of this constraint in terms of phrase structure (pages 45-68) and, after that, discusses other pieces of evidence for a VP constituent in Latin, like the possibility of coordinating “strings consisting of a direct object and a dependent non-finite lexical verb” (page 68) and the existence of VP ellipsis and VP pronominalization in Latin, among others (pages 69-73). He concludes the chapter claiming that a configurational, phrase-structure approach to Latin syntax is superior to alternative approaches, and that structural ambiguity is an important issue to be taken into account.

The second chapter (“Latin corpus linguistics and the study of language change”) is dedicated to methodological issues. At first, the author argues for a statistical treatment of diachronic data (pages 79-83) and presents his corpus (pages 83-101), containing about 3,700,000 words and 39 texts, running from roughly 200 BC until 590 AD. In order to assure that the corpus is a reliable source for the study of Latin language change, Danckaert presents a case study unrelated to his main issue, namely the development of the future perfect passive: as it is known by Latinists, future perfect passive tense (‘I will have been loved’, for example) is formed by the be-auxiliary and the past participle; however, the tense of the be-auxiliary may be either the future ‘imperfect’ (like ‘amatus ero’) or the future perfect (like ‘amatus fuero’). In the first case, there is a mismatch between the tense of the be-auxiliary and the tense of the whole periphrasis; in the second case, there is no such tense mismatch. The first pattern is diachronically older than the second one and the data of the assembled corpus correctly map this change, showing that the corpus is indeed reliable as a source for diachronic studies.

Chapter Three (“Multiple object positions and how to diagnose them”) is dedicated mainly to showing that Latin clauses have multiple possible positions for the placement of the object. At first, the author presents statistical data on the alternation VO/OV in his corpus, showing that different syntactic environments yield different (and sometimes contradictory) results. He then presents the hypothesis that there are more than two possible positions for the object, that can be disambiguated in clauses with two verbs (one finite and another non-finite). When the order VOAux is counted as OV and the order AuxOV is counted as VO, for example, the results appear to show a diachronic change from OV to VO, which is as expected from the data of the Romance languages. As the author says, “[w]e can conclude that the OV/VO alternation is not simply a matter of objects either appearing preverbally or postverbally. Instead, […] a more fine-grained classification is required to arrive at an accurate description of the empirical data” (page 136). The remainder of the chapter is dedicated to the syntactic formalization (in X-bar terms) of the multiple object positions and of the clauses with modals and auxiliaries, as well as the formal and semantic differences between clauses with be-auxiliaries and with the various kinds of modals.

The fourth chapter (“VOAux – a typologically rare word order pattern”) continues to refine the description of object placement in Latin. Many graphics are presented which show frequencies of specific order patterns (object shift, object extraposition, different positions of modal verbs etc.), in order to test all possible contexts and verify which of them show more diachronic change. The author shows that a large discrepancy is found between VPAux and AuxVP clauses: in VPAux clauses, the order VO decreases very abruptly from ca. 100 AD; in AuxVP clauses, approximately the opposite is true: the order VO increases from about the same period (although not so abruptly). In order to explain that, Danckaert presents an account of the VOAux pattern in X-bar-theory terms, and, based on Latin examples, proposes a synchronic description of this pattern’s derivation. He concludes the chapter by discussing some aspects of the loss of VOAux in Late Latin, a subject that will be treated in the next chapters.

In Chapter 5 (“Changing EPP parameters – Clause structure in Classical and Late Latin”), Danckaert presents his main explanation of the change from Classical to Late Latin. He describes two grammar systems, called ‘Grammar A’ and ‘Grammar B’, which enter into competition at some point in the history of Latin. According to his hypothesis, there is the requirement so satisfy the “clausal EPP-requirement” (p. 216), a notion that is explained in pages 229-233. As such, in Grammar A, “VP movement takes place to satisfy this EPP-requirement” (p. 225), whereas, in Grammar B, “movement of the highest verbal head applies to perform the same function” (p. 225). Besides, the author also proposes that the main ‘cause’ of the shift from Grammar A to Grammar B was “incorporation of the negator ‘non’ into the hierarchically highest verb in the clause” (p. 268).

The sixth and last chapter (“The development of BE-periphrases”) aims at describing some of the differences between modal and be-periphrases. The author presents initially the two possible patterns of passive (and deponent) be-periphrases in Latin, which he refers to as ‘E-periphrases’ and ‘F-periphrases’ (respectively, in the infinitive, ‘amatus esse’ and ‘amatus fuisse’, both meaning ‘to be loved’). The F-pattern, despite being already attested in Plautus, increases its frequency over time and is felt as a ‘new pattern’. As the author shows, both patterns have different preferences in terms of word order: the E-pattern favors the order Past Participle + ‘sum’, while the F-pattern (although less frequent) favors the order ‘sum’ + Past Participle. As such, Danckaert hypothesizes that both periphrases have distinct diachronic developments. On pages 272 to 289, he then presents rich analyses of how both patterns have emerged diachronically and concludes that the shift from E- to F-periphrasis (which survives in Romance languages) is correlated to the general loss of the synthetic passive form.

A three-page epilogue closes the text, in which the author summarizes the main points of his whole argumentation and also presents an explanation of the reason that the order OV was eventually lost. The reader is referred to a submitted paper on this subject, but the main argument is that “in Grammar B it is more difficult for the VP-internal OV-order to be cued unambiguously than in Grammar A, a state of affairs that ultimately leads to the order VO taking over completely” (page 294). At the end of the book, there is a glossary of technical terms (pages 296-306) followed by the references and indices (pages 307-356).


“The development of Latin clause structure” is a generative-theoretic analysis of some aspects of Latin syntax, from a diachronic point of view. Danckaert develops his argumentation based on a very rich and consistent corpus of Latin texts, as well as a deep understanding of the theories required to explain the data. The author successfully manages to correlate data which do not seem to be related at first, such as the position of the negation and the order of object placement, and builds solid theories to accommodate these and other data. It should also be pointed out that the work is rich in statistical data, which prove to be very important in the author’s conclusions.

Using a generative-based approach to analyze a language without native speakers is certainly a huge challenge, especially because, methodologically, these approaches are usually based on native speakers’ intuitions. Danckaert does not address this issue in his book, instead assuming that the sentences attested in the corpus are grammatical. At first, I was expecting to find at least some justification for the use of corpus data in a generative-based analysis; instead, the chapter dedicated to methodology discusses the very important question on how to use statistics to describe variation and change. As I am not a ‘generative linguist’ (and, as such, I am now fully aware of more recent developments of this approach), I assume that this issue is already settled in the literature.

It is worth pointing that the book is mainly a theoretical, generative-based approach to Latin. Classicists and Latinists without an interest in or a good understanding of generative linguistic theory may not find the book so helpful. Although the reader will find in the text explanations for all the relevant concepts of generative theory (and the glossary at the end of the book is also helpful), one usually needs some familiarity with the theory and with the tree-like representations of sentence structure. An important exception, in my opinion, is Chapter Six, which develops a deep analysis of the two passive periphrases with very few theory-specific concepts.

Some minor problems could be pointed out, although they do not disqualify the work as a whole. Firstly, on pages 54 and 55, the concept of ‘fragment question’ (or ‘fragment why-question’) is referred to, without examples or discussion of what it specifically means. This concept could be explained in a footnote, for example, in order for the reader to better understand this passage.

Secondly, there are three minor typos and mistakes:

a) On page 47, where one reads “… which represents a fairly simply case of RM”, it should be “fairly simple”.

b) When discussing OV-order structure, on pages 183-184, one reads “The basic structure of a VO-clause would be as in (4)”. But since (4) shows an OV-structure, one should read, probably, “The basic structure of an OV-clause…”.

c) When discussing the stages in Jespersen’s Cycle (page 249), one reads “Early and Classical Latin would represent stage 1a”; but there is no stage 1a in the detailed structure of the cycle on page 249. The author probably meant stage 1’.

These minor issues pointed here do not diminish at all the importance of the book. Danckaert’s work is a very deep research with important analyses and is definitely a must-read for linguists interested in diachronic studies of Latin, especially those working on generative approaches.
Bruno Maroneze completed his Ph.D. in the University of Sao Paulo in 2011. His Ph.D. thesis focuses on Brazilian Portuguese neologisms formed by suffixation. His main research interests are on Lexicology, specifically word formation, neologisms and diachronic studies of the lexicon. He is currently teaching in the School of Communication, Arts and Letters of the Universidade Federal da Grande Dourados, MS,Brazil

Format: Hardback
ISBN-13: 9780198759522
Pages: 352
Prices: U.S. $ 90.00