This slender monograph packs a lot of weight. It had surprised me to learn, as Emonds notes near the beginning of the discussion, that there has not been a comprehensive analysis of the verb and auxiliary system of English since Chomsky (Reference Chomsky1957), despite English being one of the most studied languages in the world, if not the most. This volume offers a fresh outlook on the morphosyntax of English verbs and auxiliaries, providing a solid empirical foundation for categorizing verbs, auxiliaries and modals. Emonds draws some surprising and intriguing conclusions in the process, much of which stems from a long line of Emonds’ earlier work. As such, this volume serves as the culmination of several years of Emonds’ thinking on English grammar. For instance, he concludes that inflected forms such as is, am, are, etc. are not inflected forms of be as is traditionally assumed. This claim, however, has been made before (Becker Reference Becker2004). Emonds also discusses implications of this work for both theoretical linguistics and traditional grammarians, making suggestions for grammatical descriptions intended for a wider audience (such as language learners). Throughout, Emonds also makes reference to various non-standard varieties of English and other languages around the world. Such references are necessarily brief, as the topic he sets out to cover is already of an expansive enough range that any additional discussion would unduly lengthen the volume. He raises numerous such points in footnotes, which invite interested readers to explore the implications of his proposal in other dialects and languages. His discussion strikes the right balance between covering what is necessary and raising thoughtful points for future research. While maintaining a professional and academic tone throughout the discussion, Emonds’ writing style is engaging and even humorous at times. See in particular footnote 19 in the first chapter. Readers have much to learn from the excellent organization of the academic discussion in this book.
Chapter 1 (pp. 8–45) lays out the theoretical foundation for Emonds’ discussion, which is generally couched within a generative grammar framework and roughly follows the principles of Distributed Morphology. Emonds also makes use of some of his own formalisms developed in his previous work. One important formalism he uses is a ‘cancellation feature’, which overwrites the characteristic LF interpretation of an element. For example, he assumes that verbs have a characteristic active interpretation, thus a stative verb such as know has the category VØ, where the subscript null sign indicates that the characteristic LF interpretation of the verb (+active) is cancelled. Emonds recalls much early (and sometimes forgotten!) work on the auxiliary system in English. Of particular note here is the discussion on be-raising, which goes back to a class at MIT taught by Klima in 1966. The discussion here is particularly well articulated, as it delves into the history of the development of the ideas as told through the eyes of someone who lived through this development.
Chapter 2 (pp. 46–79) discusses the past tense feature of English. Emonds makes an intriguing proposal in which there is but a single past tense feature in English, with allomorphic variation between what is traditionally called -ed and -en. Here, Emonds lays out how the verbal morphology of English is mapped into a morphosyntactic structure. An important conclusion of the discussion is that simple past tense is inserted directly on V but may be alternatively realized on I if an auxiliary is present. He also proposes that past tense can be directly inserted in I, but it has a different LF realization. Compare the following:
(1) John hasn't eaten lunch.
(2) John hadn't eaten lunch.
While past tense on V refers to the Event Time, he proposes that past tense on I (inserted in I, not realized alternatively on I) refers to the Reference Time in the sense of Reichenbach (Reference Reichenbach1947). One interesting conclusion he draws is in regard to negation. His analysis requires that I and VP be sisters, so an intervening NegP is impossible. He argues that negation is essentially a prefix to V inserted at the word level. Interesting aspects of the discussion here also include an analysis of expressions such as had better and have got.
Chapter 3 (pp. 80–116) starts with an interesting question, namely: why is it that present tense -s and progressive -ing have no irregular forms, while the past tense in English famously has dozens of irregular forms? First, he notes that the irregular forms does, says and has are not open-class verbs. Rather, he proposes they are essentially functional items (labeling them ‘grammatical verbs’). He formulates a hypothesis for constraining irregular forms to account for these observations. Anything in the ‘neighbourhood’ of X can trigger irregular inflection on X. That is, any YP dominated by any of X's projections, but nothing internal to YP. φ-features are internal to DP, so cannot trigger irregular inflection on V. Harley & Bobaljik (Reference Harley, Bobaljik, Newell, Noonan, Piggott and Travis2017) show that the number features of the internal argument trigger suppletive forms in a number of verbs in Hiaki, including walk, wander, arrive, die, stand up and kill. It is unlikely (but not impossible) that all of these verbs are functional, requiring further modifications to Emonds’ proposal. Hopi and Zuni exhibit similar patterns. The remainder of this chapter discusses the consequences of his proposal for other aspects of English grammar, touching on relevant data in French and Czech.
Chapter 4 (pp. 117–55) discusses the many environments in which the morpheme -ing is found. He begins with the proposal that non-finite clauses in English lack IP. He acknowledges (in footnote 3) that a reviewer objects to this claim, but maintains that there is no evidence for an IP. He analyzes the infinitival to marker as a preposition. Those familiar with Emonds’ earlier work will recognize this claim. I would note that interrogative infinitives as in (3) and (4) are problematic for the bare VP analysis of non-finite clauses, however.
(3) John wonders what to eat.
(4) Mary told Bill which book to read.
Emonds does touch on interrogative non-finite clauses in chapter 5, but does not discuss complex forms such as those in (3) and (4). Emonds then goes on to describe the various environments in which -ing is found, focussing on the category of the -ing affixed forms. Here are two examples:
(5) a painting on the wall
(6) a pretty daring student
Emonds argues that these forms are lexical verbs with an -ing suffix. The resultant forms are complex heads with the following structures, respectively: [NV – N] and [A V – A]. I will not review all the structures Emonds proposes for the various non-finite forms discussed here, but do note that he uses a modified version of the Theta Criterion (what he calls Anti-Transitivity) to deal with forms such as see John eating an apple. Crucially, John is not thematically related to the verb see, requiring the novel structure he proposes. Again, in all cases, the non-finite forms are VPs lacking (IP). The external argument originates in VP, following the VP-internal subject hypothesis.
Chapter 5 (pp. 156–200) discusses the differences between gerunds and infinitives. Emonds notes a number of asymmetries between gerunds and infinitives that are seldom discussed in the literature. Here is a small sample of the intriguing data he discusses:
(7)
(a) They discussed visiting Mexico/*to visit Mexico.
(b) They discussed when to visit Mexico/*when visiting Mexico.
He also notes that while gerunds, infinitival and finite clauses can all serve as subjects and objects, only gerunds can serve as objects of a preposition. Based partly on the outcome of the previous chapter, Emonds argues that only gerunds are true DPs and, as such, only gerunds can serve as the object of a preposition. It is in this chapter that Emonds clarifies his analysis of raising and control. These also are VPs, with elements such as for and to (e.g. for John to open the door) as prepositions. No higher clausal functional projections (CP and IP) are present. He briefly revisits interrogative infinitivals such as the following:
(8) They wondered whether to dress up for the party.
Again, he assumes that whether is a preposition. It is unclear how verb raising facts in French infinitives (Pollock Reference Pollock1989) or how many of the landmark properties of control (Landau Reference Landau1999, Reference Landau2015) can be reconciled with the structure Emonds proposes here; however, that is simply fodder for future research. Despite these questions, one cannot deny the vast empirical landscape that Emonds covers in this chapter, and the simple, elegant proposal he has for it.
Chapter 6 (pp. 201–43) tackles passives in English – both verbal and adjectival. He argues against Bruening's (Reference Bruening2014) conflation of verbal and adjectival passives, maintaining the core distinction between the two as follows. Verbal passives have an underlying and inherent subject while adjectival passives do not. He also discusses what he calls an indirect passive, as in example (9). He states that previous discussions on the English passive ignore examples like those in (9), which are also called experiencer have constructions, but a previous analysis can be found (Ritter & Rosen Reference Ritter and Rosen1997).
(9) Kamila had the kitchen replastered.
Chapter 7, ‘The empirical basis of theoretical advance’ (pp. 244–52), is a brief summary of the monograph.
As mentioned, this book presents a well-organized discussion on English verbal morphology that is not to be ignored. There is a vast wealth of empirical coverage. The editing is highly polished, making for an enjoyable read. I would also note the useful indexing that accompanies this volume. Finding the relevant discussion of an author or a principle can be tricky in a conventional index. Here, the author provides separate indexes of (i) cited authors, (ii) definitions, principles and the like, and (iii) English lexical entries. I was pleased with this aspect of the editorial process that De Gruyter provided.
My problems with the volume are few and should not put anyone off reading it. The few tree diagrams that are found in this monograph are rather sloppily typeset. There exist now numerous ways of typesetting high-quality trees, which is something the editorial team or the publisher should help the author with. Nevertheless, this is a rather minor editorial flaw in an otherwise well-produced book.
Given the breadth of coverage in the monograph, I was surprised not to find any discussion on the go get construction, as in (10), especially given the numerous forays into colloquial versions of English and non-standard dialects.
(10) Every morning I go get a cup of coffee.
Such constructions are discussed extensively in Bjorkman (Reference Bjorkman2016), which appears in a highly visible journal. In fact, Bjorkman's (Reference Bjorkman2011) MIT dissertation does not figure anywhere in Emonds’ monograph, despite the fact that Bjorkman provides an extensive analysis of the English auxiliary system, so one would think her analysis would be addressed here. Like Emonds, Bjorkman provides a cross-linguistic discussion of her proposals to bolster her findings. She also discusses the have/be alternation found in some Germanic and Romance languages, as well as Old and Middle English, another property missing from Emonds’ monograph. It is natural that in a discussion as extensive as Emonds’ the occasional reference will be missed; however, a discussion as important as Bjorkman's should definitely have been addressed. I end with a reiteration that the shortcomings I discuss above in no way detract from the importance of Emonds’ contribution.
