This volume begins with an introduction titled “Alternations in Co- and Subordinations” that describes the phenomena and the approaches to them used by the contributors, followed by a look at the challenges they present. Each of the twelve chapters is then outlined in the space of three pages. In this review, I do roughly the same, but with the addition of a few critical remarks and some data.

The first chapter, “Marking (Not-)at-Issue Content by Using Verb Order Variation in German” by Mailin Antomo, focuses on a question addressed in various ways throughout this volume: What licenses the occurrence of dependent V2 clauses? One tradition follows Hooper & Thompson Reference Hooper and Thompson1973: The difference between dependent verb-final and V2 clauses is often the difference between the status as an assertion or a presupposition. Antomo discusses an alternative, the At-issueness Hypothesis, to account for dependent V2 clauses in German. He comes to the conclusion that there are two types of dependent clauses: those that can be used to express at-issue content and those that cannot.

Antomo’s analysis reveals that determining whether or not a dependent V2 clause meets the criteria for expressing at-issue content is not a simple matter. Some questions remain unanswered under his hypothesis and must be left for further research. In the end, he concludes that to a great extent his hypothesis and the Hooper & Thompson thesis based on assertion make the same predictions; his, he argues, yields better coverage since it accounts for the ban on V2 in nonpresupposed contexts such as appositive relative clauses and asserted free dass-clauses. While the analysis in Antomo’s chapter sheds new light on important issues, it should have been read carefully by a qualified native speaker of English. There are many usage and punctuation errors that detract from the context and mar the presentation.

The next chapter, “The Decline of Asymmetric Word Order in Cimbrian Subordination and the Special Case of Umbrómm” by Ermenegildo Bidese & Alessandra Tomaselli, addresses the evolution in Cimbrian subordinate clauses away from the asymmetric finite verb position. In contrast to German, Cimbrian has lost the V2 restriction except with some pronominal forms. Most subordinate clauses allow verb raising to some higher position, as indicated by the position of the negative particle nèt ‘not’ (post- versus preverbal) and the target of enclisis, either the subordinating conjunction or the finite verb, as in 1a. Example 1b exemplifies the main clause pattern.

1. (1)

   

The complementizer umbrómm, which can introduce both adverbial and interrogative clauses, presents a special case because it requires the root, declarative word order:

1. (2)

   

The goals of Bidese & Tomaselli’s study are to determine i) how far the decline of the root-embedded word order asymmetry has developed and in which syntactic contexts, and ii) to what extent this tendency can be considered a general internal development rather than a contact-induced phenomenon. They come to four conclusions: i) the decline of the asymmetric word order involves the declarative modality, that is, the [-wh] feature, which induces a process that does not require the borrowing of Italian word order; ii) the decline of the asymmetric word order in embedded contexts comes by way of the extension of Cimbrian root declarative word order; iii) in the special case of umbrómm, the root declarative word order extends all the way to wh-clauses; iv) asymmetric word order still holds in the embedded context with az ‘that’ and some other complementizers, with the exception of umbrómm; the reasons for that are suggested by other research (for example, Moro Reference Moro2011), namely, this interrogative adverb is base-generated in the C-layer and thus does not undergo movement and has a special status with respect to the wh-criterion. I found this chapter to be very well constructed and argued, with a good deal of persuasive data. The limited scope chosen by the authors forces them to leave several questions unanswered. While this situation may be unsatisfying to some, it opens an opportunity for other investigations with interesting data and insights.

The third chapter, “(Anti-)Control in German: Evidence from Comparative, Corpus- and Psycholinguistic Studies” by Patrick Brandt, Beata Trawiński, & Angelika Wöllstein, investigates types of control relations between clauses and attempts to find support for the hypothesis that certain infinitival clauses have a CP- and not just a vP- or VP-structure. Finding this kind of evidence is particularly difficult in anti-control relations because complements of anti-control verbs are almost exclusively finite (versus nonfinite/infinitival) clauses introduced by a complementizer; thus, these clauses are CPs, which require no control. The difficulty of finding evidence is even greater because incoherent construal of the control relation is always possible and a coherent construal is only possible if the infinitive appears in the middle field, as in 3.

1. (3)

   

In their approach, Brandt, Trawiński, & Wöllstein compare German anti-control relations to those in Polish constructions with the complementizer źeby, which disallows control (that is, it requires anti-control). The authors argue logically that anti-control infinitives construe only incoherently and thus would not likely occur in the middle field. They conclude that, “[i]f infinitival anti-control complements in German are C-projections as their Polish cousins clearly seem to be, these differences follow straightforwardly” (p. 94). They caution, however, that the results of their judgment tasks require further testing because independent factors could be the cause.

The next chapter, “On the Question of Subordination or Coordination in V2-Relatives in German” by Nicholas Catasso & Roland Hinterhölzl, takes up a question that has a long history. It has generated a lot of interest in recent years, possibly because it presents an excellent challenge to researchers working on the interfaces (which I refer to here as simply PF and LF) and on their interaction. Although Catasso & Hinterhölzl do not explicitly frame their solution in those terms—possibly to avoid appearances of bias against any one framework—the minimalist assumption that the integration of PF and LF plays a crucial role in the derivation of the structures may be implied.

The authors’ account of V2 relatives preserves the syntactic basis of the relation between root and relative clauses in that it assumes a hypotactic analysis. For this analysis to work, a silent relative operator must be the head of the relative clause. The V2 status of the relative clause does not undermine the root-embedded/hypotactic relation because the verb-raising operation assures that the embedded event is epistemically anchored to the speaker, that is, that the embedded proposition is asserted, which is a core property of V2 clauses. The V2-status is thus compatible with the LF property of assertiveness, that is, the LF requirement on the syntax is met with this strategy. The V2 relative must be compatible with the LF requirement that the relative clause be a comment on the main clause. To meet this requirement, extraposition of the V2 relative is required. The authors state the following:

There is a problem inherent in this approach: It seems to impede the restrictive interpretation of the relative clause—this is expected from its DP-internal position. To overcome this problem, the authors propose a matching operation that identifies the event arguments in the two clauses as denoting the same set of objects. This outcome, in turn, licenses the phonological deletion of the lower NP under semantic identity; furthermore, this outcome is compatible with the extraposition of the V2 relative and derives for free the determiner restrictions on the V2 relative head noun.

This clever proposal employs further argumentation and data, including from Italian, that are too complex and extensive to be outlined here. Yet it leaves one question open: How does extraposition of the V2 relative proceed? A more detailed presentation of extraposition could possibly provide the “final touch” needed for this analysis to be fully convincing.

The next chapter, by Christian Fortmann titled “Da Capo je-desto—On the Comparative Conditional Construction in German”, deals with the comparative conditional construction most typically formed with je and desto, as shown in 4a. Fortmann presents a wealth of data and analyses in support of an approach that does not require the phrases introduced by je and desto to be clausal or left-positioned in the sentence. In the most typical analysis shown in 4b, left-dislocation is required to reposition the je-phrase in the Spec, CP.

1. (4)

   

Fortmann argues that 4b takes into account certain properties of this construction, but not the “constitutive” ones. In his investigation, he compares this construction to relative clauses and identifies a number of parallels, following work of Speyer (Reference Speyer2011). For instance, the je … desto-construction does not have to be clausal, nor does it have to occur at the left edge of a sentence.

Fortmann’s analysis relies on some assumptions that might be questionable to some syntacticians. First, it requires head movement. Second, Fortmann must posit the category KorrP (correlative phrase). However, he does not present any independent evidence for the existence of this category, leaving it rather ad hoc. Another weakness of his presentation is not directly related to the content but certainly affects its impact: There are several minor English usage problems and typos (here the editors should have been more helpful); in addition, at least two sentences cannot be clearly parsed. These are unfortunate problems for a study that otherwise has much to offer.

The next chapter, “On Some Correlations Between Interpretative and Formal Properties of Causal Clauses” by Werner Frey, deals with three types of causal clauses: i) central adverbial clauses (CAC), ii) peripheral adverbial clauses (PAC), and iii) nonintegrated dependent clauses (NonIC). The last of these are typically introduced by weil ‘because’ (both the verb-final and V2 varieties) and da ‘since’._.Frey presents detailed arguments based on a plethora of data, including prosodic properties, for the proposal that there is one-to-one mapping between the syntax of three types of causal clause and the semantic type each represents, depending on both the syntax and prosody. He claims that a CAC is base-generated inside its licensing clause in a low position and refers to a relation between eventualities. In contrast, a PAC is base-generated inside its licensing clause in a high position and refers to a relation between propositions. Finally, a NonIC is a syntactic orphan; it refers to a relation between speech acts.

There are further claims that are too subtle and complex to outline here, relating to illocutionary force and possible interpretations. In summary, Frey’s study excels at exploring the intricate possibilities of meaning and implication, and how they relate to the underlying and surface syntactic properties. I recommend it highly for those working in these areas.

In the next chapter, titled “Clause Integration and Verb Position in German”, Ulrike Freywald also focuses on types of embedded clauses and the degree of their integration with the main clause. She argues that the relative types of embedded clauses can be identified based on the position of the finite verb in the embedded clause (verb final or V2). Recall that under Frey’s analysis, the distinction is based on where the embedded clause is generated. Freywald does not discuss derivational aspects in any great detail, though crucially, she assumes a cartographic model (Rizzi Reference Rizzi and Haegeman1997, Haegemann Reference Haegeman2010, among others). Her central thesis is that the varying degrees of integration can be determined just on the basis of the “external syntax”, that is, the position of an embedded clause within the host structure. The higher the position of the embedded clause (within the host clause), the less integrated it will be: a CAC adjoins in the VP, a PAC in the TP, and a nonintegrated adverbial clause in CP. She points out that only embedded clauses that are nonintegrated and independent, with V2 and a nonhypotactic structure, can have an expanded left periphery; a complementizer may be present as long as it allows verb raising (typical only of spoken usage).

Following Freywald’s proposal, the dass-clause in 5a is within the VP, whereas in 5b, it is within a FocP, with dass itself functioning as an “illocution-marking particle” (p. 181). Freywald points out that the prosody of these two constructions differs markedly: If the dass-clause in 5b lacks the prosody of an independent assertion, it is ungrammatical.

1. (5)

   

To strengthen her argument that subordinate clauses may have the status of independent utterances, she provides examples with modal particles and topicalized elements. Thus, while her analysis shares some points with Frey’s, it arrives at some notably different conclusions.

In their contribution titled “Argument Omission in Imperative-Declarative Conjunctions”, Robert Külpmann & Vilma Symanczyk Joppe determine whether acceptability ratings from empirical studies would support the traditional classification of imperative-declarative conjunctions (IDCs) with respect to the omission of a direct object:

1. (6)

   

The above examples illustrate two different types of IDCs: one with the conjunction oder, the other with und. The results from Külpmann & Symanczyk Joppe’s study confirm the traditional classification; no significant difference between plain imperatives and the two types of IDCs above was found. They conclude that argument omission should be treated as a property of sentence type and cannot be reduced to verb mood.

In her chapter titled “Causal Clauses in Old Indo-European Languages”, Rosemarie Lühr sets out to determine whether Old Indic and Ancient Greek expressed factual, epistemic, and speech-act causality by different linguistic means. Of particular interest to the study are epistemic modality and speech-act modality. Her other goal is to establish whether these three types of causality are expressed by means analogous to the German weil-verb-final clause or to the weil-V2-clause. Lühr’s answer to the second question is negative. She arrives at the conclusion that it was more important to signal the type of discourse relevance than to make a formal distinction within the subordinate clauses. Furthermore, as opposed to German, where weil-V2-clauses are spreading at the expense of denn-clauses, the equivalents of weil in Old Indic and Ancient Greek have not ousted the equivalents of denn. As is the case with some other contributions, this chapter suffered from usage problems, punctuation errors, and typos. Two sentences cannot be parsed with certainty.

In her contribution titled “Syntax and Semantics of Causal Nachdem-Clauses in German”, Stefanie Pitsch states as her goals i) to establish the existence of causal nachdem-clauses in today’s Standard German, and ii) to analyze the syntax and semantics of both temporal and causal nachdem-clauses and arrive at a conclusion about how they are distinguished. As Pitsch points out, prescriptive grammars claim that causal nachdem-clauses are a thing of the past, except in certain southern varieties, and encourage today’s users to avoid them for clarity’s sake. Pitsch’s data support the existence of causal nachdem and reveal how the temporal usage has become the favored one. In brief, the temporal interpretation occurs more frequently because its semantics and syntax can be derived without ambiguity when used with the perfective aspect. By contrast, the causal interpretation arises (only) when the embedded verbal predicate headed by nachdem can be construed as imperfective. Given that the perfective is the default and unmarked aspect in German, the predominance of temporal nachdem is not surprising.

Pitsch goes into some detail on how each type of usage can be derived. Her derivations follow the work of Klein (Reference Klein1994) and require the category [Asp] projecting AspP, where it is determined whether the specification of nachdem is temporal or causal. Although this analysis seems to answer the questions about the properties of nachdem (which Pitsch refers to as a subjunction, contrary to standard usage), it would be interesting to know how the judgments of an empirical study on current usage fall out.

Marga Reis, in her chapter titled “Consecutive so …V2-Clauses in German”, undertakes an analysis of V2 clauses that are preceded by—and thus in some way are in the semantic domain of—the associated clause that ends in so + adverb. One question in particular is how the V2 version without the complementizer in 7a differs from the V-final version with dass in 7b.

1. (7)

   

Since no other work has been done on this question, one of Reis’ main objectives is to provide a detailed description of the relevant formal and interpretive facts. Beyond that, she compares this type of dependent V2-clause to others explored in this book. In the final section, Reis provides a potential explanatory perspective that draws on the Assertional Proto-Force Hypothesis of Gärtner (Reference Gärtner2002).

Reis’ chapter is a model of clarity, organization, and linguistic methodology. Her easy-going style, sprinkled with colloquialisms, renders the content more accessible and precise. Not only does it provide an outstanding introduction to this particular variety of dependent V2 clauses, it also points to future research topics in an open and theoretically unbiased way.

The final chapter of this volume, “Variable Binding as Evidence for Clausal Attachment” by Sophie von Wietersheim, reports on two experiments carried out at the University of Tübingen. The goal of the study was to determine whether or not some theoretical claims made in the generative literature about the difference between CACs and PACs are borne out by empirical tests. The experiments involved online tasks performed mostly by students at the University of Tübingen, who had to provide acceptability judgments on variable binding between a matrix clause and an embedded adverbial clause. One of the main objectives was to settle incompatible claims made in the literature about various subordinate clause integration tests. More specifically, the experiment sought to identify the structures that do or do not pass those tests. The participants used a five-point acceptability and naturalness scale ranging from very acceptable to very unacceptable, based on the standard acceptability scale outlined by Featherston (Reference Featherston, Featherston and Winkler2009).

Both experiments showed that the two clause types exhibit contrasting syntactic behavior. In particular, support was found for the theoretical claim that CACs allow c-command and binding into them, whereas PACs do not. This finding, in turn, confirms the view that the former are integrated, whereas the latter are not. The experiments also revealed other factors that affect acceptability judgments of variable binding. The participants appear to rely on factors other than the structural precondition of c-command to identify the binding relation between the variable and the quantifier. Unsurprising to some, linearity proved to be a very powerful factor. Wietersheim’s presentation of this study is very precise, clear and well-written. I would recommend it highly as an introduction to this kind of work.

With regard to the editorial work on this volume, I think more attention should have been paid to uniformity and correctness of manuscript style, including punctuation, standard English usage, and the format of examples. In addition, the editors should have insisted on always providing translations, including for sentences within the body of the text. Though these errors rarely contribute to lack of clarity or ambiguity, they could be interpreted as an indication that the editorial work, and by implication the content, is not of the highest standard. That would be unfair to the vast majority of the contributions. The volume overall has my high recommendation.