Asking questions investigates various aspects of questions, providing a lot of insightful data and covering a vast area of linguistics, including syntax, semantics, pragmatics, and even some methodology pertaining to the study of language.

Robert Fiengo's main theme throughout the book is that sentence types used for questions represent incompleteness, and performing a question act indicates the speaker's ignorance. Moreover, Fiengo emphasizes that one cannot study the grammar of language while neglecting its use, nor study the use of language while neglecting the grammar. I completely agree with Fiengo on this point. Indeed, topics on the interface between these different subdisciplines of linguistics feature largely in recent literature.

Questions are an ideal topic for this interdisciplinary enterprise, since questions (or interrogative sentences) have been studied extensively in the area of syntax, while the questioning act is an essential subject in the study of pragmatics. The importance of looking at these phenomena across disciplines has been widely realized, although there is not much work available which directly tackles the interface issues. Asking questions is an exacting book, whose primary aim is to fill this gap in the literature.

It is worth noting that the book is perhaps aimed more at a philosophical than a linguistic readership. Being a linguist, I expect the author to proceed as follows in the presentation of the argument: identify an empirical problem, point out problems inherent in the previous analysis, propose a new theory, and show how the new theory solves the initial problem. However, in this book, Fiengo often begins the discussion by considering abstract ideas, new notions and even anecdotes before turning to the linguistic data, thereby creating the impression of ‘drifting’ into his proposal. The book contains a lot of interesting data and their description, but it was hard for me to grasp the general properties of the data and how the proposal outlined accounts for these.

In chapter 1, ‘Introduction: Ignorance and incompleteness’, Fiengo first distinguishes two sorts of ignorance. The first type of ignorance expresses a lack of an item or a predicate, and the second type expresses a lack of the capacity to assert. Depending on which kind of ignorance the speaker wishes to express, there are two sorts of sentence types which can be used: (i) open questions and (ii) confirmation questions. Open questions represent incompleteness in structure; for confirmation questions, which are sometimes also incomplete in structure, ‘there is an incompleteness in the using of them’ (11). Open questions often involve ‘deformation’, such as wh-movement and inversion, while confirmation questions do not exhibit any deformation, cf. (1)–(2).

1. (1)

   1. (a) Who did you see?

   2. (b) Is it raining?

2. (2)

   1. (a) You are talking to who?

   2. (b) It's raining?

In chapter 2, ‘The instrumental model of talking: How to talk about talk’, Fiengo attempts to distinguish knowledge of grammar and knowledge of use. Using an analogy from the world of carpenters' tools, the author also distinguishes between types and tokens of the same sentence type. Each token is fixed in a time of utterance as well as to a particular agent. Hence, any uttered token will be different, although different agents can use the same sentence type at different times.

Fiengo notes one difference between tools and languages: sentence types are personal tools, i.e. they arise from the individual grammar of different agents. Hence, it is a considerable task to figure out how speakers decide which tokens to utter and how hearers are able to identify these tokens. There must be some conditions that constrain the ‘mapping space’ between phonetic and logical form.

There are other aspects in which language resembles tools. First, just like the same tool can be used for different purposes, the same sentence type can be used to perform different speech acts. Hence, the knowledge of grammar and the knowledge of use are distinct. Second, in terms of evolution, form follows failure rather than function, i.e. it involves a succession of imperfections, with only the useful forms of language surviving.

Fiengo also makes a distinction between how to decide what sentence type to use and how to decide what speech act to perform. If the speaker wishes to pose a question, an incomplete sentence type is useful because it represents that there is an incompleteness in what is expressed, i.e. a gap that the speaker cannot complete and that the speaker wishes to be completed by the addressee.

Lastly, Fiengo argues in this chapter that intonation is not part of the grammar. According to the author, rising intonation is an invitation for conversational turn-taking. Hence, it can be a useful question marker, especially for confirmation questions, which can be mistaken as statements. Fiengo further suggests that it is the availability of yes/no-answers that determines whether a certain utterance is an assertion or a question. For example, utterances without a rising intonation can be answered by yes or no in ‘an interrogation scenario’ like (3). Fiengo concludes that the utterances in (3) qualify as questions, even though they lack rising intonation.

1. (3) You were in Central Park. Yes. You wanted to make a statement. No. You had always hated pigeons. No. You poisoned the bird seed. No. You waited for your chance. No.

In contrast, some utterances that have a rising intonation pattern (utterances for ‘setting the stage’) cannot be followed by yes/no-answers, as illustrated in (4). Hence, these do not qualify as questions even though they have rising intonation.

1. (4)

   1. (a) So I was in the frozen food aisle?

   2. (b) When I was in graduate school?

I agree with Fiengo in that rising intonation indicates turn-taking and that we have to be cautious in incorporating intonation into the grammar. However, I am doubtful that the availability of yes/no-answers is an appropriate test for questionhood because speakers can also agree (say yes) or object (say no) to statements. For instance, in uttering (5a), the speaker is showing neither incompleteness in structure nor a lack of the capacity to assert, but is making a statement. Yet, the addressee can respond by saying no to communicate disapproval or surprise, as demonstrated in (5b).

1. (5)

   1. (a) I'm going to France next week!

   2. (b) No, you aren't!

Chapter 3, ‘Open questions, confirmation questions, and how to choose which sentence-type to use when asking them’, extends the inventory of questions presented in chapter 1 and explores a wide range of questioning acts, introducing a set of novel parameters. As developed in chapter 1, Fiengo distinguishes between conventional (form) and occasional (using of the form) incompleteness. Conventional incompleteness is characteristic of open questions, while occasional incompleteness is characteristic of confirmation questions. In this chapter, the author also introduces a new term to characterize yes/no questions. While the incompleteness of open wh-questions is the absence of an item or a predicate, the incompleteness of open yes/no questions is the absence of the ‘glue’ between the subject and the predicate. For Fiengo, the term ‘glue’ corresponds to saturation in the sense of Frege (Reference Frege and Beaney1891/1952), and is part of the sentence type of assertion.

Most part of this chapter is devoted to a list of different uses of questions: questions confirming beliefs, sarcastic and ironic questions, rhetorical questions, tag questions, repeat questions and so on. Let me highlight some of the interesting observations that Fiengo makes concerning confirmation questions. A confirmation question is asked when there might be some, but not sufficient, evidence for asserting a proposition. This corresponds to the observation made concerning biased questions, as discussed in Nilsenová (Reference Nilsenová, Bel and Marlien2002), Gunlogson (Reference Gunlogson2003), and Romero & Han (Reference Romero and Han2004), among others. Fiengo goes on to examine confirmation questions like (6), characterizing the situation in which this kind of sentence can be uttered. Here, the speaker has a certain epistemic bias (e.g. the weather is hot), sees a possible counterexample (e.g. the addressee wearing a sweater) and tries to refute the counterexample.

1. (6) Aren't you hot?

One thing that confused me is that Fiengo considers confirmation questions to be a category distinct from both open and closed questions. In the previous literature, biased questions usually include rising declaratives, preposed negative questions and tag questions, where the former two correspond to Fiengo's confirmation questions and closed questions, respectively.

Although chapter 3 extensively discusses the usage patterns associated with the different sentence types, I do not think that Fiengo actually succeeds in showing that ‘incomplete sentence types are well suited to the asking of open questions’ (80). The chapter reads more like a list of cases where those questions are asked. It would have been useful to provide a summary of the taxonomy of question use, especially for cases like biased questions, where Fiengo's classification does not match the one generally adopted in the literature.

Chapter 4, ‘Quantifiers, wh-expressions, and manners of interpretation’, focuses on wh-questions and is concerned with the ‘manners of interpretation’ of different wh-expressions. The author starts with a discussion of quantifiers and the concept of ‘manners of interpretation’. According to Fiengo, ‘each’ and ‘every’ have the same quantificational force, but differ in their manners of interpretation: ‘each’ is ‘individualizing’, while ‘every’ is ‘totalizing’. Consequently, sentences containing the quantifiers ‘each’ and ‘every’ have the same truth conditions but differ in the mode of presentation, in the sense of Frege. Fiengo then applies this distinction to the wh-phrases ‘which N’ and ‘what N’: ‘which N’, like ‘each’, has an individualizing manner, whereas ‘what N’, like ‘every’ has a totalizing manner.

In chapters 1 and 3, Fiengo argued that questions are characterized by incompleteness. In chapter 4, he claims that wh-expressions themselves are not incomplete, but that the variable bound by those wh-expressions is the site of incompleteness. Here, Fiengo also touches on the incompleteness of indefiniteness. In many languages, wh-expressions and indefinite descriptions have similar or identical forms, and the author shows how the the structure and manners of interpretation of wh-expressions interact.

The structure of multiple questions shows in which way the speaker's ignorance lies; that is, the speaker's interest determines the choice between Which jockey rode which horse? and Which horse did which jockey ride? Thus, a speaker's pragmatic interest has an impact on syntactic structure.

At the end of the chapter, Fiengo turns to Austin's (Reference Austin, Urmson and Warnock1953) quartet of speech acts (calling, describing, Exemplifying and classing), which he uses to categorize wh-expressions.

Chapter 5, ‘Syntactic structure’, deals with the syntax of questions. Fiengo first introduces the notion of a syntactic hierarchy, conceived of as ordered and unordered sets of linguistic items. On the basis of these set-theoretic notions, he defines relations of dominance and prominence. Dominance determines a scopal relation; hence, it is truth-conditionally relevant. Prominence, on the other hand, determines a structure pertaining to speech acts. In the case of multiple wh-questions, the more prominent wh-expression shows the speaker's interest as to how the pair-list answers are to be structured. Fiengo argues that movement results from splitting powers of syntactic expression. These powers seem to me to be equivalent to syntactic features, although note that for Fiengo, there is no feature-driven movement.

In chapter 6, ‘On the questioning speech-acts and the kinds of ignorance they address’, Fiengo tries to apply Austin's (Reference Austin, Urmson and Warnock1953) analysis of speech acts to different kinds of questions investigated in the previous chapters, and explores the kinds of ignorance they express and how ignorant the speaker can be.

To sum up, Asking questions investigates various aspects of question sentence types and questioning acts, proceeding from the assumption that questions represent incompleteness. It would have been nice to have a section of the book devoted to a comparison with other analyses of questions, such as Hamblin (Reference Hamblin1958), Karttunen (Reference Karttunen1977) and Groenendijk & Stokhof (Reference Groenendijk and Stokhof1984). The biggest problem I had with this book is that Fiengo introduces a vast number of new notions, categorizations and parameters, and therefore, the taxonomy of interrogative sentences and questioning acts proposed in this book is very complicated indeed. Having more information displayed in tabular form would be helpful to the reader in following Fiengo's exposition. Also, when arguing for a particular analysis, Fiengo appeals too often to the reader's intuitions about the conceptual matters, rather than providing linguistic data that clearly falsifies an alternative explanation.

Nonetheless, Fiengo argues quite convincingly that syntax cannot ignore pragmatic factors and vice versa. In particular, his point is well taken that, in order to understand the structural patterns of multiple wh-questions, one needs to refer to the speaker's interest on how the answers should be structured because the truth conditions of the two variants are exactly the same. The book also presents a wealth of information on the use of interrogative sentence types, and offers insightful observations. It will be mandatory reading for researchers seeking comprehensive data coverage on questions and an interesting new approach to the analysis of questions and general interface issues.