The first task of a grammar or grammar sketch is to identify the language being described, and to provide certain particulars concerning its ethnolinguistic context. It is also important to orient the reader to previous literature and other research that has been done on the language.

The name of the language

Self-referent or auto-denomination are the anthropological terms for the name a group of people uses to refer to themselves. Often this name can only be translated as “people,” or “human beings.” It may also have hierarchically related meanings. For example, the word e'ñapa in Panare (a Carib language of Venezuela) means “person” when used in opposition to the term në'na “wild animal” or “evil spirit.” The same term means “indigenous person” when used in opposition to the term tato “outsider”/“white person.” Finally, the term can also refer strictly to Panares, when used in opposition to terms referring to neighboring indigenous groups. Only the context can disambiguate.

The terms by which language groups are known to outsiders are usually drawn from the outsiders' language, and are often derogatory in nature, e.g., in Peru the group now known as the Urarina used to be called the Chimaco a Quechua term meaning “unreliable.” Such terms are often not recognized by the people themselves, and, as in the case with Urarina, the self-referent can sometimes be substituted for the outsiders' term. On the other hand, the term Panare mentioned above is a Tupí word meaning “friend.” So the outsiders' form of reference is not always derogatory. If there is a well-established tradition in the literature of using the outsiders' term, a linguistic researcher should not try to change it, unless the people themselves are offended by the general term and clearly would prefer to be known by the self-referent.

Historical background and definitions

Morphology is the study of shapes. For example, one can talk about the morphology of camels – different species of camels have different morphologies, i.e., they have different body shapes. Morphology in linguistics has to do with the shapes of words. How are words shaped in such-and- such a language? What systematic rules determine when and how they may adjust their shapes? Traditionally, morphology has also been concerned with the “categories” (i.e., “operations” or “functions”) represented by adjustments in the shapes of words, as distinct from those operations represented by lexical or analytic processes (see Introduction, section 0.2.3).

In the rest of this section I will briefly define some terms used in discussions of morphology. After that, an outline for a possible chapter on morphological typology will be suggested.

A morpheme is a minimal shape. The classical definition of a morpheme is a minimal formal shape or piece that expresses meaning. For example the English word dogs contains two morphemes: dog, which embodies the main semantic content of the expression, and -s, which embodies the meaning of plurality. The form dog itself is not further divisible into meaningful component pieces, therefore it is a morpheme – a minimal shape. In most situations this definition works fine. However, more current approaches acknowledge the fact that particular meanings are not necessarily directly linked to particular pieces of form. For example, in Maasai (an Eastern Nilotic language of Kenya and Tanzania) many morphemes are not pieces of form at all; rather, they are tone patterns. Example la is in the active voice while lb is the contrasting “middle” voice (examples courtesy of Jonathan Ololoso).

The following is a list of reference grammars that may serve as examples of how a grammatical description may be organized. These grammars have been judged as “successful” by an informal panel of professional and student linguists who have actually used reference grammars in their research. They provide alternative organizational schemes to the one offered in the body of the present work. In general, the criterion for whether a grammar is “successful” or not is whether reliable information can be gleaned from it fairly quickly by readers who may not be at all familiar with the language being described. Other helpful characteristics include an insightful description of the sociological and cultural context in which the language is used, well-glossed examples, transparent terminology, and an inclusive index.

In formulating an outline for a grammar sketch or reference grammar, it is very important to keep in mind that the inclusiveness of the work will be in direct proportion to the author's familiarity with the language, and resources available for fieldwork. While a complete reference grammar for every language on Earth is ideal, exigencies of fieldwork and limitations on funding and time make it necessary at times to limit the scope of a description. For a language that is completely undocumented, a concise ten-page sketch may be extremely useful, while for a language that has been well studied, and may be spoken by a large number of speakers, a more detailed reference grammar would probably be necessary. It is important for a fieldworker to accurately estimate the level of detail of a proposed reference grammar in proportion to available resources.

The purpose of this book

This book is a guide and a bridge. I hope it will be a better guide than Squirrel, and a better bridge than the water boa. It is a guide for linguistic fieldworkers who desire to write a description of the morphology and syntax of one of the many under-documented languages of the world. It is a bridge designed to bring the extensive knowledge of linguistic structure that exists in the literature to bear on the complex and often confusing task of describing a language.

As this introduction is being written, there are reported to be about 6,000 languages spoken on Earth (Grimes 1992). About 2,000 of these have received close attention by linguistic researchers. The other 4,000 (roughly speaking) have only sporadically been described by linguists, and many have not even been recorded in written form for future generations. Krauss (1992) estimates that 3,000 of the 6,000 or so languages spoken today will become extinct in the next century. The human and intellectual tragedy of language extinction has been well articulated by Krauss, Hale (1992), and others. It is not surprising that the 3,000 languages facing extinction come overwhelmingly from the 4,000 or so that have not been consistently described.

In the previous chapter, we looked at the syntax of head movement. In this chapter, we look at a very different kind of movement operation, known as operator movement because it applies to expressions which contain an (e.g. interrogative or negative) operator of some kind.

So far, we have implicitly assumed that CP comprises a head C constituent (which is filled by a complementizer in some structures and by a preposed auxiliary in others) and an IP complement. However, one question which such an analysis begs is where the bold-printed preauxiliary constituents are positioned in structures such as (1) below:

1. (1) (a) What languages can you speak?

2. (b) No other colleague would I trust

Each of the sentences in (1) contains an inverted auxiliary (can/would) occupying the head C position of CP, preceded by a bold-printed phrase of some kind (viz. what languages and no other colleague). Each of the pre-auxiliary phrases contains a word which is sometimes classed as an operator: more precisely, what is an interrogative operator (or whoperator) and no is a negative operator. Expressions containing (e.g. interrogative or negative) operators are – for obvious reasons – called operator expressions (hence what languages and no other colleague in (I) are operator expressions).

It seems clear that each of the operator expressions in (1) functions as the complement of the verb at the end of the sentence.

Grammar is traditionally subdivided into two different but inter-related areas of study – morphology and syntax. Morphology is the study of how words are formed out of smaller units (traditionally called morphemes), and so addresses questions such as ‘What are the various component parts (= morphemes) of a word like antidisestablishmentarianism, and what kinds of principles determine the ways in which the parts are combined together to form the whole?’ Syntax is concerned with the ways in which words can be combined together to form phrases and sentences, and so addresses questions like ‘Why is it OK in English to say Who did you see Mary with?, but not OK to say *Who did you see Mary and?’ (a star in front of an expression means that it's ungrammatical). ‘What kinds of principles determine the ways in which we can and cannot combine words together to form phrases and sentences?’

However, grammar is traditionally concerned not just with the principles which determine the formation of words, phrases and sentences, but also with the principles which govern their interpretation – i.e. with the principles which tell us how to interpret (= assign meaning to) words, phrases and sentences.

Overview

The aim of this chapter is to outline contemporary ideas on the nature of grammar and the acquisition of grammar. The approach adopted here is that associated with the principles-and-parameters model developed by Noam Chomsky during the 1980s and 1990s, in works ranging from his 1981 book Lectures on Government and Binding to his 1995c book The Minimalist Program.

Grammar

Grammar is traditionally subdivided into two different but inter-related areas of study – morphology and syntax. Morphology is the study of how words are formed out of smaller units (traditionally called morphemes), and so addresses questions such as ‘What are the various component parts (= morphemes) of a word like antidisestablishmentarianism, and what kinds of principles determine the ways in which the parts are combined together to form the whole?’ Syntax is concerned with the ways in which words can be combined together to form phrases and sentences, and so addresses questions like ‘Why is it OK in English to say Who did you see Mary with?, but not OK to say *Who did you see Mary and?’ (A star in front of an expression means that it's ungrammatical.) ‘What kinds of principles determine the ways in which we can and cannot combine words together to form phrases and sentences?’

However, grammar is traditionally concerned not just with the principles which determine the formation of words, phrases and sentences, but also with the principles which govern their interpretation – i.e. with the principles which tell us how to interpret (= assign meaning to) words, phrases and sentences.

Overview

So far, our discussion of syntactic structure has tacitly assumed that all constituents in a given structure are overt. However, we now turn to argue that syntactic structures may also contain empty (= covert = null) categories – i.e. categories which have no overt phonetic form, and hence which are inaudible or silent. As we shall see, empty categories play a central role in the theory of grammar which we are outlining here.

PRO subjects

We begin by looking at the structure of clauses which might be argued to contain an empty subject. In this connection, compare the structure of the bracketed infinitive clauses in the (a) and (b) examples below:

1. (a) We would like [you to stay]

2. (b) We would like [to stay]

3. (a) We don't want [anyone to upset them]

4. (b) We don't want [to upset them]

5. (a) They will expect [students to pass the exam]

6. (b) They will expect [to pass the exam]

Each of the bracketed infinitive complement clauses in the (a) examples in (1–3) contains an overt (italicized) subject. By contrast, the bracketed complement clauses in the (b) examples appear to be subjectless. However, we shall argue that apparently subjectless infinitive clauses contain an understood null subject. (By saying that a constituent is null or empty or covert, we mean that it has no overt phonetic form and so is silent.)

In this chapter, we are concerned with the structure of phrases and sentences – i.e. with the way in which words are combined together to form phrases and sentences. To put our discussion on a concrete footing, let's consider how an elementary two-word phrase such as that produced by speaker B in the following mini-dialogue is formed:

1. (1) SPEAKER A: What are you trying to do?

2. SPEAKER B: Help you

As speaker B's utterance illustrates, the simplest way of forming a phrase is by merging (a technical term meaning ‘combining’) two words together: for example, by merging the word help with the word you in (i), we form the phrase help you. The resulting phrase help you seems to have verblike rather than nounlike properties, as we can see from the fact that it can occupy the same range of positions as the simple verb help, and hence e.g. occur after the infinitive particle to: cf.

1. (2) (a) We are trying to help

2. (b) We are trying to help you

By contrast, help you cannot occupy the kind of position occupied by a pronoun such as you, as we see from (3) below:

1. (3) (a) You axe very difficult

2. (b) *Help you are very difficult

So, it seems clear that the grammatical properties of a phrase like help you are determined by the verb help, and not by the pronoun you.

Overview

In chapters 3 and 4, we discussed the ways in which words are projected into syntactic structures. In this chapter, we shall be concerned with the principles which determine the morphological form of words. The kinds of question we shall ask here include (for example) why we say We are winning not *Us are winning, or *We is winning, or *We are win. Why is it that we require we rather than us here, are rather than is, and winning rather than win? We shall suggest that the morphological properties of words can be characterized in terms of sets of grammatical features, and features must be checked in an appropriate manner: this chapter thus provides an introduction to the concept of checking. It should be noted that work on checking theory is as yet in its infancy, so that many of the ideas and descriptive details in this chapter are inevitably somewhat sketchy and speculative.

Interpretable and uninterpretable features

Before we explore feature-checking, however, let's first consider the overall organization of a grammar, and the role which features play in it. We assume that the sentence structures formed by successive merger operations must ultimately be mapped (i.e. converted) into two different kinds of structural representation for the sentence: (i) a representation of those aspects of the structure of the sentence which determine its phonetic form (= PF); and (ii) a representation of those aspects of the structure of the sentence which determine its logical form (= LF, i.e. linguistic aspects of the meaning of the sentence).

So far, our discussion of syntactic structure has tacitly assumed that all constituents in a given structure are overt. However, we now turn to argue that syntactic structures may also contain empty (= covert = null) categories – i.e. categories which have no overt phonetic form, and hence which are inaudible or silent. As we shall see, empty categories play a central role in the theory of grammar which we are outlining here.

We begin by looking at clauses which might be argued to contain an empty subject. In this connection, compare the structure of the bracketed infinitive clauses in the (a) and (b) examples below:

1. (1) (a) We would like [you to stay]

2. (b) We would like [to stay]

1. (2) (a) We don't want [anyone to upset them]

2. (b) We don't want [to upset them]

Each of the bracketed infinitive complement clauses in the (a) examples in (1–2) contains an overt (italicized) subject. By contrast, the bracketed complement clauses in the (b) examples appear to be subjectless. However, we shall argue that apparently subjectless infinitive clauses contain an understood null subject. (By saying that a constituent is null or empty or covert, we mean that it has no overt phonetic form and so is silent.) The kind of null subject found in the bracketed clauses in the (b) examples has much the same grammatical and referential properties as pronouns, and hence is conventionally designated as PRO.

In the previous chapter, we looked at the syntax of subjects, arguing that these originate in a θ-marked specifier position within VP, and typically move into a case-marked specifier position within IP by application of raising. In this chapter, we turn to look at further instances of raising, and at an additional type of movement operation traditionally termed passivization. We go on to argue that raising and passivization are two different variants of the same A movement (= argument movement) operation, and we look at the syntax of A movement.

To get our discussion onto a concrete footing, consider the alternations illustrated below:

1. (1) (a) It seems [that he understands her]

2. (b) Heseems [to understand her]

3. (2) (a) It would appear [that they are lying]

4. (b) They would appear [to be lying]

5. (3) (a) It happened [that she came across an old love-letter]

6. (b) Shehappened [to come across an old love-letter]

7. (4) (a) It turned out [that Mary was right]

8. (b) Maryturned out [to be right]

The bold-printed verbs in these examples have a that-clause complement in the (a) examples, and an infinitive complement in the (b) examples. But what is puzzling about sentences like (1–4) is that the italicized expression which functions as the subject of the bracketed complement clause in the (a) examples surfaces as the subject of the matrix clause (i.e. the clause containing the complement clause) in the (b) examples: for example, he is the subject of understands in (Ia), but the subject of seems in (Ib).