Introduction

This chapter discusses the contribution of linguistic typology to the study of language universals. Language universals research and linguistic typology are closely related fields, and are often not distinguished very clearly. Yet the difference between them can be characterized in the following way (Comrie, 1989): language universals define the restrictions on cross-linguistic variation; linguistic typology studies the restrictions on cross-linguistic variation; so typological research can be seen as the primary method used in uncovering language universals.

After a brief description of the way in which language samples are selected in cross-linguistic research in Section 2, the basic concepts and the methodology used in linguistic typology are introduced in Section 3, in which the notion of implicational hierarchy is taken as the point of departure. The implicational hierarchies uncovered through typological research are generally assumed to reflect true language universals, which means that they may be expected to show up in other linguistic domains as well, such as the historical development of languages, the process of language acquisition, language contact phenomena, and the distribution of linguistic phenomena within a single language. Section 4 of this chapter is dedicated to this issue. Conclusions are presented in Section 5. Wherever possible, the examples I present are taken from my own work, since this allows me direct access to the primary data.

Language sampling

Typological investigations make use of representative samples of the approximately 6,000 languages of the world.

Introduction

As traditionally understood, universals of language are cross-linguistic generalizations concerning synchronic grammars, and their explanations usually appeal to functional principles thought of in a synchronic domain. It stands to reason, however, that any synchronic pattern must have a diachronic dimension, since that pattern had to come into being in some way. One could even argue, as I did in Bybee (1988), that we cannot be sure of the validity of a functional explanation for a synchronic universal unless we can confirm that that functional consideration was applicable in the formation of the synchronic pattern. That is, all explanations of synchronic universals must have a diachronic dimension.

In the current chapter, I outline a position on the role of diachrony in universals, whose logical consequence is that the true universals of language are not synchronic patterns at all, but the mechanisms of change that create these patterns. This position is an extension of the theory of diachronic typology formulated and practiced by Joseph H. Greenberg, to whom this chapter is dedicated.

In several papers, Greenberg proposed a method for the study of typology and universals which he called dynamic comparison or diachronic typology. In this method, typological patterns are shown to emerge from common diachronic changes that arise in related and unrelated languages. It has become clear subsequently that what Greenberg elaborated in the many domains of language that he studied was not a comparative methodology so much as a theory of language that has great potential for explanation.

The debate on language universals

Introduction

For the last several decades we have been living in what has been called, for better or for worse, the postmodern era, a cultural movement or climate of social sensitivity, which, in contrast to the traditional values of the rationalistic, globalizing version of Modernism inherited from the Enlightenment, defends ideological positions based on heterogeneity, dispersion, and difference. Over the past years, contingency and individuality have gradually taken precedence over permanence and universality. As Harvey (1989) so accurately states, the views that are presently most highly valued in the postmodern world are generally those that concede greater importance to particularism and fragmentation, focus on the individual nature and interest of the parts rather than the whole, and are ultimately conducive to the disarticulation or deconstruction of all human sociocultural and economic activities. In the same way that moral values and instruction are not thought to be universally applicable, many well-known scholars of this era, even in the realm of science – especially the social sciences (e.g. the work of Lyotard) and, to a lesser extent, physics and mathematics (in line with Spengler) – affirm that there are no general principles that can be objectively evaluated independently of the spatiotemporal context in which they were initially proposed.

What are universals?

In everyday life we are struck by the differences between languages. When we overhear a conversation between people speaking a language we have no knowledge of, we find it completely impenetrable. There is something disturbing, even sinister, about a language you do not understand at all. The fear this engenders is part of what tempts authoritarians everywhere to ban the use of unfamiliar languages, and enforce the use of the familiar. Unfamiliar languages are also frequently looked down on as unstructured and incapable of serving to convey more than simple messages. Linguists confront this fear and contempt, and it is part of the value of the mainstream traditions of linguistics as a humanistic discipline that they insist on the equal worth of every language. Every language is considered equal, in the sense of meriting equally serious consideration, akin to the way that all people are declared to be “equal in dignity and rights” in the Universal Declaration of Human Rights (United Nations, 1948). In contrast to lay emphasis on linguistic differences, for linguists the demonstration of similarities between languages functions at the same time as one of the ways in which the proposition of equality is supported and as one of the reasons why linguists hold this view. We feel justified in constructing general theories within which the peculiarities of individual languages can be described, rather than taking it as our task to build quite separate theories for different languages.

Any mention of linguistic universals means the continuation of a journey begun many years ago, and refers to a topic of debate among both linguists and philosophers, which has been a constant in the history of linguistics throughout the ages.

The debate regarding universals is one of the most fundamental chapters – perhaps the most fundamental – in the history of grammar, and its genesis can be traced back to the very dawn of linguistic reflection. Furthermore, it is a subject that transcends boundaries between academic disciplines since it is one of the cornerstones of the philosophical debate between rationalism and empiricism. Consequently, it is of vital interest not only to linguists, but also to philosophers, psychologists, anthropologists, psychobiologists, and ethnologists – in other words, to researchers of all academic disciplines that are involved in what is known today as Cognitive Science.

However, linguistic universals are currently in the limelight because any linguistic theory that aspires to explanatory adequacy must offer a satisfactory answer to the question of why languages that are so apparently different on the surface at the same time present undeniable regularities in their underlying structure. It is no longer a question of merely discussing the existence of universals, but rather of making their existence compatible with the epistemological premises of different theoretical approaches. This book is an explanation of how these approaches have dealt with this task.

General considerations

My aim in this chapter is first to present an overview of my assessment of the state of the art in syntactic typology, and then to illustrate the general points in Section 2 by means of the investigation of one particular phenomenon, namely the syntactic typology of relative clauses, with particular regard to areal characteristics of European languages in comparison with other languages of the world.

What distinguishes the typological approach to syntax, or indeed to grammar as a whole, from other approaches? One feature of typological work, indeed arguably the defining characteristic of such work, is the serious attention paid to cross-linguistic diversity. In order to understand Language, it is essential to understand languages. Although all approaches that consider themselves general-linguistic in orientation at least pay lip-service to the cross-linguistic applicability of their tenets, it is the typological approach that sets out to examine data from as wide a range of languages as possible, in order to ensure that we have the best basis possible for deciding what logical possibilities are actually attested among the languages of the world, and thus for assessing how wide-ranging our characterization of the phenomenon in question must be. It is not sufficient for an approach to handle English but not Japanese, or vice versa. And indeed, in Section 2 of this chapter, I will be taking data from a fair range of languages, backed up by data from far more languages in the overall typological literature on relative clauses, in order to draw certain specific conclusions about the typology of relative clauses.

Introduction

The history of morphology in grammatical theory is somewhat checkered. For the American structuralist tradition, morphology was central. In the Chomskyan generative tradition, syntax is central and morphology has been either relegated to phonology and syntax or expelled from linguistics altogether. There is thus a good deal of controversy about the status of morphology itself. It is not even entirely clear what the subject matter of morphology is. Informally, we tend to think of the domain of morphology as the word, but the concept of word is notoriously difficult to pin down and it is clear that the pre-theoretical notion corresponds to several distinct and partially independent technical notions (see Dixon and Aikhenvald, 2002a, for a recent review of the issues). On the other hand, for American structuralists such as Bloomfield (1933), morphology seems to have been about morphemes. However, the morpheme concept is no more secure than the word concept, and the utility of the morpheme has been flatly denied by many recent theorists.

Even without these conceptual uncertainties, we must be careful to distinguish typological claims about morphology (words or morphemes) from claims about other aspects of grammar which are often reflected in morphology. Consider the expression of number: oversimplifying considerably, we can say that if a language has a trial number it has a dual; if a language has a dual number it has a singular–plural distinction (Corbett, 2000, 38f.).

Preliminaries

Lexicology is the study of words and their relationship to each other. A first step, then, is the determination of the boundaries of a word. While we all have an intuitive sense of what a word is, defining the term in a scientific manner is much more difficult. The typesetter might describe the word as any unit surrounded by spaces, in which case don't is one word but do not is two. This leads to such questions as: Do compound forms of a verb (Ishall have eaten) constitute one word or three? Is the term arc-en-ciel ‘rainbow’ one word or three? We see that lexical units, sometimes referred to as lexemes, might include more than one ‘word’.

At the same time, a single form might represent different words. The verb lead /lijd/ is a different word from the noun lead /lijd/ ‘dog's leash’. Even within the same grammatical category, ball (sporting equipment) is a different word from ball (a formal dance). Given these problems, Touratier (1998) proposed a set of defining characteristics to distinguish the ‘word’ from a ‘morpheme’, insisting on the autonomy of the word. Through the characteristic of ‘moveability’, a word can be moved about in its sentence context, while bound morphemes cannot be (see section 3.3.1). The second criterion is ‘separability’, by which we recognize that a word is independent because it can be separated from the surrounding constituents of a phrase, i.e. it is not part of a fixed phrase.