This book comprises a collection of seventeen papers from a workshop on linguistic complexity held at the Max Planck Institute for Evolutionary Anthropology in Leipzig in April 2007. In addition to the papers there is an interview with Daniel Everett by one of the editors, Geoffrey Sampson, and a brief ‘Envoi’ by the editors at the end.
The purpose of the workshop and of the book that arose from it is to challenge the view that all languages are equally complex. This ought to be an empirically testable hypothesis after a workable definition of linguistic complexity has been established. Many of the chapters in the book make a serious effort to provide such a definition, and the hypothesis is refuted by most of the definitions offered here. Besides the challenge of defining linguistic complexity, the hypothesis that all languages are equally complex leads to further questions: Can complexity be measured, and if so, how? How do we compare complexity in different components of the language? How invariable is invariance?
The Leipzig workshop was convened in response to ‘scepticism about a longstanding linguistic axiom—that all languages and language varieties are similar in complexity’ (vii). It is not made explicit, however, where this axiom comes from. The idea that all languages are equally complex may be implied by the view that all languages are equally advanced, that they have similar properties, and that there are no ‘primitive’ languages. But of course there is no such necessary implication.
The volume starts with an introductory chapter by one of the editors, Geoffrey Sampson (Ch. 1). This is where we would expect to find a first attempt at the problem of defining complexity, and of comparing complexity between languages and language varieties. But there is no such attempt. We naturally do not expect Sampson to provide clear answers to such problems, but we should at least expect him to discuss them, and to problematize the intricate notion of complexity. Instead, he shies away from the whole enterprise with statements like ‘it seems to me very difficult to define the job which grammar does in a way that is specific enough to imply any particular prediction about grammatical complexity’ (2). So how can we then argue for or against equal complexity, let alone publish an entire book about it? Nevertheless, Sampson goes on to argue strongly against the claim that all languages are equally complex. As the following chapters demonstrate, there are good empirical and theoretical grounds for such an argument. But much of Sampson’s own reasoning is difficult to follow because his argumentation is in part very imprecise. That is, it is not always clear whether he is talking about the grammar of a language in general, or the particular use of that grammar in specific contexts or genres. For instance, he presents an excerpt from Classical Latin poetry as an example of Latin syntax (3). It may very well be that Classical Latin, even the everyday spoken variety, was more complex than, say, modern Italian, but poetry hardly serves as the testing ground. Similarly, he claims that ‘Biblical Hebrew made strikingly little use of subordinate clauses’ (6–7), but there must have been some subordinate clauses in the language, and subordination was part of Hebrew grammar at the time. So how did Hebrew differ in complexity from other languages with subordinate clauses? Sometimes Sampson [End Page 748] uses lexical facts to argue against claims about grammar. The fact that vocabulary learning goes on throughout a lifetime is not an argument against the theory of an innate grammar (8–9), and the observation that ‘small children use simpler grammar than adults’ (14) is hardly an example of varying complexity among languages.
The question of recursion is important to Sampson’s argumentation. Generative grammar claims that recursion is part of universal grammar (UG). The existence of a language without recursion, therefore, would falsify UG and would be an argument against generative grammar. This is one of...