In this book, Stroik and Putnam take on Turing's challenge. They argue that the narrow syntax – the lexicon, the Numeration, and the computational system – must reside, for reasons of conceptual necessity, within the performance systems.
AUTHOR: Newmeyer, Frederick TITLE: Possible and Probable Languages: A Generative Perspective on Linguistic Typology PUBLISHER: Oxford University Press YEAR: 2005
Mayrene E. Bentley, Department of Languages & Literature, Northeastern State University, Broken Arrow, Oklahoma.
_Possible and Probable Languages_ (PPL) serves as an excellent reference book or supplemental reading for a typology course or a syntax course focused on the question of language variation and/or acquisition. PPL provides a comprehensive overview of the principal research on language universals and the motivation behind the research within the framework of generative syntax and more recently, Optimality Theory. Citing Chomsky's evolving views on Greenberg's universals, Newmeyer details how Chomsky's ideas changed during the 70s, 80s, and 90s, on the relationship between cross-linguistic variation and the principles of universal grammar (UG). In summarizing Chomsky's contributions as well as the work of other prominent linguists, Newmeyer takes issue with particular aspects of the claims of functionalism and formalism in hopes of showing that neither approach alone can account for language variation. PPL's goal is clear: both language use and a formal structural system are necessary to explain the relationship between grammatical form and outside forces shaping the world's languages.
''On the Possible and Probable in Language'' is a summary of language universals and their explanation from functionalist and formalist viewpoints. Newmeyer provides examples of possible and improbable features of languages from the typological literature as well as ''The Universals Archive'' that is housed at the University of Konstanz and available electronically. These universals provide a backdrop for a discussion of experiments where subjects tried to learn ''unnatural'' rules with respect to real and unreal languages. These experiments suggest two things: 1) that humans adapt more readily to learnable rules and hence, there is a link between innateness and possible grammars; 2) although a rule may appear as ''improbable,'' it does not imply that it is impossible, but simply non-occurring.
''Parameterized Principles'' looks at how early generative syntax neglected what typology had brought to light in the 60s regarding cross-linguistic variation and instead, sought solutions to various syntactic questions in theoretical assumptions based on language-specific discoveries. Newmeyer walks the reader through numerous examples of how UG later returned to answering questions of cross-linguistic variation in word order, gapping, etc., by delving into markedness, deep vs. surface structures, parameterized principles, and currently, optional features on functional heads. However, it becomes apparent that UG's goal is antithetical to the hard-sought goal of explanatory elegance since numerous constraints were needed to account for and predict an ''unexpected (and previously unnoticed) clustering of typological properties'' (p. 44-45).
''Parameters, Performance, and the Explanation of Typological Generalizations'' zeros in on particular principles of UG such as ''Head Parameter,'' and ''Subjacency,'' and how these principles and their corresponding parameter settings fall short when accounting for linguistic variation and the correlation of other language features. Newmeyer maintains that Greenberg alone has been successful in correlating the basic word order of a language with other morphosyntactic features. Newmeyer's extensive grasp of the literature in these areas makes for a very convincing argument while leading the reader to conclude that ''the degree of grammatical variation is in fact highly constrained, but much more by performance factors than by UG'' (p.75). Newmeyer concludes this chapter by reviewing Hawkins' (2004) hypothesis of ''Performance-Grammar Correspondence'' as the means for explaining efficiency in language.
''In Defense of the Saussurean View of Grammar'' sets out to defend research in formal grammar and the notion that mental grammar is integral to language use. Newmeyer begins, however, by showing that ''language users are sensitive to the frequency of grammatical forms'' (p.130). He then seeks to counter the supposed ''disparity'' between the formal structures that grammars describe and the utterances people make by arguing that speakers have a full argument structure (SVO, SOV, etc.) although they may produce only parts of the full-blown structures. Newmeyer cites examples from Merchant (2004) on seven different languages that illustrate how speakers respect morphological case even when replies are fragments. He then provides additional examples that show how speakers rely on inference to decipher the language code, which is ultimately only a partial conveyor of what speakers need grammars to do. With respect to language change, Newmeyer devotes several pages to data (reflexives and differential object marking) that attempt to disprove that a need to lessen ambiguity has led to certain morphosyntactic features; he claims that speakers rely instead on ''the well-established hypothesis that within a given domain, more frequent combinations of features require less coding than less frequent ones'' (p. 158). Although Newmeyer spends time arguing that if ''speakers were really driven to reduce potential ambiguity'' (p. 159), they would put objects before all other linguistic constituents since an object is more predictive of verb type than the subject is, he chooses not to invoke the argument he uses a few pages later that ''our interpretation of a structurally ambiguous string is determined in large part by real-word contextual factors that have nothing at all to do with grammar, no matter how broadly defined'' (p.162). He ends the chapter by decrying stochastic grammars and advocating that language evolution had to have begun with conceptual structures that were translated into sounds via grammar such that as communication evolved, use and grammar evolved imperfectly.
''The Locus of Functional Explanation'' addresses specifically whether an atomistic or holistic functionalism explains the ''functional pressures and the typological distribution of formal elements that represents a response to those pressures'' (p. 174). Newmeyer is upfront that his position rests on the side of the holistic approach, which he explains as one that claims no direct link between function and use. The link between function and use is indirect and evolves through language use and acquisition and ultimately plays itself out typologically according to Newmeyer. He argues that atomistic functionalism (AF) is unable to provide all the possible external factors motivating language, and that the literature has obscured this fact to a certain degree. The last part of the chapter is a rebuttal to the claims of Optimality Theory (OT); in particular, Newmeyer takes issue with those who argue that the constraints put forth in OT can be accounted for functionally, i.e., FOT. He replies specifically to claims made by Aissen (1999, 2003) that differential object marking can be explained by OT constraints. The OT machinery required to account for the various hierarchies implied within the given data essentially contradict the functional explanations, i.e., they are too convoluted since ''there is no way for a correct form to emerge without competition between rival forms. Therefore, one needs to set up a proliferation of candidate sets, simply to ensure that a 'winner' results.'' (p. 223).
PPL bears witness to the fact that the functionalist viewpoint is being heard and taken seriously. While Newmeyer brings together the major research initiatives of the last forty years and seeks to highlight what each does best, his overview of OT was much too brief for anyone to understand it without some background in the representative tableaux. His motive for having a discussion of OT in the book is not to laud its contributions to the field of syntax in the past decade, but to show its inadequacy as a means for explaining typological generalizations within a theory of interacting constraints.
Many in the formal linguistic community will bristle at Newmeyer's conclusions. Others will recognize the merit in what he says and may, with time, begin to change the direction of theoretical linguistics as well. They can't dismiss the book. It is meticulously researched, comprehensive, and written by a linguist of stature. Newmeyer's clarity of thought and expression make the book more accessible than many of the original articles. His juxtaposing of various views provides a larger context for evaluating the research efforts of those linguists cited in the book. PPL is an excellent resource for graduate students and linguists who want to keep pace with recent developments in the field of linguistics, but haven't the time to absorb all the literature across competing frameworks. Moreover, it is a summary of Newmeyer's views on how the forces of language use and formal grammar join together as language is acquired and evolves over time to eventually provide the typological patterns found throughout the world's languages.
Editing comments: Glossing of examples was not consistent such as p. 38 where the Dutch is given a word-for-word gloss, but the Italian isn't. Page 34 inserted English ''the'' in the German example in (11b) while on page 36, example (16) omitted the gloss ''person.'' Despite careful introduction to theoretical terms in UG, Newmeyer refers to ''LF'' with no explanation on p. 44 and to ''X-bar schema'' on p. 49, but explains it on p. 105. Omission of the word ''to'' occurs in the last paragraph of p. 83. The expression ''neutralize adjectives'' is first used on p. 85 but not defined until p. 86. E-language is defined on p. 105, but not I-language. The tables and tableaux are not always appropriately placed within the text and cause the reader to ''search and find'' to get the needed examples as he or she reads, especially on pp. 199-200.
Aissen, Judith (1999) ''Markedness and Subject Choice in Optimality Theory,'' Natural Language and Linguistic Theory 17:673-711.
_____ (2003) ''Differential Object Marking: Iconicity vs. Economy'' Natural Language and Linguistic Theory 21:435-83.
Hawkins, John (2004) Efficiency and Complexity in Grammars. Oxford: Oxford UP.
Merchant, Jason (2004) ''Fragments and Ellipsis,'' Linguistics and Philosophy 27:661-739.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Mayrene Bentley is an Associate Professor at Northeastern State University.
Her research interests are typology, Eastern Bantu languages, and Cherokee.