Book Review: Grammar as Science

Written by on April 7, 2019 in Book Review, Natural Language Processing, Uncategorized with 0 Comments

Title: Grammar as Science

Author: Richard K. Larson

Publisher: The MIT Press

Year: 2010

I love studying English Grammar. That is one of the reasons I enjoy working in the area of NLP. Machine Learning techniques apart, I firmly believe that a good understanding of the conventional approaches to modelling syntax and semantics is essential for achieving success in this field. 

I have a collection of many good books on syntax and semantics, but I do not miss the opportunity to buy a book if it comes across as interesting. The book I am going to review today, “Grammar as Science”  is one such book. It is not a recent publication; it was published in the year 2010, but only recently did I come across a reference to it and immediately bought the book.

This is easily one of the best introductory books on syntactic theory, and it is inspired (similar to many other books on the subject) by Noam Chomsky’s work. What is different about this book compared to others is that this book discusses syntax with primary focus on scientific theorising and scientific thought. As the author says in the preface “ … it is an introduction to syntax as an exercise in scientific theory construction”.

The book has a total of 28 Units (chapters) grouped into 7 major Parts (sections). Part 1 lays the ground work by introducing Linguistics and Syntax. Part 2 describes the idea of Phrase Structure Rules and shows how to define Grammars and test them. Chapter 4 in this part gives practical guidelines on how to systematically formulate a grammar (manually) from sentence examples and shows why judging well-formedness is not that simple. 

The book’s core idea of scientific theorizing is introduced and elaborated in Part 3. This, in my opinion, is the most important section in the book. It shows how we can compare theories and evaluate them. As part of this discussion, the author explains the idea of “tree” structure, constituencies and word/phrase categories. Constituency tests involving conjunction, ellipsis and dislocation are clearly explained. These concepts help us in comparing and revising grammars. 

Part 4, consisting of just two chapters, goes into the details of how to construct arguments in support of a specific grammatical structure that we may arrive at in the course of language analysis. Part 5 expands on this theme and goes into the process of “explaining” a chosen syntactic structure. It is in this section that the author introduces the “lexicon” and how it can associate multiple “features” with words, thus enabling us to come up with an acceptable structure for a given sentence. Other important aspects covered in this part are “adjuncts” and “complements”. 

Part 6 builds upon the previous section and discusses “Complement sentences” in much more detail. I found the explanation regarding hidden-subject analysis quite interesting. There is also a whole chapter devoted to the structure of NP. Towards the end of this part, the author introduces “X-Bar” theory. I recommend this nice video if you want to learn more about this theory.

Part 7, the last section in the book, delves into the more advanced topic of “wh-movement” and associated constraints. “Gaps”, usually associated with movement, are beautifully explained. Two entire chapters in this section are devoted to discussing the different constraints on movement. wh-movement is a complex topic and if you want some additional material, you can watch this video.

The topics covered in the book may be regarded by many beginners as somewhat abstract and complex. But the author deserves praise for adopting a unique presentation style using an excellent graphic layout. Illustrations are remarkable and are very down-to-earth. This approach renders the concepts easy to grasp and paves the way for a pleasurable reading experience. 

This book on classical syntactic theory might look outdated in the current scenario, where a lot of focus is on ML-based NLP. So, this is not the book for you if you are looking for advanced contemporary topics on NLP. Also, this is not the book for you if you are interested in getting into the nitty-gritty details of English grammar. For that I strongly recommend the following two books:

1. A Comprehensive Grammar of the English Language, by Randolph Quirk, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik, Longman, 1985.

2. The Cambridge Grammar of the English Language, by Rodney Huddleston and Geoffrey K. Pullum, Cambridge University Press, 2002.

However, If you are just getting started in Computational Linguistics, this book would be a great addition to your reading list. 

The book’s website refers to a companion software called “Syntactica”, a Java application that facilitates exploration of syntactic theory as described in the book.  I did not have the time to play with it, but might do so in the future.

Overall, a very well written book! 

Have a great weekend!

Tags: , ,


If you enjoyed this article, subscribe now to receive more just like it.

Subscribe via RSS Feed

Leave a Reply

Your email address will not be published. Required fields are marked *