Much has been written in the past 15 years on the subject of endangered languages. Linguists have published academic findings on what causes languages to die, what keeps them alive, and how to save them if they are at risk. Despite the alarming estimate that half of the world's approximately 6,900 languages will disappear during the coming century (p. 3), the plight of endangered languages has yet to take hold of the public's imagination. Harrison's book may help.

Written for both a scientific and a popular readership, the book calls for greater effort to record and save endangered languages because of what they can tell us about the human condition. Every language is a repository of thousands of years of accumulated knowledge about the world in which it is spoken. This information is encoded in the language itself so that each generation can benefit and thrive in challenging environments. Often the information helps extend the world's scientific knowledge as well.

The clichéd scenario is finding a cure for fatal diseases among plants in the Amazon known only to local tribes. Taking a more sweeping view, Harrison describes other kinds of information encoded in little-known languages around the world. Each chapter of the book is organized around one of these topics, such as animal species and behavior, topography, timekeeping, oral tradition, and arithmetic. For example, the Kayapó of Brazil are intimately familiar with 56 varieties of bees, which are linguistically grouped depending on sound, color, flight patterns, and even scent. For nomads like the Tuva in Siberia, the lay of the land has become part of their speech. Even simple sentences describing motion require a speaker to note whether the route ascends or descends and to relate it to the flow of the nearest river. And a group with no need for higher mathematics may package complex arithmetic in its numbering system. In Bukiyip of Papua New Guinea, two separate number systems are used, one in base three and the other in base four, depending on the items counted. As Harrison points out, these languages not only provide details about local environments, they also shine new light on human cognition and language.

Knowing that facts win arguments while personal stories win hearts, Harrison sets out to “pursue hard scientific questions while keeping the human factor in view” (9). As a result, he does not apply theoretical models of language endangerment, nor does he evaluate methods of language maintenance. His premise, that the loss of a language impoverishes our collective knowledge, is not in fact new. However, he has created a valuable addition to the literature by demonstrating, with apt and illustrative detail, the sad consequences for human knowledge when a language dies. The writing is lively and engaging, and the many examples, which are drawn from his own fieldwork and that of others, are vivid and instructive. Theoretical linguists may wish for a stronger framework or tighter structure, but the book's informal style is more likely to attract and educate the public. Likewise, enthusiastic readers may wish for more information on how to help further the cause. Nonetheless, the book may well become the bridge that helps the non-academic world finally understand the language crisis we face.