The cover blurbs for Guy Deutscher’s “The Unfolding of Language” (Metropolitan Books, 320 pages, $26) imply a Richard Attenborough-style tour through the variety among the world’s languages, peeking in on why objects have genders in languages like German and French, why there is no word for “have” in most languages, and so on. But Mr. Deutscher’s actual goal is more particular. He seeks to answer a certain question: If erosion of sounds and endings is so typical of how languages change over time, then why don’t people eventually end up with just a mouthful of dust?

For example, in Latin courses we grapple with nouns taking different endings according to how they are used in a sentence: “puer,” “boy”; “pueri,” “the boy’s”; “puero,” “to the boy”; and so on. Nouns fall into five classes, all with their own lists of endings. But in Latin’s descendant, French, all these endings except for the plural “-s” are history.

Mr. Deutscher assumes we want to know how languages build new machinery to replace what gets lost. Linguists such as myself find that a fertile question indeed, but I wouldn’t bet many laypeople are exactly on fire about it. Mr. Deutscher’s task, then, is to show that the answer to this question is worth their time. He gives it the old college try. At one point, Mr. Deutscher warns that the next section is going to be tough going and gamely urges the reader to “make yourself a strong cup of coffee and read on.” This is a general problem with the book: The writing requires too much strong coffee. Even interested lay readers will have trouble sticking with it.

Mr. Deutscher tries hard to render linguistics jargon in friendlier terminology: “squeezing,” “appendages,” etc. But not consistently enough: He often loses sight of how opaque even basic linguistic concepts can be. “The verbal adjective ‘running,’ while still subordinate to the noun, nevertheless acquires some participants of its own” – “participants”? “The change from p to f is part of a longer change of weakening: p > f > h”- but why is one sound “weaker” than another one? Mr. Deutscher seems to think that the glossary in the back takes care of snags of this kind, but people do not enjoy breaking their stride to look terms up every few pages.

Another problem is text flow. Mr. Deutscher knows much and mines great examples. Too often, however, he presents them as butterfly-collecting digressions. At one point he gives a tour of all of the major languages of the Semitic family (which includes Arabic and Hebrew) before addressing an issue about Semitic verbs that involves only a few of the languages. Why the tour at all, then, other than that Mr. Deutscher happens to be a Semitic expert? I myself, for example, savored his extended explanation, based on his own work, of how in the world a language like Hebrew came up with verbs that show tense not by endings (“walk,” “walked”) but by vowels in between the consonants (“KoTeV,” “writes”; “KaTaV,” “wrote”). I expect, however, that it will wear out most other readers.

To get across that erosion is part of how languages change, surely we need some examples. But Mr. Deutscher devotes 40 pages to one demonstration after another, even drifting into describing irregular verbs in ancient languages. In case you were wondering, irregular verbs are often byproducts of the erosion of sounds. But non-linguists are unlikely to catch that connection without studying the text closely.

Mr. Deutscher’s feel for the difference between a book for the public and a graduate seminar is unsure. For example, he shows how the verb “go” as in “I’m going to the store” started being used in sentences like “I’m going to laugh,” where “go” is no longer a verb about movement, but just a marker of the future (“I’m going to laugh” does mean that you are moving toward laughing). This is one of those ways a language builds at the same time as it wears down: English got a new future-marker.

So far, so good, and Mr. Deutscher laudably leaves out the jargon linguists would discuss this in. But then he homes in on the arcane question of how we can know just when “going to” stopped being a verb and became a tense marker (his conclusion: no before-after line can be drawn). That issue has inspired countless articles and conferences in the linguistics world, but people picking Mr. Deutscher’s book up in Barnes & Noble have no reason to care.

Mr. Deutscher actually hits a kind of stride in the last main chapter, in which he shows how language developed from a “Me Tarzan, you Jane” stage to the language of Newsweek. He takes us from a sentence like “father stone take meat cut girl give girl eat finish sleep” to “With a sharp stone he cut some pieces of meat for the girl, who ate them up before falling fast asleep.” Mr. Deutscher shows how languages develop prepositions, adjectives, and subordinate clauses such as “before falling fast asleep.” Here the book gains momentum for the first time, but it is too late.

Perhaps Mr. Deutscher could have built the whole book on this plan, starting by showing, as the cover blurbs imply, how language becomes so complex, and then moving into how languages also simplify at the same time. This way, the descriptions of erosion would fit more clearly into an arc of argument. Instead, he has written a book that is a dandy treat for a linguist, but for most, will read as a rambling affair whose ultimate payoff is only fitfully clear.

Mr. McWhorter last wrote in these pages on integration.