Let us begin with full disclosure: I admire Henry Petroski, and I’m predisposed to think well of just about anything he publishes. This is mainly due to the excellent books and engrossing articles he has written in the past. But it is also because we are members of a small community of engineers who write about engineering from a humanistic viewpoint-with the general reader in mind as well as fellow professionals. Through his prodigious example, Petroski has become this group’s widely acclaimed leader.

It is hard to believe that only a dozen years have passed since the publication of Petroski’s first book, To Engineer Is Human: The Role of Failure in Successful Design. This literary endeavor captured the fancy of the nontechnical world and provided the impetus for a BBC television special. Beyond Engineering, a potpourri of essays, followed in 1986. Then came three books that completely broke through the barriers that had so long separated engineering writing from popular culture: The Pencil: A History of Design and Circumstance (1990), The Evolution of Useful Things (1992), and Engineers of Dreams: Great Bridge Builders and the Spanning of America (1995).

Granted, there are a number of talented people writing about technology today, more than there were just two or three decades ago. David McCullough’s books about the Panama Canal and the Brooklyn Bridge are superb popular history, as are Richard Rhodes’s works on the atomic and hydrogen bombs. Tracy Kidder’s The Soul of a New Machine well deserved its Pulitzer Prize. But while this growing attention to the history of technology is heartening, none of these authors are practicing engineers.

Does it matter? Shouldn’t we be satisfied if engineering is represented in the literary arena by professional writers, many of whom have had technical training? Although the answer is partially yes, in an important way it is no. Only a practicing engineer can “feel” what it is to do engineering and “know” engineering with heart and soul.

Petroski seeks to convey such intimacies in his latest work, Invention by Design, a deftly conceived textbook for college courses. Instead of taking his readers into the murky depths of technical definitions, as others have done, Petroski offers nine engineering undertakings, each one of which exemplifies a particular aspect of the engineering process. He begins with a brief history of the paper clip, through which he introduces the concept of design. A key source of design is discontent with an existing product. To illustrate this, Petroski focuses on the need in the late nineteenth century for a replacement for straight pins, which were then used to hold papers together. Eventually, the classic Gem clip, with which we are all familiar, became the standard for this task. Even though the clip is widely accepted and has often been praised for its grace and beauty, Petroski points out a half-dozen deficiencies, such as its inability to hold more than a few papers, and offers as evidence the many patent filings for new variations.

Next, Petroski uses the evolution of the pencil point to explain the process of engineering analysis. This seemingly simple object has been improved through study of its geometry, its material composition, and the methodology of its manufacture. Through this example, Petroski shows that engineers cannot rely solely on creative design plus pure mathematics, but must evaluate their ideas, taking into account the physical imperfections of substances and other contingencies of the material world.

Even conceptual ideas and analytical reviews, however, are not substitutes for the endless physical tinkering required to come up with a successful product. Nor can this process-development-be isolated from the financial and marketing considerations inherent in engineering. Petroski illustrates how such business forces prompted engineers to invent and then reinvent the zipper, eventually leading to innovations such as Velcro, the plastic zipper, and the resealable plastic bag. Even readers who think they are familiar with this process will be amazed by the interactions among technical work, patent struggles, and competitive merchandising.

“Engineering has many dimensions,” Petroski next writes, “but the idea of failure spreads across all of them.” Engineers make progress by designing to preconceived limits and then learning from failure. He explains, for example, that engineers refined the aluminum can’s structural shape with the aid of machines that test it to failure under pressure. Interestingly, the can’s pop-top opener is designed to “fail in a controlled way”-to tear open conveniently but not to leak or break off in the hand.

The importance to engineered products of supporting infrastructures-“networks” is the author’s term-is explained through the saga of the fax machine. This device was successfully designed long before it could be widely marketed, its utility to consumers limited by the signal-carrying capacity of telephone lines.

The computer’s crucial role in engineering is illustrated in the chronicle of the Boeing 777. For that project, computer-aided design (CAD) was carried to new heights. A “paperless” design strategy meant that 238 teams, each as many as 40 engineers, could call up details at any of 7,000 linked workstations.

To show how engineering affects, and is affected by, social forces and politics, Petroski devotes two chapters to, respectively, water supply and bridge building. Here he ranges far and wide, discussing Roman aqueducts, Parisian sewers, the bridges of San Francisco, and other historically significant projects.

Finally, in considering the phenomenon of complex systems, he discusses the past, present, and possible future of the modern skyscraper-its structural designs, mechanical ingenuities, elevator networks, and relationship to the environment. He concludes by reflecting on the World Trade Center in New York City, whose two towers must accommodate 50,000 employees and 80,000 visitors daily. This load taxes not only technological systems within the towers but such community systems as transportation, mail, and telephone.

Petroski’s mosaic approach provides a wonderfully comprehensive synthesis of what it is that engineers actually do. While he includes more diagrams, mathematics, and theoretical mechanics in this work than in his books for the general public, there is nothing here to intimidate the ordinary in-formed reader. He maintains a light touch that is sadly lacking in most works about engineering-certainly works intended for students. Happily, an encounter with any one of Petroski’s books should lead readers to explore his others.